ASABE Logo

Article Request Page ASABE Journal Article

Imagining African Agrifood Systems: Looking Forward - An e-Book of Executive Summaries, 2023 AMAA Conference & Technology Showcase

Margaret W. Gitau1, Senorpe Hiablie2, Klein Ileleji1, and Ajit Srivastava3

1 Purdue University, West Lafayette, IN

2 Shell International Exploration and Production Inc., Houston, TX

3 Michigan State University, East Lansing, MI

Imagining African Agrifood Systems: Looking Forward - An e-Book of Executive Summaries, 2023 AMAA Conference and Technology Showcase, Paper No. 10.13031/AMAA.2023, pages 1-93 (doi: 10.13031/AMAA.2023). St. Joseph, Mich.: ASABE.

    Cover Design

    Senorpe Hiablie

    Shell International Exploration and Production Inc., Houston, TX

    Publishing Team

    Senorpe Hiablie

    Shell International Exploration and Production Inc., Houston, TX

    Margaret W. Gitau

    Purdue University, West Lafayette, IN

    Joseph Walker

    American Society of Agricultural and Biological Engineers

© 2024 by the respective authors and editors

Each article in this book is the intellectual property of its respective author(s), who retain copyright over their contributions. The compilation, arrangement, and editorial content of this book, as well as any additional materials provided by the editors, are the intellectual property of the editors, who retain the copyright for the book as a whole.

This book is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0), which applies to the collective work and each article individually unless otherwise specified by an author.

To view the full terms of the license, visit https://creativecommons.org/licenses/by-nc-sa/4.0/

For permissions beyond the scope of this license, including those of individual articles, please contact the respective author(s) or the editors as appropriate.

Peer Review Acknowledgement

Each article in this book underwent an internal peer review conducted by designated pillar leads or technical session leads. The editors are grateful to the pillar leads and technical session leads for their critical insights and contributions to this work.

Disclaimer

The views, opinions, and arguments expressed in this work are those of the respective authors of each article/chapter and do not necessarily reflect the views of the editors, publisher, AMAA, ASABE, or any affiliated organizations or institutions. Each author is solely responsible for the accuracy and originality of their content. While efforts have been made to ensure the quality and integrity of the book as a whole, the editors and publisher assume no liability for errors, omissions, or interpretations made by individual contributors.

Published by:

American Society of Agricultural and Biological Engineers

2950 Niles Road, St. Joseph, MI 49085

Abstract.  This e-Book of Executive Summaries emanates from a conference held by the AMAA in 2023 with the objective to “create a platform that fosters collaboration among diverse stakeholders, local actors, and visionaries to design innovative and transformative initiatives for modernizing African agrifood systems.” The conference was built on special sessions held at ASABE’s Annual International Meetings in June 2020, July 2021, and July 2022 and was the first in what AMAA envisions as a series of events to be hosted in different parts of Africa by the AMAA. The content captures highlights and key take-home messages from the conference and features selected papers presented around AMAA’s four Foundational Pillars— (1) Technology, Methodology, and Innovation; (2) Business Development and Entrepreneurship; (3) Capacity Building and Workforce Development; and, (4) Infrastructure and Policy Frameworks—as well as cross-cutting themes that are important in modernizing African agrifood systems.

Keywords.modernizing African agriculture, smallholder farmers, agrifood systems, agrifood value chain, sustainable development, agricultural transformation, agricultural technology, innovations in agriculture, entrepreneurship in agriculture, workforce development.

ABBREVIATIONS

AbbreviationMeaning
ACEDEAfrican Center for Engineering Innovation and Design Education
AEEAgricultural and Environmental Engineering
AfDBAfrican Development Bank
AFREAgricultural, Food and Resource Economics
AMAAAlliance for Modernizing African Agrifood
ASABEAgricultural and Biological Engineers
ASATAgricultural Scalability Assessment Toolkit
ASMCAppropriate Scale Mechanization Consortium
AVCAgrifood value chains
BAEBiosystems and agricultural engineering
BMCBusiness model canvas
CFUColony-forming units
CNNConvolutional Neural Network
CVECountering Violent Extremism
DAUSTDakar American University of Science and Technology
DJDisc jockey
DSPDigital signal processors
FAFHFood away from home
GDPGross Domestic Product
HSTHermetic storage technologies
IITAInternational Institute of Tropical Agriculture
IoTInternet of things
IPIntellectual Property
IRMAInstitute for Rural Management Anand
IWRMIntegrated Water Resource Management
JKUATJomo Kenyatta University of Agriculture and Technology
KEBSKenya Bureau of Standards
KNUSTKwame Nkrumah University of Science and Technology
KPIKey Performance Indicators
LCDLiquid crystal display
LDCLess developed countries
LMICLow- to middle- income countries
MITMassachusetts Institute of Technology
MSUMichigan State University
NARESNational Agricultural Research and Extension Systems
NGONon-governmental organizations
NIAENigerian Institute of Agricultural Engineers
OSUOhio State University
PASAEPan African Society of Agricultural Engineers
PICSPurdue Improved Crop Storage
PSUPennsylvania State University
RASRecirculating aquaculture System
SDGSustainable Development Goals
TARDISSTransformation of American Rubber through Domestic Innovation for Supply Security
TIEETechnology Innovation and Engineering’ Education and Entrepreneurship
TRATechnology Readiness Assessment
UNESCOUnited Nations Educational, Scientific and Cultural Organization
YALIYoung African Leaders Initiative

CONTRIBUTORS

Soji Adelaja is the John A. Hannah Distinguished Professor in Land Policy in the Department of Agricultural, Food and Resource Economics (AFRE) at Michigan State University (MSU). He founded and served as the Director of MSU’s Land Policy Institute from 2006 to 2011. He was Special Adviser on Economic Intelligence at the Presidency in the Federal Republic of Nigeria, where he also led federal intelligence efforts in economic and social security, food security, economic stability, conflict prevention and management, and Countering Violent Extremism (CVE). He also chaired the Presidential Initiative for the Northeast (PINE) aimed at stabilizing and revitalizing the region of Nigeria most devastated by the Boko Haram insurgency. Prior to MSU, Soji spent 18 years at Rutgers University where he served as Executive Dean of Agriculture and Natural Resources, Dean of Cook College, Executive Director of the New Jersey Agricultural Experiment Station, and Chair of AFRE, and founded the EcoPolicy Center, Food Policy Institute, and Food Innovation Center. Soji received his BS in Agricultural Mechanization from the Pennsylvania State University (PSU), and an MS in Agricultural Economics, and an MA and a Ph.D. in Economics from West Virginia University.

Shadrack Kwadwo Amponsah is an accomplished Agricultural Engineer and Senior Research Scientist at the CSIR-Crops Research Institute in Kumasi, Ghana. He has over 15 years of experience in agricultural mechanization, including the design, assessment, and implementation of farm machinery and postharvest technologies. Dr. Amponsah is a passionate advocate for knowledge transfer and has actively engaged in training programs that empower farmers and artisans with modern agricultural practices. He is also a dedicated mentor to the next generation of agricultural engineers.

Emmanuel Yaovi Hunnuor Bobobee is an Associate Professor of Agricultural Machinery Engineering at the Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana. He obtained OAPI Patent 17219 for inventing a mechanical cassava harvester (www.tekcassavaharvester.com) ideal for harvesting cassava and all tropical root and tuber crops. Emmanuel consults on agricultural mechanization, and emerging technologies in agriculture. Emmanuel holds PhD in Agricultural Machinery Engineering from the Swedish SLU, Uppsala, an MSc in Agricultural Machinery Engineering from Cranfield University UK, and a BSc (Hons) in Agricultural Engineering, from KNUST Kumasi, Ghana.

Dieudonné Baributsa is a Professor of Entomology at Purdue University with over 25 years of experience in international development. He serves as the Director of the Purdue Improved Crop Storage (PICS) Program. For the past 15 years, Dr. Baributsa has led efforts to commercialize postharvest technologies across sub-Saharan Africa and other regions. He has managed over $30 million in funding to reduce postharvest storage losses in more than 40 countries, improving food security and the incomes of smallholder farmers.

R. Andres Ferreyra is Industry Standards and Collaborations Lead for Syngenta. He chairs ISO Technical Committee 347 (Data-driven agrifood systems) and ISO International Workshop Agreement 47 (Reference architecture for data-driven agrifood systems). Dr. Ferreyra has developed technology for agrifood and natural resources systems for 30+ years and participated in the international agricultural data standards community since 2010. In his words, “I care with a sense of urgency about people, food, foodways, culture, and living sustainably in nature.”

Victoria M. Garibay. As engineer by training, Victoria prefers to apply a practical problem framing outlook to her work in bridging gaps between science and policy. Her doctoral thesis on East African water data scarcity and recent research on relationships between climate shocks, adaptation responses, and poverty share a common theme—to expand the accessibility of computational modelling for the development of more efficient strategies. Experience in natural, computational, and social science in academic and intergovernmental settings provides her a uniquely comprehensive perspective on solution development, with her personal priority placed on issues faced by understudied and underrepresented populations.

Kifle G. Gebremedhin is an International Professor Emeritus of Biological and Environmental Engineering at Cornell University. He is an elected Fellow of ASABE, faculty fellow of the Cornell Institute of Food Science, and the Atkinson Center for Sustainable Future. Gebremedhin is inducted to the Rural Builder Hall of Fame by the National Rural Builder Magazine and is listed in Who is Who in the East by Marquis. He is the recipient of numerous major national awards for teaching excellence, research, computer software development, mentoring, service, and leadership. He has published a total of 255 research publications. He is nationally and internationally known for his research work on modeling animal thermal stress physiology; heat and mass transfer in biological systems across scales; and analysis and design of post-frame buildings including diaphragm action. He has made several invited presentations and keynote speeches nationally and internationally.

Abel Girma, Founder and Managing Director of the Impact Fund for Modernizing African Agri-Food Systems (IF4MAAS) is a visionary entrepreneur and advocate for African sustainable development. With expertise in impact investing, business development, and operations management, Abel leads IF4MAAS in its role of blended financing for sustainable agrifood projects. Abel's strength lies in bringing together diverse stakeholders, promoting collaboration, and leveraging his cross-cultural communication skills. His educational background includes an Executive MBA and MS in Mechanical Engineering. Abel is experienced in multi-discipline engineering consulting, skilled in new product development, new technology integration, supply chain management, strategic sourcing, and global procurement that bridges diverse sectors from Aerospace to Agriculture.

Margaret W. Gitau is Professor of Agricultural and Biological Engineering at Purdue University with research focus on water resources, in particular: water quality; integrated hydrologic and water quality modeling; and, data-driven decision making and management, including developing strategies and solutions for data scarce areas. Prof. Gitau is Co-founder and Co-lead of the AMAA, Founding President/President of the Mentoring Network for African Women in Academia (MTAWA), Carnegie African Diaspora Fellow, Purdue Insights Fellow, and has served as an Expert Committee Member with the National Academies of Sciences, Engineering, and Medicine.

Senorpe Hiablie is a Researcher in Shell’s Biotechnology and Process Development Group where she provides research and technical due diligence support for scoping novel feedstock options for biofuels and biochemicals as well as process integration support for current operations. She has served as past secretary, vice-president, and president of the African Network Group of the ASABE (ANGASABE). She is a co-founder of the ASABE Alliance for Modernizing African Agrifood Systems initiative (AMAA) and currently serves as its secretary. As a previous research fellow at Project Drawdown, she developed solutions for avoiding methane emissions in livestock systems. She was a post-doctoral researcher with the U.S. Department of Agriculture where she contributed to a national assessment of the sustainability of the U.S. beef value chain. Her academic training was in Agricultural and Biological Engineering (Pennsylvania State University, PhD), Marine Estuarine Environmental Sciences (University of Maryland Eastern Shore, MSc), and Oceanography and Fisheries (University of Ghana, Legon, BSc).

Klein Ileleji is a Professor and Extension Engineer in Agricultural and Biological Engineering at Purdue University, where he has served for over 23 years. He is also the co-founder, CEO and CTO of JUA Technologies International Inc, a company developing solar dehydration technologies and mobile off-grid power generators for agrifood systems. He is the 2017 recipient of The Andersons (NC-213) Cereals and Oilseeds Award of Excellence and also the 2021 recipient of the ASABE Kishida International Award.

Satish Joshi is a Professor in the Department of Agriculture, Food and Resource Economics, Michigan State University, East Lansing. He received a Ph.D. in Public Policy Analysis and Management from Carnegie Mellon University. His research interests include business strategy and sustainability and environmental and energy policy analysis. His publications have appeared in Journal of Agricultural Economics, Strategic Management Journal, The Accounting Review, and Environmental Science and Technology. He teaches courses in strategic management in agrifood businesses, corporate environmental management, and environmental and resource economics.

Nicholas S. Kiggundu is an Associate Professor at Makerere University in the Department of Agricultural and Biosystems Engineering. With a focus on irrigation, hydrology and hydraulics, renewable energy and environmental management. Nicholas has authored/co-authored numerous peer-reviewed articles in high-impact journals (h-index = 24). His current research revolves around hydrological modeling, development of a water-energy-nutrient decision support tool for smallholder aquaculture farmers, and development of renewable energy solutions. Nicholas’ goal is to be able to contribute to the sustainable use of natural resources for improved livelihood through teaching, research, and outreach.

James Williams Kisekka is theEast Africa Office Director/Team Leader for Aid Environment. He holds an MSc in Forestry from Makerere University. He has expertise in planning and implementing Integrated Water Resource Management (IWRM) and natural resource management at various levels. Since 2014 James has played a central role in developing and managing Aid Environment’s projects and programs, setting up systems to monitor their impact, and in general coordination. The projects and programs focus mostly on the nexus between agriculture, rural development, and environmental resources management. James is passionate about natural resources-based business enterprises.

Victor Kongo is a highly accomplished Water Resources Engineer and international consultant in water resources management, climate resilience, sustainable development, and food security. Dr. Kongo’s extensive expertise has been utilized by organizations such as the UN, World Bank, Bill and Melinda Gates Foundation, as well as numerous governments. He has successfully led and implemented over 50 projects for international organizations, governments, and research institutions, and has been instrumental in capacity building and resource mobilization initiatives in Africa. Dr. Kongo is the Executive Director for Global Water Partnership Tanzania and Technical Advisor to the Ministry of Water, Energy and Minerals in Zanzibar.

Ranjani Krishnan is Associate Dean of Faculty and Administration, and Ernest W. & Robert W. Schaberg Endowed Professor of Accounting and Information Systems, in the Eli Broad College of Business, Michigan State University. Ranjani's research has been published in journals such as The Academy of management Journal, Journal of Accounting and Economics, Journal of Accounting Research, Management Science, Strategic Management Journal, and The Accounting Review. She teaches many masters and MBA level courses in management accounting and analytics.

Francis Kumi is a senior lecturer at the Department of Agricultural and Mechanical Engineering, University of Cape Coast in Ghana. His area of specialization is in Agricultural Machinery Systems, Mechanization and Precision Agriculture. He has more than ten years’ experience in training students, engineers, farmers, and other stakeholders to apply engineering principles to agriculture and related sectors.

Bancy Mati is a leading expert in land and water management and Professor at the Jomo Kenyatta University of Agriculture and Technology (JKUAT) in Kenya. She is also a consultant with Resource Plan, and Chairperson of the Association of Irrigation Acceleration Platform (AIAP). She holds a PhD degree in Agricultural Engineering, Food Production and Rural Land Use; MSc. in Land and Water Management and BSc. in Agricultural Engineering. Prof. Mati is an active consultant, researcher, trainer and development worker. She has written over 190 publications variously in books, book chapters, refereed journal papers, consultancy reports and conference proceedings. She is passionate about improving water management for agriculture, environment and livelihoods, of which she has dedicated her working life.

Daniel Moriasi is Research Leader and Hydrologist at the USDA-ARS National Laboratory of Agriculture and The Environment in Ames, Iowa.  Prior to this, he was a Research Hydrologist at the USDA-ARS Grazinglands Research Laboratory in El Reno, Oklahoma. He has authored 138 publications, including 108 peer-reviewed journal articles. Dr. Moriasi holds a B.S. in Agricultural Engineering from Egerton University in Kenya. He also holds an M.S. in Biological and Agricultural Engineering and Ph.D. in Engineering Science (Agricultural Engineering) from Louisiana State University. Additionally, he completed a postdoctoral fellowship in Agricultural Engineering at USDA-ARS in Temple, Texas.  

Subira E. Munishi is a dedicated environmental and water resource engineer with a proven track record of driving innovation in the water sector. She is a visionary leader with expertise in strategic planning and team management, and has over 18 years of experience in water resources assessment. As a professional researcher, consultant, and lecturer at the University of Dar es Salaam's Water Resources Engineering Department, she is committed to advancing technological solutions in the water sector where she has published her works in many areas including on the assessment of water quality in the peri-urban areas of Dar es Salaam.

Judith Njoku Nkechinyere holds a B.Eng. in Petroleum Engineering from the Federal University of Technology, Owerri, Nigeria, and an M.Sc. in Aeronautics, Mechanical, and Electronics Engineering from Kumoh National Institute of Technology, South Korea. She is pursuing a PhD in IT Convergence Engineering, specializing in digital twins, AI, and intelligent transportation systems. Judith also served as the Internship Coordinator for the Climate Smart and Decision Support Systems (CLIMDES) group at Michigan State University.  

Ajay Shah is a professor in the Department of Food, Agricultural and Biological Engineering at the Ohio State University (OSU), where he leads the BioSystems Analysis Lab. He is also the director of the NSF ERC: Transformation of American Rubber through Domestic Innovation for Supply Security (TARDISS). His expertise is in engineering solutions for enhancing the sustainability of agricultural production and biobased systems.

Ajit Srivastava is a researcher, educator, innovator, and a visionary leader in biosystems and agricultural engineering (BAE) with a focus on agricultural mechanization and renewable energy systems.  He is a CANR Distinguished Professor in the department of Biosystems and Agricultural Engineering (BAE) at Michigan State University.  He has served as BAE department chair at MSU for 18 years broadening the department’s focus to include food, environment, water, and energy systems including agriculture.  He was the founding co-director of the MSU Global Center of Food Systems Innovation (GCFSI), a USAID funded project.  He has served as a co-PI of the USAID Appropriate Scale Mechanization Consortium (ASMC).  He is the lead author of a textbook on agricultural machines.  Srivastava has provided leadership to ASABE’s Alliance for Modernizing African Agrifood (AMAA) Systems initiative.  He is a Fellow of ASABE.

Jaden Tatum recently completed her PhD at the Ohio State University, where she is currently a research scholar in the lab of Ajay Shah. Her research is focused on improving food system resilience in global and local contexts, and her research areas include sustainable energy for controlled environment agriculture and appropriate technologies for small-scale farmers.

Daniel Uyeh is an assistant professor in the Department of Biosystems and Agricultural Engineering at Michigan State University. His research focuses on developing climate-smart decision-support systems to address the complex challenges of climate change. He actively engages with scholars, policymakers, and industry experts worldwide. His research aims to bridge the gap between technological innovation and practical application, providing stakeholders with the tools to make informed, data-driven decisions, thereby promoting sustainability and resilience in agricultural systems.

David Waititu Wairimu is a final year student at Jomo Kenyatta University of Agriculture and Technology (JKUAT), pursuing a Bachelor's degree in Agricultural and Biosystems Engineering. David focuses on leveraging emerging technologies to address critical challenges in agriculture and sustainability. His key research interests include additive manufacturing, artificial intelligence, and climate change. David is particularly interested in the intersection of technology and agriculture, exploring how smart systems can optimize agricultural production and promote sustainable practices in response to global climate change.

Rosamond Frempoma Yarquah is a research and teaching assistant at the University of Cape Coast, Ghana, where she works with lecturers and students to apply technology to the agricultural sector to improve food security in Ghana and the world. She is passionate about using her skills and knowledge in agriculture and technology to address the challenges and opportunities in the agricultural industry.

    Eric Zama is a researcher in Environmental Science and Engineering, affiliated with the University of Bamenda, Cameroon. Currently, he is a research scholar in Agricultural and Biological Engineering at the University of Illinois, Urbana-Champaign, USA, where he focuses on optimizing bioreactor-biochar systems for agricultural drainage water treatment. Dr. Zama has authored numerous publications on biochar's agricultural and environmental applications, building on his PhD in Environmental Science from the prestigious Chinese Academy of Sciences.

ACKNOWLEDGEMENTS

ASABE Past Presidents

ASABE Board of Trustees

ASABE Headquarters Staff

African Network Group of the ASABE (ANGASABE)

Pan African Society for Agricultural Engineering (PASAE)

Nigerian Institute of Agricultural Engineers (NIAE)

Association Senegalaise des Ingenieurs de l’Agriculture (ASIA)

International Commission of Agricultural and Biosystems Engineering (CIGR)

Kenya Society of Environmental, Biological, and Agricultural Engineers (KeSEBAE)

Conseil Ouest et Centre Africain pour la Recherche et le Développement Agricoles (CORAF)

Institut Sénégalais de Recherches Agricoles (ISRA)

Dakar American University of Science and Technology (DAUST)

Ecole Nationale Supérieure d’Agriculture (ENSA)

International Institute of Tropical Agriculture (IITA)

SPECIAL THANKS

Kumar Mallikarjunan

Terry Howell

Joseph Akowuah

Michael Ngadi

Daniel Uyeh

Taisha Venort

Dolores Landeck

Michael Faborode

Folarin Alonge

Emmanuel Njukwe

EDITORIAL, HIGHLIGHTS, & KEY TAKEAWAYS

MODERNIZING AFRICAN AGRIFOOD SYSTEMS

Modernizing African Agrifood Systems

Margaret W. Gitau1*, Ajit Srivastava2, Klein Ileleji1, Senorpe Hiablie3, Adesoji Adelaja4, Kifle Gebremedhin5

1Agricultural and Biological Engineering, Purdue University, West Lafayette, IN

2Biosystems and Agricultural Engineering, Michigan State University, East Lansing, MI

3Shell International Exploration and Production Inc., Houston, TX

4Agriculture, Food and Resource Economics, Michigan State University, East Lansing, MI

5Biological and Environmental Engineering, Cornell University, Ithaca, NY

*Corresponding author email: mgitau@purdue.edu

KEYWORDS. modernization, African agrifood systems, holistic approach, technological innovation

Preamble

Agriculture is the single most important economic activity in Africa today and it is the bedrock of Africa’s rural economy (Adelaja et. al., 2021). Agriculture: employs over 60% of Africa’s working population; contributes—in several countries—30% to 60% of the Gross Domestic Product (GDP); and, accounts for about 13% of merchandize exports from Africa (Our World in Data, 2024a,b; Statista, 2024a). Given that 60% of the world’s uncultivated arable land is in Africa (Lambin et al., 2013; Lowder et al., 2016), African agriculture is poised to be an engine of economic growth, enhanced living standards, poverty reduction, and increased food and nutrition security (AfDB, 2024). Along with its backward linkages (input value chain) and forward linkages (food value chain), agriculture is key to Africa’s economic transformation.

However, African agriculture faces many challenges. These include: small farm sizes; low agricultural productivity, limited R&D expenditures, and the ineffectiveness of the National Agricultural Research and Extension Systems (NARES); various input use and supply challenges; strong trade barriers and limited intercountry trade; low levels of domestic and foreign direct investments; limited access to credit, finance and insurance; political instability and insecurity; climate- and water-related challenges; limited development of rural infrastructure; and, limited access to electricity and other forms of energy. These challenges have led to high food product imports which are expected to reach $110B by 2025 (World Bank, 2022). Based on information from the African Development Bank (AfDB), it is expected that Africa's food and agriculture market could reach US$1 trillion by 2030—up from US$280 billion a year in 2023—if these challenges can be addressed (White and Case, 2023).

Technology is poised to play a major role in the transformation of African agriculture. Because of the urgency caused by challenges such as population growth, rapid urbanization, growth in the middle class and related diet transformation, and climate change, we must take a bold, innovative value-chain oriented approach to transform the African agrifood system by developing and deploying appropriate, modern technologies and methodologies. As a global and strategic initiative of the American Society of Agricultural and Biological Engineers (ASABE), the Alliance for Modernizing African Agrifood System (AMAA) has taken a holistic approach in exploring the development and adoption of modern technologies to accelerate improvements in the agricultural value chains in sub-Saharan Africa. Our efforts are well-aligned with UN Sustainable Development Goals (SDGs), particularly: Goals 1 (No Poverty); 2 (Zero Hunger); 5 (Gender Equality); 6 (Clean Water and Sanitation); 7 (Affordable and Clean Energy); 8 (Decent Work and Economic Growth); and, 9 (Industry, Innovation, and Infrastructure) (see UNDESA, 2024).

This e-Book of Executive Summaries emanates from a conference held by the AMAA in 2023 with the objective to “create a platform that fosters collaboration among diverse stakeholders, local actors, and visionaries to design innovative and transformative initiatives for modernizing African agrifood systems”. The conference built on special sessions held at ASABE’s Annual International Meetings in June 2020, July 2021, and July 2022 and was the first in what we envision as a series of events to be hosted in different parts of Africa by the AMAA. This collaborative effort—with partners including Pan African Society for Agricultural Engineering (PASAE), Association Sénégalaise des Ingenieurs de l’Agriculture (ASIA), Nigerian Institute of Agricultural Engineers (NIAE), Kenya Society of Environmental, Biological, and Agricultural Engineers (KeSEBAE), Conseil Ouest et Centre Africain pour la Recherche et le Développement Agricoles (CORAF), Dakar American University of Science and Technology (DAUST), Ecole Nationale Supérieure d’Agriculture (ENSA), International Institute of Tropical Agriculture (IITA), and International Commission of Agricultural and Biosystems Engineering (CIGR)— aimed to address the challenges faced by the continent’s agricultural sector and develop comprehensive projects to drive its modernization. By leveraging the expertise and perspectives of various participants, the Alliance has laid the groundwork for implementing transformative programs on the continent. A summary of the conference is presented in Srivastava et al. (2024).

The sections that follow capture highlights and key take-home messages from the conference, and feature selected papers presented around AMAA’s four Foundational Pillars— (1) Technology, Methodology, and Innovation; (2) Business Development and Entrepreneurship; (3) Capacity Building and Workforce Development; and, (4) Infrastructure and Policy Frameworks—as well as cross-cutting themes that are important in modernizing African agrifood systems. For more details on proceedings and outcomes of prior events, please see Gitau et al. (2021).

Highlights And Key Take-home Messages

The AMAA Conference, Summit, and Technology Showcase with the theme, “Imagining African Agrifood Systems: Looking Forward”, was held on November 14-17, 2023. It comprised: (a) a full day pre-conference session with two training workshops, one on Smart Agriculture and the other on Entrepreneurship; (b) two full days comprising technical presentations and working sessions; (c) a half-day summit with invited speakers and AMAA partners; and, (d) a Technology Showcase that ran through two days of the conference featuring recent innovations in Africa and beyond.

Highlights

Key take-home messages

Pillar I: Technology, Methodology, and Innovation

This pillar focuses on appropriate scale technologies and methodologies aimed at addressing specific agricultural production, irrigation and water resources management, energy, storage, handling, postharvest and food processing technologies and methodologies for African agrifood systems including digitalization of the agrifood system. In addition to encouraging adoption of existing technologies, this pillar aims to promote innovation from African researchers in conjunction with innovation hubs located on the continent.

Tatum et al. developed an integrated grain drying and storage system (iGDSS) to address pest infestation and high moisture content as two key causes of postharvest losses, and tested this in Tanzania with good initial results.

Baributsa addressed the challenges in commercializing technologies within the continent and presents an account of processes, challenges, and opportunities in commercializing chemical-free storage bags to reduce postharvest losses reporting sales of over 50 million bags and counting.

Yarquah and Kumi described the development of an automated chicken egg incubator to provide ideal conditions for hatching eggs, noting the need for alternative power supply e.g. solar to maintain system efficiency.

Bobobee presents a light weight, tractor-powered, mechanical cassava harvester developed to streamline cassava harvesting in line with industrial needs as related to cassava processing and export, reducing the amount of time needed to uproot a plant from up to 600 seconds to one second or less and achieving a capacity as good as 1.9 hours/hectare harvested.

Amponsah et al. present Recirculating Aquaculture Systems (RAS) as the future of fish production in Africa, noting the existence of favorable solutions in sub-Saharan Africa.

Pillar II: Business Development and Entrepreneurship

This pillar facilitates technology evaluation, transfer, and scaling and supports the development of entrepreneurs and innovators in the area of agriculture and agrifood systems. It helps create international public/private partnership by identifying such opportunities and playing an advocacy role to funders in setting investment priorities and funding allocation strategies. Pillar activities include support for innovators and entrepreneurs through business incubation and technical support services.

Joshi and Srivastava discussed structural transformation in agrifood value chains and their impact on technology development, presenting an example of successful development in a milk value chain in India—which has resulted in on average 27 million liters of milk being processed daily and annual revenues of 6.7 billion USD, and is now being replicated across the country.

Wairimu et al. demonstrated the use of deep learning models that they have developed for high efficiency automated vaccination systems for disease prediction, detection, and prevention in the poultry value chain noting the opportunities it presents for intelligent on-farm poultry systems and the potential for new industry to build such systems.

Krishnan et al. explored a variety of factors influencing scaling—an aspect that is critical to achieving impact from technology, methodology, and innovation—and demonstrated the impacts of scaling showing up to 85% labor reductions and up to 150% increases in grain yields.

Girma et al. proposed a novel development finance and impact assessment model (IF4MAAS) that acts as an intermediary to leverage support from foundations, multinational donors, and corporate organizations for value chain integration and scalable agricultural infrastructure investment, ultimately, accelerating progress towards sustainable, productive, and inclusive agrifood systems in Africa.

Pillar III: Capacity Building and Workforce Development

This pillar develops, promotes, and enables capacity building and workforce development efforts in African agrifood systems. We define capacity as encompassing knowledge-based capacity (including education and training, research, extension and outreach programs, as well as cross-country collaborations), infrastructural support, capital capacity (including tuition and technology support), and social capacity (opportunities and barriers based on perceived norms). Workforce is defined as including farmers, growers, and pastoralists, and agricultural support occupations. This pillar focuses on skills development to build a cadre of Africans that go on to create jobs in various agricultural value chains, and includes youth, women, and other vulnerable groups in Africa. Additionally, this pillar identifies priorities for appropriate capacity building and workforce development systems that fit African agriculture.

Njoku et al. explored remote training solutions as offering a paradigm shift in capacity building and workforce development, in particular, as related to artificial intelligence and machine learning, reporting over 95% retention rates and about 85% success rates in publishing for interns in a program that they have developed and implemented.

Zama et al. made the case to revamp capstone design curricula in Africa, including incorporating partnerships between industry and academia, establishing pathways for commercialization, and developing innovation centers of excellence that support state-of-the-art technologies as well as technology transfer.

Srivastava and Uyeh present a concept for a Minor in Smart Ag at Michigan State University that aims to train students to apply their disciplinary expertise to the broad field of smart agriculture. The minor is a collaborative effort led by Biosystems and Agricultural Engineering in partnership with Mechanical Engineering, Electrical and Computer Engineering, and Computer Science and Engineering—in line with the inter- and trans-disciplinary nature of Smart Ag.

Krishnan et al. describe a one-day, seven-session training workshop on entrepreneurship and business model development that was offered to conference participants, providing them with an understanding and the ability to articulate the purpose, role and importance of business models, and the opportunity to critically evaluate, design, and recommend business models for an entrepreneurial new venture.

PILLAR IV: Infrastructure and Policy Frameworks

This pillar focuses on identifying potential partners and key players, and advocates for funding for the development of infrastructure and enabling policy frameworks for technology-related research and development in Africa, including the basic physical and organizational structures and facilities that are required to achieve the desired goals. This pillar advocates for policies through research efforts that support local infrastructural development such as, but not limited to, rural roads, electrification, and storage and processing infrastructure to increase market access, efficiency and productivity of agrifood systems.

Ferreyra and Lehmann noted the lack of data standardization in agrifood industries beyond agricultural machinery and supply chain traceability, and call for the Global South in general, and Africa in particular, to get involved with international data standards for agrifood systems for lasting, positive impacts that can advance the SDGs.

Gitau et al. noted the dire need for enabling policies surrounding climate and water data to improve data-driven decision making and management with respect to water resource systems, thus, protecting the integrity of these systems and ultimately, agrifood systems that are very much dependent on the water resource.

Gender integration and Youth Engagement

On average, women account for 40% to 50% of the agricultural labor force in sub-Saharan Africa (FAO, 2011; IFAD, 2011), but have limited access to inputs (such as fertilizers, improved seeds, and tools), resources (such as land and financing), and training opportunities resulting in overall lower yields from farms that are managed by women (Van De Velde et al., 2020). About 60% of Africa’s population is younger than 25 years (Rocca and Schultes, 2024). This younger generation is quick to embrace technological innovations presenting immense opportunities for their engagement in modernizing agrifood systems in Africa. With current unemployment rates in Africa averaging around 7% and being close to 30% in Southern Africa (Statista, 2024b,c) this group will need to be especially innovative and self-reliant, and look to entrepreneurship for income generation. Aspiration 6 of Africa’s Agenda 2063 incorporates the need to harness the potential of Africa’s women and youth in all sectors by eliminating bottlenecks that hinder the effective engagement of these groups (African Union Commission, 2015).

In his keynote address, Dr. Njukwe highlighted CORAF’s efforts to nurture enterprising youth in agricultural technologies by engaging youth through coaching and mentoring by researchers and the private sector and providing internships with private agrifood companies. Furthermore, he noted their use of gender-sensitive communication and instruments such as policy briefs and factsheets that help capitalize on gender impacts.

Girma et al. (Pillar II) noted IF4MAAS strategy for empowering women and youth including providing access to financing, training, and markets.

Krishnan et al (Pillar II) investigated technology scaling impacts on women empowerment and documented savings in time and effort which then freed up the women to engage in other activities.

Looking Forward

Because of the urgency created by challenges such as climate change, population growth, rapid urbanization, and growth of the middle class and the related diet transformation, we must take a bold, innovative value-chain oriented approach to transform African agrifood systems by developing and deploying appropriate modern technologies and methodologies. To achieve the desired results and outcomes:

ALL stakeholders, government, business and industry, and higher learning institutions need to collaboratively work together. In the process, it is inescapable not to be mindful of the final beneficiaries, farmers, growers, food processors, and the agents of change—youth and women.

Capacity building and workforce development particularly in the general area of digitalization such as generation of big data, data management and analytics, machine learning and artificial intelligence, etc. These are and will be game changers in all sectors of the economy, and the present and next generation of African youth need be equipped with these technologies and methodologies.

Entrepreneurship will provide an alternative to formal employment particularly in regions and countries with high unemployment rates. Capacity building efforts will also be needed in this regard, including curricula adjustments to include courses in and related to entrepreneurship and business management.

Water—its availability and quality—is an aspect that does not get its due attention when it comes to modernizing agrifood systems. Discussions often revolve around the need for irrigation (or increasing irrigation), which is important; much more attention is needed on the development and management of water resources to ensure that water is available in a sustainable manner to enable that much needed irrigation. Furthermore, the quality of water, particularly as it affects population health and well-being, is a factor that needs much attention; a healthy population is needed to participate in the modernizing of African agrifood systems. This need goes beyond the expectation that water will be treated, to keeping pollutants out of environmental waters (lakes, rivers, streams, wetlands, etc.) and water sources dedicated to water supply.

Governments need to: provide the basic infrastructure to support the production and flow of agricultural commodities, and related inputs and services; enact enabling policies to allow producers and other actors in the agrifood space to be competitive at all scales across sectors; and, simplify administrative procedures or eliminate bureaucracies to encourage innovation, and empower farmers, producers, and other key actors in the agrifood industry.

The AMAA provides a platform for the various actors in the African agrifood value chain to: come together and collaborate for the common vision of transforming African agrifood systems so that agriculture becomes an engine of economic growth and development; provide food and nutrition security; and, promote prosperity in a sustainable manner.

Acknowledgments

We extend our most sincere thanks to ASABE and its leadership for supporting the conference, to our host organizations including ISRA and CORAF in Senegal, and to our sister organization PASAE.

References

Adelaja, A., George, J., D’errico, M., Hodbod, J., Jones, L., Jayne, T., and Mulenga, B. 2021. Growing impacts of shocks on Sub-Saharan Africa’s agri-food systems and the mitigating role of resilience, African Agriculture Status Report 2021, Chapter 3; Alliance for Green Revolution in Africa (AGRA): Nairobi, Kenya.

African Development Bank (AfDB). 2024. Feed Africa Strategy 2016 – 2025: Scaling-Up Climate-Smart Agriculture Adoption for Smallholder Farmers in Africa, ACSA Progress Report (2024). https://www.afdb.org/en/documents/feed-africa-strategy-2016-2025-acsa-progress-report-2024

African Union Commission. 2015. Agenda 2063: The Africa We Want. Popular Version. The African Union Commission, Addis Ababa, Ethiopia. Available at: https://au.int/Agenda2063/popular_version. Last accessed, October 2024.

Food and Agriculture Organization (FAO). 2011. The Role of Women in Agriculture. ESA Working Paper No. 11-02. Agricultural and Development Economic Division, The food and Agriculture Organization of the United Nations. Available at: www.fao.org/economic/esa. Last accessed, October 2024.

Gitau, M.W., Asem-Hiablie, S., Ileleji, K., & Srivastava, A. 2021. The Alliance for Modernizing African Agrifood Systems. Resource Magazine. American Society of Agricultural and Biological Engineers. November/December 2021. https://bt.e-ditionsbyfry.com/publication/?i=727239&ver=html5&p=4.

International Fund for Agricultural Development (IFAD). 2011. Women and rural development. International Fund for Agricultural Development, Rome, Italy. Available at https://www.ifad.org/documents/d/new-ifad.org/women-and-rural-developmentpdf. Last accessed, October 2024.

Lambin, E.F., Gibbs, H.K., Ferreira, L., Grau, R., Mayaux, P., Meyfroidt, P., Morton, D.C., Rudel, T.K., Gasparri, I. and Munger, J. 2013. Estimating the world's potentially available cropland using a bottom-up approach. Global environmental change, 23(5), pp.892-901.

Lowder, S.K., Skoet, J. and Raney, T. 2016. The number, size, and distribution of farms, smallholder farms, and family farms worldwide. World Development, 87, pp.16-29.

Our World in Data. 2024a. Share of the labor force employed in agriculture. https://ourworldindata.org/grapher/share-of-the-labor-force-employed-in-agriculture?region=Africa. Accessed December, 2024.

Our World in Data. 2024b. Share of GDP from agriculture. https://ourworldindata.org/grapher/agriculture-share-gdp?region=Africa. Accessed December, 2024.

Rocca, C. & Schultes, I. 2024. Africa’s Youth: Action needed now to support the continent’s greatest asset. Mo Ibrahim Foundation. Available at: https://mo.ibrahim.foundation/sites/default/files/2020-08/international-youth-day-research-brief.pdf. Last Accessed, October 2024.

Srivastava, A., Gitau, M., Hiablie, S. & Ileleji, K. 2024. Imagining the Future at the 2023 AMAA Conference. Resource, May/June 2024, ASABE, St. Joseph, MI. https://bt.e-ditionsbyfry.com/publication/?i=821591&p=16&view=issueViewer

Statista. 2024a. Agriculture as a share of total merchandise exports from Africa from 2015 to 2023. Available at: https://www.statista.com/statistics/1498540/share-of-agriculture-in-total-goods-exports-in-africa/. Last Accessed, December 2024.

Statista. 2024b. Unemployment rate in Africa from 2012-2024. Statista Research Department. Available at: https://www.statista.com/statistics/1319860/unemployment-rate-in-africa/. Last Accessed, October 2024.

Statista. 2024c. Unemployment rate in Africa from 2012-2024, by region. Statista Research Department. Available at: https://www.statista.com/statistics/1321581/unemployment-rate-in-africa-by-region/. Last Accessed, October 2024.

United Nations Department of Economic and Social Affairs (UNDESA). 2024. Sustainable Development. Available at: https://sdgs.un.org/goals. Last Accessed, October 2024.

Van De Velde, P., Stanley, V., & Stickler, M.M. 2020. Invisible Farmers: Why recognizing and supporting women farmers is key to food and nutrition security. World Bank Blogs (blogs.worldbank.org). Last accessed, October 2024.

White and Case LLP. 2023. Africa Focus: Delivering growth through sustainable development. Available at: https://www.whitecase.com/insight-our-thinking/africa-focus-summer-2023. Last accessed, October 2024.

World Bank. 2022. African Food Systems: The Importance of Climate Adaptation.  Available at: https://www.worldbank.org/en/news/infographic/2022/05/10/african-food-systems-the-importance-of-climate-adaptation. Last accessed, October 2024.

PILLAR I: TECHNOLOGY, METHODOLOGY, AND INNOVATION

Design and Evaluation of an Integrated Grain Drying and Storage System For Smallholder Farmers

Jaden Tatum1*, Ajay Shah1

1 The Ohio State University

*Corresponding author: tatum.70@osu.edu

KEYWORDS. hermetic storage, appropriate technology, natural air grain drying

Background And Proposed Solution

High rates of post-harvest losses of staple grain crops in developing countries threaten rural livelihoods and global food security. Most post-harvest loss in developing countries has been found to occur during on-farm storage, through degradation by microbial activity or storage pests (Sheahan & Barrett, 2017). While microbes naturally occur in and on grain, high moisture puts grain at risk for developing mycotoxins, harmful secondary metabolites of some fungal species (Mahuku et al., 2019). Hermetic storage techniques are a chemical-free alternative to pesticides and create an air-tight which allows natural respiration to deplete oxygen levels in the grain to levels that reduce or eliminate insect activity. Multi-layered flexible storage bags have been developed and promoted as a tool for smallholder farms to reduce impacts of storage pests, but these bags are vulnerable to puncture by rodents and some insects and must be replaced periodically (Jones et al., 2011). To maintain grain quality throughout storage, the grain must be adequately dried, and the bags re-sealed perfectly after grain is accessed. For subsistence farmers who typically rely on traditional on-ground sun drying and must access grains regularly throughout the year, these conditions may not be guaranteed.

To combat two major causes of postharvest losses - pests and elevated moisture- a novel technology has been developed and evaluated which provides both mechanized drying and rigid hermetic storage. This integrated grain drying, and storage system (iGDSS) has been designed as an appropriate technology for developing country contexts, using as its basis 55-gallon drums which are widely available and affordable in many areas of sub-Saharan Africa. Evaluations of iGDSS have been carried out to determine its technical feasibility, economic viability, and desirability to end-users.

P430#yIS1

Figure 1. Integrated Grain Drying and Storage System Components

Experiments and Results

To evaluate technical performance, experimental evaluations of drying of dent corn in iGDSS were performed, followed by 6-month storage trials evaluating changes to key grain quality parameters under simulated pest pressure and grain access periods. Experimental grain drying trials determined the drying rate, drying energy, and electric drying efficiency of the grain dryer. In ten replicates, dent corn was dried from 22.4 to 14.6% in 2.0 days using continuous unheated air. The average drying rates were 0.16% moisture removed per hour, and 7.19% moisture removed per kWh. The evenness of airflow in the on-floor duct design was evaluated using a digital manometer to measure air pressure at 60 different points in the barrel, based off the methods of pressure loss determination in a series of duct configurations (Williamson, 1965). The low drying time allows for modular adoption of the system by end-users, with farmers needing to purchase only one fan to dry multiple sets of 150 kg capacity barrels.

After drying in all replicates was completed, one half of treatments in iGDSS were inoculated with 10/kg adult Sitophilus zeamais, and all systems were sealed. Throughout storage, the oxygen and carbon dioxide levels in storage containers were monitored and used to determine the oxygen intrusion inherent with sampling. Grain quality parameters of moisture content, test weight, and presence of colony-forming units was carried out bi-weekly. Moisture content of maize was determined using an oven-drying method with three replicate samples (30 g) prepared in brown paper bags and kept in a 100°C oven for 72 hours. Test weight was measured by using a digital handheld density tester. The number of colony-forming units (CFUs) on the grains surface was determined using a surfactant rinse and incubation on a fungi-promoting medium (Mcdonough et al., 2011).

Throughout the 6-month storage period, there was a statistically significant (p < 0.05) increase in moisture content within all treatments of stored grain, from an average of 13.7% to 14.7%, which was in line with the 1-2% increase reported in other hermetic studies, which can result from natural respiration in the grain (Mlambo et al., 2017; Njoroge et al., 2014). Over time, the levels of test weight shifted in response to the changes in moisture content. Within all replicates, colony counts remained stable across the six-month storage period, and there were no significant differences between the initial and final CFU counts across all treatments. This is very positive, as it indicates that when grain is well-dried before storage, fungal levels are well controlled in hermetic storage systems, even allowing for bi-weekly access and storage pest infestation at 10 weevils/kg. Notably, there was no significant difference in any grain quality parameters at 6 months between treatments with and without storage pests, indicating that iGDSS was effective in reducing the impacts of 10/kg maize weevils in hermetically sealed and periodically accessed maize (Tatum & Shah, 2024). The oxygen and carbon dioxide readings indicated an average oxygen intake of 0.08 to 0.09% O2/day in the three days following sampling in the non-inoculated and inoculated treatments, respectively (Tatum & Shah, 2024). This is below recommended threshold levels of 0.15% O2/day which preserves the effectiveness of the modified atmosphere management (Navarro & Navarro, 2014).

On-ground fieldwork in Tanzania informed the economic analysis and gave insight into the perceptions of end-users and the utility of iGDSS in their context. With the help of local community development officers and sociology faculty, six communities consisting of primarily subsistence maize farmers were identified, and 280 farmers were surveyed about current farming and postharvest practices, as well as perceptions of existing hermetic storage technologies and desirability and proposed cost of grain drying infrastructure. Preliminary costs of iGDSS for farmers were found using insights from farmer interviews along with market visits.

Among respondents, it was found that 62% of households were already practicing airtight storage in metal drums, validating the availability and appropriateness of these drums for the basis for the iGDSS technology. It was also found that a substantial portion, around 45%, of the farmers surveyed, reported that they would be willing to pay for forced-air drying, including subsistence-only households that did not anticipate marketing benefits from well-dried grains but still saw value in mechanized drying. Preliminary economic analysis found that the capital cost to adopt iGDSS for the average small farmer surveyed was around $120 USD, including the purchase of one fan and three barrels and their retrofits.

Conclusions and recommendations

A novel grain drying, and storage technology was evaluated through experimental trials, preliminary economic evaluations, and engagement with potential end-users in Tanzania. Experimental evaluations through storage and access trials – simulating grain removal by subsistence farmers - determined that grain drying proceeds quickly and evenly, that fungal growth is repressed and that even when the system was opened to remove grain, the oxygen introduced was below thresholds to preserve the effectiveness of the modified atmosphere management. On-ground fieldwork in Tanzania informed the economic analysis and gave insight into the perceptions of end-users and the utility of iGDSS in their context.

This low-cost and effective grain drying infrastructure has the potential to greatly reduce postharvest losses of staple grain crops in developing countries and lead to greater global food security. Future work should include on-farm evaluation of this technology and further work in retrofitting the on-floor duct system into barrels with close-headed designs.

Acknowledgments

The graduate student was supported by NSF Award No. 1922666 during the study period. The authors would also like to express thanks to the Ohio State University Sustainability Institute, Office of International Affairs and College of Food, Agricultural, and Environmental Sciences for student grant funding to support this work.

References

Jones, M., Alexander, C., & Lowenberg-DeBoer, J. (2011). An Initial Investigation of the Potential for Hermetic Purdue Improved Cowpea Storage (PICS) Bags to Improve Incomes for Maize Producers in Sub-Saharan Africa (Issue #11-3, pp. 1–44).

Mahuku, G., Nzioki, H. S., Mutegi, C., Kanampiu, F., Narrod, C., & Makumbi, D. (2019). Pre-harvest management is a critical practice for minimizing aflatoxin contamination of maize. Food Control, 96, 219–226. https://doi.org/10.1016/j.foodcont.2018.08.032

Mcdonough, M. X., Campabadal, C. A., Mason, L. J., Maier, D. E., Denvir, A., & Woloshuk, C. (2011). Ozone application in a modified screw conveyor to treat grain for insect pests, fungal contaminants, and mycotoxins. Journal of Stored Products Research, 47(2011), 249–254. https://doi.org/10.1016/j.jspr.2011.04.001

Mlambo, S., Mvumi, B. M., Stathers, T., Mubayiwa, M., & Nyabako, T. (2017). Field efficacy of hermetic and other maize grain storage options under smallholder farmer management. Crop Protection, 98(2017), 198–210. https://doi.org/10.1016/j.cropro.2017.04.001

Navarro, S., & Navarro, H. (2014). The biological and physical aspects of hermetic storage: A critical review. 11th International Working Conference on Stored Product Research, 337–354. https://doi.org/10.14455/DOA.res.2014.58

Njoroge, A. W., Affognon, H. D., Mutungi, C. M., Manono, J., Lamuka, P. O., & Murdock, L. L. (2014). Triple bag hermetic storage delivers a lethal punch to Prostephanus truncatus (Horn) (Coleoptera: Bostrichidae) in stored maize. Journal of Stored Products Research, 58(2014), 12–19. https://doi.org/10.1016/j.jspr.2014.02.005

Sheahan, M., & Barrett, C. B. (2017). Review: Food loss and waste in Sub-Saharan Africa. Food Policy, 70(2017), 1–12. https://doi.org/10.1016/j.foodpol.2017.03.012

Tatum, J., & Shah, A. (2024). Performance of Small-Scale Hermetic Storage Systems Under Periodic Access. Agriculture, 14(10), 1839-1854. https://doi.org/10.3390/agriculture14101839

Williamson, W. F. (1965). Pressure losses and drying rates in grain ventilated with various on-floor duct systems. Journal of Agricultural Engineering Research, 10(4), 271–276. https://doi.org/10.1016/0021-8634(65)90070-3

Commercialization of Hermetic Bags in Sub-Saharan Africa: The PICS Experience

D. Baributsa1*

1Department of Entomology, Purdue University, W. Lafayette, IN

*Corresponding author: dbaribut@purdue.edu

KEYWORDS. smallholder farmers, grain storage, postharvest losses, hermetic technologies, scale-up

Abstract

Commercializing technology, the process of bringing an innovation from the laboratory to the marketplace, is a significant challenge in developing countries. The Purdue Improved Crop Storage (PICS) technology, an airtight bag, was designed to mitigate postharvest storage losses of grains and other commodities. PICS bags, a chemical-free storage method, reduce losses caused by insect pests from an estimated 30% to less than 1%. Additionally, the technology helps mitigate mold development, which leads to aflatoxins. PICS bags have been disseminated in over 40 countries across sub-Saharan Africa, South and East Asia, Latin America, and the Caribbean. More than 12 million farmers have been trained in using PICS bags through village demonstrations. In addition, media has been utilized to increase awareness among farmers and other users. To date, over 50 million bags have been sold by manufacturers, distributors, and vendors. By 2023, the return on investment for farmers is estimated to be around $3.0 billion.

Introduction

Postharvest storage management of agricultural products is a significant challenge in developing countries, particularly among smallholder farmers. Most staple crops produced and stored on smallholder farms, such as maize, wheat, rice, and beans, are susceptible to insect pests. On average, farmers lose up to 30% of their grains during postharvest. This means households have less grain (of lower quality) to consume and sell later in the year. This results in food and nutrition insecurity and loss of income. Smallholder farmers use different grain protection techniques, including traditional/local methods, pesticides, and hermetic systems, to preserve their grain. Postharvest storage losses are acute in developing countries because most technologies tailored to the needs of low-resource farmers are not available, scalable, affordable, or often ineffective. In addition, some of these methods, such as pesticides, may have health hazards due to their toxicity to humans or the environment when not properly used.

Hermetic storage technologies (HSTs) as an alternative to traditional and chemical control methods have gained significant interest among farmers, the private sector, governments, and development agencies. Hermetic storage containers come in different forms and sizes. Commonly used HSTs include silos and drums (metal and plastic), cocoons, and plastic bags (Bakoye et al., 2020; Gitonga et al., 2013; Kotu et al., 2018; Villers et al., 2008). Hermetic storage bags are chemical-free methods that owe their effectiveness to the airtight environment conditions created inside the container during storage. Biological processes such as respiration and metabolic activities, driven mainly by the presence of insects and other organisms, lead to the depletion of oxygen and the release of carbon dioxide inside hermetic containers (Murdock et al., 2012). Hence, the hypoxic environments created inside hermetic containers become unfavorable to the development and reproduction of insect pests and thus minimize or stop grain damage. The airtight nature of hermetic bags prevents the entry of oxygen, creating an environment where pests and mold cannot thrive.

Hermetic bags are the most widely disseminated and adopted hermetic storage methods among smallholder farmers in developing countries. The use of hermetic bags to store grain has significantly increased in the last 15 years, spearheaded by the development and dissemination of the Purdue Improved Crop Storage (PICS) bag (Baributsa & Concepcion Ignacio, 2020). Smallholder farmers’ interest in hermetic bags is driven by several factors, including: (i) the severity of storage losses, (ii) the efficacy of these technologies, and (iii) their cost effectiveness. In addition, hermetic bags significantly address the food safety challenges posed by the conventional method of treating grains with insecticides or poorly storing grain, which results in moldy grain, leading to aflatoxin contamination.

We share Purdue's experience in commercializing PICS technology in developing countries. We briefly discuss the approach to creating demand and developing the supply chain among smallholder farmers in developing countries. We describe the role of each actor in facilitating the commercialization of PICS technology in developing countries, particularly in sub-Saharan Africa. We share some of the achievements and lessons learned in commercializing postharvest technology. Smallholder farmers represent a strong and reliable market, but creating demand for new products requires sustained investment from governments and development partners.

The Pics Technology

The PICS technology is a triple-layer hermetic bag made of two inner liners and one woven bag (Baributsa et al., 2015). This hermetic technology saves farmers time, labor, and exposure to multiple applications of pesticides (which lead to health hazards) and prevents grain from being contaminated with mycotoxins (aflatoxins). It also guarantees grain protection for a longer time, allowing farmers to take advantage of better market prices.

a) The PICS bag’s value proposition

PICS bags provide an airtight storage solution that significantly extends grain shelf life and allows smallholder farmers to store grains for months or years. Farmers invest in PICS because it is a solution that addresses a real problem/need, “storage loss at the household level”. During PICS awareness and training, our messages have been focused on several benefits, including:

b) Use of PICS bags

The PICS technology was initially developed to protect cowpeas against insect attacks during storage. Later, its use was expanded to store other dry grains and commodities. Currently, PICS bags are used to prevent insect attacks, mitigate mold growth, maintain aroma, preserve color, and improve fermentation. There are more than 20 commodities that can be stored in PICS bags, including maize, rice, wheat, sorghum, millet, cowpea, soybean, common beans, pigeon peas, Bambara nuts, groundnuts, chickpeas, cocoa beans, coffee, hibiscus seed, wheat and maize flour, animal feed, kocho fermentation, and dried fruits and vegetable (Baoua et al., 2014; Murdock & Baributsa, 2014; PICS, 2024). PICS bags are easy to use (Baributsa et al., 2015). The product stored in PICS bags, particularly grains, should be dried to the recommended moisture content (e.g., 13.5% for maize).

Approach to commercialization

The commercialization of PICS technology required partnerships with various stakeholders to create the demand and ensure availability. Development partners, including non-governmental organizations (NGOs) and national government agencies (research and extension), played a vital role in educating farmers about the benefits of PICS bags and their proper use through training and demonstrations. While the demand was being created, local and regional manufacturers were engaged to produce and supply the bags to distributors, vendors, and retailers who made them available to farmers in villages and rural markets.

a) The role of partners

Figure 1 shows the interconnectivity among three main players involved in commercializing the PICS bags: smallholder farmers, development partners, and private sector actors.

b) Creating the demand

Several approaches were used to create demand for PICS bags among potential customers and users, including village training and media awareness (radio, television, posters, flyers, and news articles) (Baributsa, et al., 2014).

Village training was the primary focus for creating the demand for hermetic bags and included four steps: awareness creation, demonstration, follow-up during, and open-the-bag ceremonies. During awareness creation, a public event, farmers were introduced to hermetic storage and PICS bags. The demonstration was also a public event showing farmers how to use PICS bags. Five households (HHs) in each community were provided five PICS bags at no cost, but they had to volunteer their grains for a three to six-month storage period (they could not consume or sell them). Follow-up involved visits by extension agents to the five HHs during storage to ensure the grain was safe and intact. Open-the-bag ceremonies were public events organized for the whole community to witness PICS bags' efficacy in preserving grain. Providing a limited number of PICS bags to a few households has proven effective in creating demand among community members. On average, 20% of participants in village training purchase PICS bags in the first year.

Media, particularly radio, was used to create PICS bag awareness among farmers. Radio messages were conveyed through jingles, radio talk shows, phone-in programs, and disc jockey (DJ) mentions during morning shows. Tapping into local community radio stations helped communicate with farmers in their local languages. Though the ownership of television is low among farmers, we used it primarily to target urban farmers and other customers such as development agencies. Posters were printed and given to extension agents and vendors as a training aid. At the same time, flyers and newspaper articles were intended for city dwellers who may be serving or have connections with farming communities.

Training and media activities have significantly increased the adoption of hermetic bags. About five years after PICS bags were introduced in West and Central Africa (2012), the adoption rate among farmers across the region and in Niger was 18% (Moussa et al., 2014). By 2018, the adoption rate in Niger had risen to 48.4% (Rabé et al., 2021). Similarly, in East Africa, 25% of Tanzanian farmers had adopted PICS bags within five years of their introduction (2019). By 2022, the adoption rate in Tanzania had increased to 39.6 % (Zacharia et al., 2024). The rise in PICS sales over the years reflects these growing adoption rates.

c) Getting PICS bags to farmers

The PICS supply chain is key for adoption. It offers minimal value in teaching farmers about technologies that are unavailable in their villages or local markets. It was critical to scale both village training and supply chain to increase adoption (Baributsa et al., 2014; Baributsa & Concepcion Ignacio, 2020). The development of the supply chain involved identifying local or regional plastic manufacturers in each country. Identified manufacturers or suppliers signed licensing agreements to ensure quality bags were produced and supplied to smallholder farmers. PICS manufacturers were likely found among those that produced polyethylene and polypropylene products. The capacity to scale PICS manufacturing was due to its low point of entry in terms of machinery. Any plastic manufacturer with basic equipment could produce hermetic bags. This was both a strength and a weakness as any plastic factory with the capacity to manufacture hermetic bags could produce both PICS or imitation/fake bags. Given that PICS was a new product, most manufacturers had two key questions- where is the market, and how big is it? Hence, building the demand was key in convincing manufacturers to launch the production of PICS bags. In some cases, when manufacturers were reluctant, the project financed the production of a limited number of bags to prove that there was a market.

To make the bags available among farmers, efforts were made to recruit distributors, vendors, and retailers who would buy the bags from manufacturers either in cash and carry or on credit (if they had existing relations). In some cases, distributors were given PICS bags on consignment by projects or provided with revolving funding. Just as for farmers, awareness building was needed among the private sector actors. Meetings were held with the supply chain actors to discuss issues related to developing an effective supply and distribution system to make the bags available to farmers. These included production process and lead time, pricing, logistics (e.g., transportation), quality control, inventory management, etc. Additionally, media was used to advertise PICS bag retail points in towns, villages, and markets, helping farmers locate local vendors.

d) Investments

Funding in the form of grants was provided by several donors to implementers, including local and internal NGOs, national and international agricultural research institutions, universities, and the private sector to create awareness and the demand for HSTs among farmers. In addition, some donors provided grants through revolving funds to supply chain actors to stimulate the manufacturing and or distribution of hermetic bags. Some innovative funding approaches, such as the AgResults in Kenya, were used as a pull mechanism to incentivize (“pay for results”) the private sector to invest in promoting hermetic storage among smallholder farmers (Tanager, 2018; Mainville & Ness-Edelstein, 2021). Economic incentives were given to companies that achieved specific development outcomes (quantity of maize stored in hermetic technologies). Further, some organizations have provided funding to the private sector through social impact investments by acquiring shares in these companies. These investments are focused on generating social benefits such as improving smallholder farmers’ food security and incomes through reduced postharvest losses.

e) Barriers to purchasing

Farmers' decisions to invest in PICS bags are influenced by several factors related to both crop production and the technology itself.

Figure 2. Sales of PICS bags in different regions of the world from 2007 to December 2023

Impact of commercialization

As noted earlier, PICS bags were originally developed for cowpea storage. In the early version of the bags, the “C” in PICS stood for cowpea. After the bags were tested and proved to be effective in preserving other grains, the “C” was changed to Crop. Expanding the use of PICS bags from cowpeas to other crops provided an opportunity to grow the market of the technology from West and Central Africa (the largest producer of cowpeas - about 80% of world production) to more regions in sub-Saharan Africa and the world where major cereals and legumes are produced. Market-building activities that began in 2014 in East and Southern Africa (e.g., Tanzania, Ethiopia, Kenya, Uganda, and Malawi) and other regions significantly increased the demand for PICS bags to store maize (Figure 2). Sales grew rapidly in East Africa, going from 9% in 2014 to 71% in 2023 of the total PICS bags sold worldwide. Most farmers (more than 80%) who purchased PICS bags used them to store cowpeas in West Africa and maize in East Africa. This implies that in each region, there is a crop that is driving the demand for PICS bags. Both crops have huge postharvest storage challenges due to insects. However, maize has surpassed cowpea in driving the demand for PICS bags due to its large production (20.2 million MT for maize vs 6.0 million tons for cowpea) and its importance as a staple crop across in East Africa (Anago et al., 2021; FEWSNET, 2022; IITA, 2024).

Total sales of PICS bags reached 50 million in December 2023 (PICS, 2024). Sales grew rapidly from 2 to 7.3 million PICS bags from 2015 to 2020, respectively. Though these sales are important, they represent a small portion of the potential markets. The sale of 5 million bags per year represents only 2.5% of the 20 million tons of maize produced in East Africa. The potential market is 20 million bags per year if only 10% of the total production was stored in PICS bags. Sales continue to grow in most regions due to continued awareness created by development partners and the private sector and improvement in the supply chain (bags being made available to farmers in shops and markets in rural areas). If sales had followed a normal trend, the pick would have been in 2021. Project purchases and exports in several countries inflated the sales in 2020.

Low sales in 2019 and 2023 were due to several factors, including low crop production and increased competition from imitations and fake bags in some markets such as Nigeria and Kenya. Poor performances of the supply chain in some countries have provided opportunities for competing products to reduce the PICS bags’ market share. For instance, in Kenya and Nigeria, there are more than 10 brands of hermetic bags in each market. In Nigeria, most of these products are imitations or fake bags (Figure 3) of substandard quality (Figure 4). In addition, the prices of these substandard bags are lower than those of PICS bags (Figure 4), indicating that many manufacturers of imitation bags compromise on quality to compete. When the quality is compromised by reducing the thickness of the liners, these bags often fail to protect stored products. As a result, insects can continue to develop and eventually create holes in the liners. To address the issue of low-quality hermetic bags in the market, some countries have developed standards for these products. For example, the Kenya Bureau of Standards (KEBS) has collaborated with manufacturers and suppliers to establish hermetic bag standards for East Africa (KEBS, 2019).

Figure 3. Images of PICS and imitation bags sold in Nigeria in 2019

Other factors affecting PICS bag sales include a mismatch between demand and supply, as well as manufacturers' limited capacity to produce enough bags during peak times (Omotilewa & Baributsa, 2022). Order backlogs at PICS manufacturing facilities, sometimes lasting several weeks, have significantly impeded sales in certain countries. While smallholder farmers are highly aware of PICS bags, resolving these supply chain challenges could greatly boost sales. To improve supply and availability at the last mile, manufacturers and distributors are developing strategies such as offering incentives (e.g., discounts) to distributors and vendors to stock bags before the harvest season, creating regional depots to address logistical challenges, advertising retail points, and building a dense distribution network in rural areas (Omotilewa & Baributsa, 2022).

Figure 4. Prices and thicknesses of PICS and imitation bags sold in Nigeria in 2019.

Despite these challenges, the demand for PICS bags continues to grow in various countries. Farmers continue to demand PICS bags because they understand the technology's benefits in terms of improving food and nutritional security, food safety, and income. Households using PICS bags store longer and are more food secure than those who do not (Gitonga et al., 2013; Omotilewa et al., 2018). On average, a farmer using a PICS bag to store 100kg of cowpeas in West Africa or maize in East Africa has a cash flow of about $26 and $20, respectively (Baributsa & Njoroge, 2020; Moussa et al., 2014; Rabé et al., 2021). With 50 million PICS bags sold by December 2023, each used at least three times, it is estimated that farmers have made or saved more than $3.0 billion.

Key lessons learned in commercializing hermetic bags

Conclusion

The commercialization of PICS bags has had a profound impact on agricultural practices across sub-Saharan Africa. PICS sales have significantly increased in the last 10 years though this represents a minute share of the potential market. Initially developed for cowpea storage, the technology's success in preserving other crops expanded its reach from West and Central Africa to East and Southern Africa, significantly increasing demand. Despite challenges such as supply chain inefficiencies and competition from counterfeit products, PICS bags have become essential for smallholder farmers, improving food security, safety, and income. Key lessons from this journey include the importance of partnerships, capacity building, adaptive research, and protecting innovations through IP rights. For sustainable commercialization, a robust input supply chain and profitable business models are essential, ensuring that demand drives supply and fosters broader adoption.

Acknowledgments

We extend our gratitude to the Bill and Melinda Gates Foundation for supporting Purdue University with the PICS 1, 2, and 3 grants, which enabled the commercialization of hermetic bags in several sub-Saharan African countries. We also appreciate the efforts of development partners, including governments, NGOs, and other donors, in reducing postharvest storage losses among smallholder farmers.

References

Anago, F. N., Agbangba, E. C., Oussou, B. T. C., Dagbenonbakin, G. D., & Amadji, L. G. (2021). Cultivation of Cowpea Challenges in West Africa for Food Security: Analysis of Factors Driving Yield Gap in Benin. Agronomy, 11(6), 1139. https://doi.org/10.3390/agronomy11061139

Bakoye, O. N., Baoua, I. B., Seyni, H., Amadou, L., Murdock, L. L., & Baributsa, D. (2020). Comparative Study of Cowpea Storage Technologies in the Sahel Region of Niger. Insects, 11(10), 689. https://doi.org/10.3390/insects11100689

Baoua, I. B., Amadou, L., Ousmane, B., Baributsa, D., & Murdock, L. L. (2014). PICS bags for post-harvest storage of maize grain in West Africa. Journal of Stored Products Research, 58, 20–28. https://doi.org/10.1016/j.jspr.2014.03.001

Baributsa, D., & Concepcion Ignacio, M. C. (2020). Developments in the use of hermetic bags for grain storage. In D. E. Maier (Ed.), Advances in postharvest management of cereals and grains (Dirk Maier, Issue July, pp. 171–198). Burleigh Dodds Science Publishing. https://doi.org/10.19103/AS.2020.0072.06

Baributsa, D., & Njoroge, A. W. (2020). The use and profitability of hermetic technologies for grain storage among smallholder farmers in eastern Kenya. Journal of Stored Products Research, 87, 101618. https://doi.org/10.1016/j.jspr.2020.101618

Baributsa, D., Abdoulaye, T., Lowenberg-DeBoer, J., Dabiré, C., Moussa, B., Coulibaly, O., & Baoua, I. (2014). Market building for post-harvest technology through large-scale extension efforts. Journal of Stored Products Research, 58, 59–66. https://doi.org/10.1016/j.jspr.2014.02.012

Baributsa, D., Baoua, I., Abdoulaye, T., & Murdock, L. L. (2015). A Guide on the Use of PICS Bags for Grain Storage. In Purdue Extension Bulletin: Vol. E-265-W. http://extension.entm.purdue.edu/publications/E-265.pdf

Baributsa, D., Díaz-Valderrama, J. R., Mughanda, D., Lubanzadio, A., Nshombo, J. P. C., Sperling, L., & Baoua, I. B. (2021). Grain Handling and Storage in Lubero and Rutshuru Territories in the North Kivu Province, the Democratic Republic of Congo. Sustainability, 13(17), 9580. https://doi.org/10.3390/su13179580

Baributsa, D., Djibo, K., Lowenberg-DeBoer, J., Moussa, B., & Baoua, I. (2014). The fate of triple-layer plastic bags used for cowpea storage. Journal of Stored Products Research, 58, 97–102. https://doi.org/10.1016/j.jspr.2014.02.011

FEWSNET. (2022). Regional supply and market outlook: East Africa: Maize. The Lancet, 1–12.

Fintrac. (2016). Smallholder Grain Storage in Sub-Saharan Africa: A Case Study on Hermetic Storage Technology Commercialization in Kenya. In Fintrac Topic Papers (Vol. 3, Issue February). http://www.fintrac.com/sites/default/files/2017-09/SmallholderGrainStorage.pdf

Foy, C., & Wafula, M. (2016). Scaling Up of Hermetic Bag Technology (PICS) in Kenya: Review of successful scaling of agricultural technologies. In Report for USAID: Contracted under AID-OAA-M-13-00017. E3 Analytics and Evaluation Project.

Gitonga, Z. M., De Groote, H., Kassie, M., & Tefera, T. (2013). Impact of metal silos on households’ maize storage, storage losses and food security: An application of a propensity score matching. Food Policy, 43, 44–55. https://doi.org/10.1016/j.foodpol.2013.08.005

IITA. (2024). Cowpea (Vigna unguiculata). https://www.iita.org/cropsnew/cowpea/#1620977076242-b594b658-46f3

KEBS. (2019). Hermetic Storage Bags — Specification. Kenya Bureau of Standards. http://www.puntofocal.gov.ar/notific_otros_miembros/ken829_t.pdf

Kotu, B. H., Abass, A., Hoeschle-Zeledon, I., Mbwambo, H., & Bekunda, M. (2018). Returns to improved storage and potential impacts on household food security and income: evidence from Tanzania. https://hdl.handle.net/10568/99493

Mainville, D., & Ness-Edelstein, B. (2021). AgResults Evaluation: Kenya On-Farm Storage Challenge Project. Sustainability Assessment. https://agresults.org/learning/29-agresults-kenya-on-farm-storage-challenge-project-summary-report/file

Moussa, B., Abdoulaye, T., Coulibaly, O., Baributsa, D., & Lowenberg-DeBoer, J. (2014). Adoption of on-farm hermetic storage for cowpea in West and Central Africa in 2012. Journal of Stored Products Research, 58, 77–86. https://doi.org/10.1016/j.jspr.2014.02.008

Moussa, B., Lowenberg-Deboer, J., & Baributsa, D. (2010). Adoption of Hermetic Storage for Cowpea in Niger and Burkina Faso in 2010. In 5th World Cowpea Conference (p. 2010).

Murdock, L. L., & Baributsa, D. (2014). Hermetic storage for those who need it most -subsistence farmers. In F. . Arthur, R. Kengkanpanich, W. Chayaprasert, & D. Suthisut (Eds.), 11th International Working Conference on Stored Product Protection (pp. 310–323). http://spiru.cgahr.ksu.edu/proj/iwcspp/iwcspp11.html. https://doi.org/10.14455/DOA.res.2014.56

Murdock, L. L., Margam, V., Baoua, I., Balfe, S., & Shade, R. E. (2012). Death by desiccation: Effects of hermetic storage on cowpea bruchids. Journal of Stored Products Research, 49, 166–170. https://doi.org/10.1016/j.jspr.2012.01.002

Omotilewa, O. J., & Baributsa, D. (2022). Assessing the Private Sector’s Efforts in Improving the Supply Chain of Hermetic Bags in East Africa. Sustainability, 14(19), 12579. https://doi.org/10.3390/su141912579

Omotilewa, O. J., Ricker-Gilbert, J., Ainembabazi, J. H., & Shively, G. E. (2018). Does improved storage technology promote modern input use and food security? Evidence from a randomized trial in Uganda. Journal of Development Economics, 135(July), 176–198. https://doi.org/10.1016/j.jdeveco.2018.07.006

PICS, 2024. Purdue Improved Crop Storage (PICS) Program. https://ag.purdue.edu/department/entm/extension/pics-network/_media/4-147-pics-rack-card-digital-10-20-24.pdf

Rabé, M. M., Baoua, I. B., & Baributsa, D. (2021). Adoption and Profitability of the Purdue Improved Crop Storage Technology for Grain Storage in the South-Central Regions of Niger. Agronomy, 11(12), 2470. https://doi.org/10.3390/agronomy11122470

Ricker-Gilbert, J., Moussa, B., & Abdoulaye, T. (2021). Using Branding to Signal Quality in Informal Markets. Evidence from an Experimental Auction in the Sahel. AgEcon Search, 1(3), 11.

Tanager. (2018). AgResults Kenya On Farm Storage Pilot Project: Airtight (Hermetic) devices. Summary Project Report (Issue December 2018). https://agresults.org/wp-content/uploads/2024/02/AgResults-Kenya-Project-Summary-Report_Final.pdf

Tubbs, T., Baributsa, D., & Woloshuk, C. (2016). Impact of opening hermetic storage bags on grain quality, fungal growth and aflatoxin accumulation. Journal of Stored Products Research, 69, 276–281. https://doi.org/10.1016/j.jspr.2016.10.003

Villers, P., Navarro, S., & DeBruin, T. (2008). Development of hermetic storage technology in sealed flexible storage structures. In G. Daolin, S. Navarro, Y. Jian, T. Cheng, J. Zuxun, L. Yue, & W. Haipeng (Eds.), Proceedings of the 8th International Conference on Controlled Atmosphere and Fumigation in Stored Products (pp. 21–26). Sichuan Publishing Group.

Walker, S. (2017). AflaSTOP: Drying and storage for aflatoxin prevention.

Williams, S. B., Baributsa, D., & Woloshuk, C. (2014). Assessing Purdue Improved Crop Storage (PICS) bags to mitigate fungal growth and aflatoxin contamination. Journal of Stored Products Research, 59, 190–196. https://doi.org/10.1016/j.jspr.2014.08.003

Yewle, N.R.; Stroshine, R.L.; Ambrose, R.K.; & Baributsa, D. 2023. Short-term hermetic storage of wet maize and its effect on quality. Foods 12, 891. https://doi.org/10.3390/foods13050760

Zacharia, H., Feleke, S., Nyaa, M., & Baributsa, D. 2024. Adoption and impacts of hermetic storage on household food insecurity among maize producers’ organizations in Tanzania. Discover Agriculture 2:98. https://doi.org/10.1007/s44279-024-00116-0

Development and Testing of an Automated Chicken Egg Incubator

Rosamond F. Yarquah1 and Francis Kumi1

1Department of Agricultural Engineering, University of Cape Coast, Ghana.

Corresponding author: ryarquah@stu.ucc.edu.gh

KEYWORDS: automated, egg incubator, chicken, hatchability, temperature, humidity

Abstract

In this study, an automated chicken egg incubator was designed and constructed to replicate the ideal conditions needed for an egg to hatch successfully. The incubator system had a Raspberry Pi microcontroller integrated to automate the process. The DHT11 sensors in the system monitor the temperature and humidity readings and the bulb is turned on and off based on their readings. When the temperature readings exceed the threshold placed, the light bulbs go off and the servo motors open the vents to allow warm air out of the system while the DC fans also turn on to blow out the excess heat through the vents on the back of the frame. This is done to avoid overheating the eggs. The data gathered from sensors are pushed to the Adafruit IO cloud services for real-time visualization and monitoring. The temperature recorded was from 35oC to 38.5oC. Also, the humidities ranged from 44.5% to 55% for the first 19 days after which it fell to 47% for the remaining two days due to erratic power supply experienced during this period, leading to the non-hatchability of the eggs. Nonetheless, the results give a good indication that the system would work efficiently under a stable power supply, and it is being planned to provide it with a solar power system.

Introduction

The poultry industry is one of the productive industries in the agricultural sector in Ghana as it is a major source of income for the Ghanaian farmer and its products are a major source of protein in our diets (Adei & Asante, 2012). Day-old chicks are needed to start any poultry business, therefore commercial businesses depend on hatcheries for their day-old chicks while subsistence farmers depend on broody hens for their chicks (Kingori, 2011). In Ghana, although there are local hatcheries that produce day-old chicks, the quality is generally low and, as a result, most poultry farmers prefer to buy the imported day-old chicks (Boschloo, 2019). Only a few farms can afford the importation of day-old chicks due to their high price.

Thus, the availability of chicken egg incubators could be very crucial for producing day-old chicks locally instead of importing them into the country or relying on the natural means of having the mother hen hatch eggs from time to time. Mostly, farmers that allow the hen to brood on the eggs lose several chicks, as the hen spends the time needed to lay more eggs to brood her eggs. A chicken egg incubator creates the right conditions needed for an egg to hatch successfully. Chicken egg incubators can help to improve poultry production through a regular flow in income and by helping these farmers become possible rural entrepreneurs (Azahar et al., 2020).

An automated chicken egg incubator is even more beneficial in obtaining a high rate of hatchability due to its ability to retain the optimum temperature needed for hatching. In doing so, temperature is one of the most important factors in determining or influencing embryo development and hatchability (Decuypere & Michels, 1992). In Ghana, there are a number of locally fabricated egg incubators, which are often not automated, and the hatching conditions are not controlled making them unreliable to use. Several farmers are therefore craving for a technology that could help them automate the process under controlled temperature and humidity. Such a facility could also automate the turning of the eggs alongside the heat and humidity which are suitable conditions needed to ensure a high hatchability rate (Mohammed-Nurudeen, 2020).

It has been observed that small-scale poultry farmers rely on natural incubation or commercial hatcheries for young birds for breeding making them unable to expand their business (Kyeremeh & Peprah, 2017), as most of them are unable to purchase imported chicken egg incubators due to their high prices. Moreover, those who build chicken egg incubators are unable to regulate their temperatures due to low levels or no automation (Mohammed-Nurudeen, 2020). This study therefore aimed to develop and test an automated chicken egg incubator for smallholding poultry farmers in Ghana.

Materials and Methodology

Materials

The materials used for the construction of the chicken egg incubator includedwood,clear glass, Raspberry Pi microcontroller board, stepper motors, DC motor fans, servo motors, incandescent bulbs, DHT11 temperature and humidity sensors, Liquid Crystal Display (LCD), electronic relays, breadboards, screws, and nails. The Python Programming Language was used for writing the codes for controlling the system and stored on Adafruit.io Cloud.

Methodology

The incubator was designed using SOLIDWORKS software. The incubator was a wooden rectangular box of length x height of 80 cm x 100 cm, and a thickness of 80 cm (Figure 1), having a stand and a glass door to give a clear vision for monitoring the activities going on inside the system. The frame of the incubator system has three compartments. The first two compartments hold the trays that have the eggs. These trays consist of racks which can move in one dimension, and this causes the eggs to roll when the motors move them (Figure 1). Each tray has a capacity of 50-100 eggs. The last compartment holds the water trough, to regulate the humidity level of the system. Two holes of 0.5 cm radius were drilled on the chamber wall for ventilation with a 45 cm distance between them.

The incubator chamber contained electronic gadgets fixed in to control the automation (Figure 2). The Raspberry Pi was used as the main control unit to which all the other electronic components in the system were connected and given specific tasks to perform through code. The stepper motors were used to turn the egg trays from side to side which in turn rotated the individual eggs put in between the tray plates, to evenly distribute the heat over the eggs. However, getting the motors to move the trays was a bit of a hassle because the friction caused by the two wood pieces rubbing against each other was much greater than the strength of the motors. Papers were therefore put under the racks to reduce the friction. Also, for the top compartment, rubber ties were used to elevate the rack a little to further reduce the friction.

The DC motors were used to cool down and circulate the heat produced by the bulbs through the system to ensure an even distribution of the temperature; the fans remove excess heat with the help of the servo motors that cover the vents on the back of the system and open them to allow warm air out. Two incandescent light bulbs were used in the top and bottom trays as the main source of warmth for the eggs.

The DHT11 sensors were used to monitor and collect three different temperature and humidity readings from the system, two on the inside, thus the top and bottom trays and one on the outermost part of the system. This was done to compare the temperature difference on the outside and inside. When the heat exceeds a threshold of 39oC, the light bulbs go off and the 220V DC fans are turned on to blow out the excess heat through the vents at the back of the frame. The 16×2 LCD served as the main monitoring point for the temperature and humidity values. Relays were used to convert small electrical input into high-current output, due to the light bulbs and fans which require high voltage, that the microcontroller cannot provide. Breadboards were used for building temporary circuits. They were used to design and test the initial circuit built, allowing for components to be removed and replaced easily to demonstrate their action, and then to reuse the components in another circuit. The nails and screws were used to provide support for the wooden frame and all other components that require support on the side of the system, that is the DC fans, stepper motors, and bulbs (Figure 2).

The Raspberry Pi comes with Python pre-installed which was used to program and control the sensors and actuators used in this project. Adafruit IO platform was used to visualize and monitor the incubator temperature and humidity readings in real time over the internet.

Figure 1. Design of the box, door and egg tray (Measurements in mm)

Figure 2. Pictorial view of the Automated Chicken Egg Incubator.

Results and Discussion

From the results, Figure 3 (b) shows that the incubator chamber was mostly kept within the temperature range of 35oC to 38.5oC. It could be observed from the humidity levels (Figure 3(a)) that the values in the incubator increased from the initial 44.5% to 55% for the first 19 days after which it fell to 47%. The humidity values were not consistent due to the major challenge of power outages experienced during the incubation period. Power outages is a problem normally faced in the country. In undertaking this project, a huge challenge faced was the frequent and long-duration power fluctuations with no alternative backup power. When there is a power cut, the whole system completely goes off, and temperature and humidity levels in the incubator chamber drop. It was also observed that when the temperature was low, the heat was not enough to vaporize the water leading to low humidity levels on some days. The outer temperature of the incubator chamber over the 21

Figure 3. (a) The inner and outer relative humidity (%) and (b) The inner and outer temperature (oC) of the incubator chamber.

days, varied between a minimum of 29°C and a maximum of 33°C, even though the inner temperature of the incubator remained relatively constant, varying between 35.5°C and 38.5°C. This indicates that heat lost by conduction through the walls of the incubator was minimal. Researchers recommend that the ideal humidity level should be about 50-55% for the first 18 days of incubation, and for the final 3 days, about 65% (Ogunwande et al., 2015).

This study yielded no hatchability. This could have resulted from the drop in humidity and temperature at certain points which led to the dead developing embryos. Other parameters that could have contributed to this performance include poor storage of the eggs, before loading them inside the incubator (Mansaray & Yansaneh, 2015). The embryonic mortalities may have also been due to the presence of fungal or bacterial infection on the eggs or in the incubator chamber, or the presence of micro cracks on the eggshells through which the embryos may have been vulnerable.

Conclusion

An automated chicken egg incubator was designed, constructed, and tested using locally sourced materials to make it relatively affordable to the average poor farmers, actively engaged in the poultry industry in Ghana. The temperature recorded (35oC to 38.5oC) fell within an acceptable range. However, humidity problems especially at the final stages of the hatching process (days 19 onwards) were encountered mainly due to frequent and long-duration power outages resulting in no hatchability. A follow-up study is therefore underway to solve the problems of humidity levels encountered to ensure the successful hatching of fertile eggs in the next test.

Acknowledgements

The completion of this project could not have been possible without the participation and assistance of so many people whose names may not all be mentioned. A big thank you goes to Mr. Isaac Mensah, Mr. Jason Appiatu, and their team from the Department of Computer Science and Information Technology at the University of Cape Coast, Ghana, for providing the resources, time, and support to ensure the success of this project.

References

Adei, D., & Asante, B. K. (2012). The challenges and prospects of the poultry industry in Dormaa District. Journal of Science and Technology, 32(1),104-116.

Azahar, K. B., Sekudan, E. E., & Azhar, A. M. (2020). Intelligent Egg Incubator. International Journal of Recent Technology and Applied Science, 2(2), 91-102.

Boschloo, R. (2019). Analysis poultry sector Ghana 2019. Retrieved from https://www.rvo.nl/sites/default/files/2019/12/Update-poultry-report-ghana-2019.pdf

Decuypere, E., & Michels, H. (1992). Incubation temperature as a management tool: A review. World's Poultry Science Journal, 48(1), 28-38.

Kingori, A.M. (2011). Review of the factors that influence egg fertility and hatchability in poultry. International Journal of Poultry Science,10(6), 483-492.

Kyeremeh, F., & Peprah, F. (2017). Design and construction of an Arduino microcontroller-based egg incubator. International Journal of Computer Applications, 168(1), 15-23.

Mansaray, K. G., & Yansaneh, O. (2015). Fabrication and performance evaluation of a solar-powered chicken egg incubator. International Journal of Emerging Technology and Advanced Engineering, 5(6), 31-36.

Mohammed-Nurudeen, M. (2020). A local poultry farmer comes to the rescue of colleagues with a locally made incubator. My JoyOnline. Retrieved February 08, 2022, from https://www.myjoyonline.com/local-poultry-farmer-comes-to-rescue-colleagues-with-locaaly-made-incubator/

Ogunwande, G. A., Akinola, E. O., & Lana, A. R. (2015). Development of a biogas-powered poultry egg incubator. Ife Journal of Science, 17(1), 219-22.

Mechanical Cassava Harvester Technology Development and Commercialization in Africa – Challenges and Prospects

Bobobee Y.H. Emmanuel1

1Kwame Nkrumah University of Science & Technology, Kumasi.

Corresponding author: emmanuel.bobobee@gmail.com; emmanuel.bobobee@knust.edu.gh

URL: www.tekcassavaharvester.com

KEYWORDS. smallholder farmers, mechanical cassava harvesters, ridge landform, conservation, precision agriculture

Abstract

Cassava (Manihot esculenta Crantz) is a nutritious carbohydrate source and an important crop after maize and rice. Sub-Saharan Africa produces over 64% of global cassava, an essential source of food and income. Cassava provides a livelihood for nearly one billion farmers, countless off-takers, processors, and traders worldwide. Cassava in Africa is a traditional food security, and an emerging profitable industry crop dominated by smallholder farmers, who use traditional tools and methods of production. Cassava harvesting all year round when planted randomly, is a challenge to its industrial processing and for export. Manual harvesting during the dry season when the ground is hard is slow and associated with physical exertion and high root damage. A mechanical harvesting device to break the labour bottleneck and physical pain associated with manual harvesting is needed. Research on mechanised cassava production, especially harvesting is very slow, and until recently there existed no known commercial mechanical cassava harvesters developed by Africans for African farmers. The main objective of this paper was to introduce and deploy a tractor-mounted cassava harvester developed at the Kwame Nkrumah University of Science and Technology in Ghana.

The device is light (300 kg), cuts soil one meter wide, and can be pulled by any medium horsepower-rated tractor (60-75 hp) available to most farmers and tractor owners in Africa. In operation, the harvester penetrates beneath the root cluster to depths of 25 to 39 cm and develops an average draught force of 10 - 11 kN at a tractor speed of 5 km/h, and a wheel slip of 9-11 percent. The device uproots best on ridged fields when the ground is hard at a rate of one plant per second or less producing a field capacity of 1.9 - 2.5 h/ha and tuber damage of 5-10 percent. Manual harvesting in hard soils takes 5-10 minutes to uproot a plant and requires a mandatory rest period of 30 - 45 mins/h and a harvesting capacity of 60–104 person-days per hectare. After mechanical harvesting, the field is left ploughed because of the deep excavation. This generates additional savings on fuel, time, production cost, and other resources for the next season's crop. The field capacity and efficiency of the TEK MCH compare favourably with other single-row harvesters in other regions of the world. Ridge planting is recommended as it gives higher root yield, low root damage, and is compliant with conservation and precision agricultural practices, and climate action (SGD 13).

Introduction

Cassava (Manihot esculenta Crantz) is one of the most significant crops in the world, and the fifth largest staple crop worldwide (Parmer et al. 2017).  Cassava is important due to its extensive product development lines and importance within the human diet in different tropical and subtropical regions (Otálora et al., 2024). The crop is a major source of flour and starch suitable for numerous industrial applications and animal feed.  Global cassava production in 2022 after Adebayo (2023), was estimated at 300 million tonnes with Africa, Asia, South America, and Oceania producing 64%, 27%, 8% and, 8%, respectively on over 26 million hectares. Vilpoux et al. (2017) reported cassava with the highest average annual growth of approximately 3% in the last decade, is Africa’s most important tuberous crop serving as an all year round cheap and reliable food security staple on the continent. However, cassava production in Africa is dominated by smallholder farmers, whose operations are influenced by using low input technologies and methods leading to low yield per hectare (Wahab et al., 2022). Other challenges faced by the smallholder cassava farmers include traditional random planting not compliant with mechanisation, treating agriculture as a way of life, not as a business, an aging agricultural labour force, aversion of the youth to take agriculture as a business, and dry season harvesting that is a predominant, slow, painful, and stressful activity full of drudgery (Bobobee et al., 2019). Thailand, which adopts mechanised cassava production strategies is the third-leading world cassava producer, and world exporter of cassava products with an export share of 42.1% in 2020 (Onyediako & Adiele. 2022).

According to Abass et al. (2011), cassava is one of the most consumed staple food crops in sub-Sahara Africa with a per capita consumption of 105 kg/year. In Ghana, cassava is an important food crop accounting for 153 kg per capita consumption and substantially contributing nearly 22% to the national agricultural gross domestic product (Bayitse, et al., 2017). Growing high-yielding cassava varieties can spur sustainable developments as recognized by Otekunrin & Sawicka (2019). Paying attention to the development of high-yielding elite varieties will promote traditional starchy staples for food security, value addition, and industrial processing. However, the dominance of the smallholder farmers and their adherence to the age-old traditional cultivation in a haphazard manner prevents the majority from enjoying the full benefits of any modern mechanised operations. Traditional smallholder farmers have average farm holdings of less than 0.5 ha with yields varying considerably from location to location in the same agroecological zone. Marechera & Muinga (2017) further confirmed that cassava root yields are determined by a combination of ecological factors and the availability of labour-saving methodologies and engineering technologies for production and processing.

Among the cassava cultivation tasks from seedbed preparation to harvesting, mechanization of cassava harvesting has been identified by cassava researchers as an important and promising area for intervention to realize the full potential of this crop (Marechera, & Muinga, 2017).

Manual cassava harvesting requires approximately 53–100 person-days per hectare because it is slow and associated with drudgery and high root damage. (Bobobee et al., 2018; Amponsah et al., 2014 and Nweke et al., 2002). This tends to increase the total cost of production because more farmhands are required to harvest to meet industrial demands during dry periods. Peipp and Maehnert (1992) also noted that the most difficult operation in cassava production is manual harvesting. Without question, a mechanical revolution is now needed to break the labour bottleneck in cassava production and harvesting among African farmers, who are planting improved varieties (Nweke et al., 2002). Until recently, there were no commercial mechanical cassava harvesters developed in Africa for Africans.

Objective

The main objective of this paper was to introduce and demonstrate a tractor-mounted mechanical cassava harvester, christened TEK Mechanical Cassava Harvester, developed at the Kwame Nkrumah University of Science and Technology in Ghana, for use by smallholder cassava farmers in Africa. Specifically, for the device to benefit smallholder farmers, they must change their traditional haphazard planting to planting in rows or on ridges.

The TEK Mechanical Cassava Harvester (TEK MCH)

The performance of the TEK mechanical harvester (Figure 1), developed at the Department of Agricultural and Biosystem Engineering of the Kwame Nkrumah University Science and Technology (KNUST), Kumasi, Ghana, was discussed in greater detail by Bobobee et al. (2018) and Amponsah, et al. (2014).

Figure 1. The TEK MCH is being deployed in the field.

For the harvester to work satisfactorily and benefit the end-users, especially the smallholder farmers and tractor owners, a major requirement is that cassava planting must transition from the traditional haphazard and random format to planting in lines preferably on ridges. as shown in Figure 2. The ridges must be wide apart to accommodate the track width of the tractor used to prepare the seedbed. To obtain the recommended planting population density of 10,000 plants/ha on the ridge, the inter-row spacing along the ridges should be the reciprocal of the tractor track width. Ennin et al., (2009), confirmed that planting on ridges had the advantage of higher cassava root yield and has the potential for mechanization to further decrease drudgery and increase the scale of production compared to planting on the flat. The harvester was tested in Ghana and deployed in Ivory Coast, Jamaica, Nigeria, and South Africa. More cassava farmers in other parts of the continent and the world are showing interest in its use. In addition to cassava, the implement harvests yam and sweet potato easily provided they are planted in line or on ridges.

Figure 2. Traditional haphazard planting (a) versus the recommended ridge planting (b).

Ridge cultivation is recommended because of the many advantages such as (a) higher yield since the loose soils of the ridges help the roots to develop rapidly before bulking, (b) rapid evacuation of excess moisture during heavy rainstorms, and retention of little precipitation during the dry season in the furrows (Figure 3), and (c) serving as a guide to the operator during mechanical harvesting to minimize tuber damage. When planted on the flat, the stems of mature cassava take different orientations, and they are not normally in a straight line thus can compound root damage during mechanical harvest.

Figure. 3. Ridging and conservation agricultural benefits for cassava.

Heart rate measurement to calibrate drudgery

The heart rate is used as a non-invasive approach to calibrate drudgery in doing physical activities including manual harvesting (Bobobee et al. 2019; Bobobee and Gebresenbet. 2007). The use of the Polar Heart Rate sensor to measure the resting, working, and recovery heart rates of manual cassava harvesters has provided sufficient evidence to confirm that manual cassava harvesting is a painful activity demanding longer rest periods for the worker to recuperate as shown in Figure 4. The higher heart rate profiles in Figure 4, were those of two women harvesting 10 plants each, and they took approximately 70 minutes per person to accomplish their tasks. The lower heart rate profiles in Figure 4, were for the tractor operator when he harvested close to half a hectare on the same day as the women and harrowed the same field during seedbed preparation. This confirms that the tractor operator experienced similar stresses for both operations, which were lower than those of the manual harvesters.

Figure 4. Heart rate profiles of manual and mechanical cassava harvesting operations.

Compared with manual harvesting, the heart rates of tractor operators are considerably lower enabling them to perform harvesting tasks over larger tracts of cassava fields when manual harvesters spend over 70 minutes each to uproot 10 cassava stems. This high amplification of work by mechanised tractor operators, enables cassava to be harvested easily, especially during the dry season to support food security and rapid processing.

One major advantage of mechanical cassava harvesting is that the deep excavation of the cassava roots leaves the field ploughed (Figure 5), necessitating secondary seedbed preparation to establish the next season's crops.

Figure 5. Mechanical cassava harvesting leaves the field ploughed.

The performance characteristics of the TEK MCH with similar tillage implements are shown in Table 1. The depth of penetration of the harvester is normally higher than ploughing. To avoid tuber damage, the digging blade penetrates deeper beneath the root cluster to excavate and expose it for easy collection. Apart from the high depth of penetration and wheel slip, there are no significant differences (p<0.05) for all other performance characteristics of the harvester and the mouldboard plough. The harvester performs like any primary tillage implements. The disc harrow, a secondary tillage implement, is wider but works shallower. This confirms that the harvester does not require a dedicated high-power-rated tractor to pull it. This is an advantage of the TEK MCH so that smallholder farmers and tractor owners on the continent can afford to invest in the implement to boost cassava production and harvesting.

Table 1. Comparison of the performance of the Mouldboard Plough, Disc harrow, and the TEK MCH

ParameterMouldboard PloughDisc HarrowTEK MCH
Speed (km/h)5.359.055.01
Fuel rate (l/h)21.9420.8518.49
Specific fuel consumption (l/ha)22.9213.0020.56
Work rate (ha/day)9.5616.299.01
Engine power (kW)57.9653.3953.42
Drawbar power (kW)26.0220.3226.75
Wheel ship (%)9.111.6216.12
Working depth (mm)250140320

Parameter Mouldboard Plough Disc Harrow TEK MCH

Table 2 highlights the performance of cassava harvesters used in different regions. Apart from the two-row automatic harvester from Thailand having a higher capacity, the field capacity and efficiency of the TEK MCH compare favourably with existing single-row cassava harvesters from other cassava growing areas in the world.

Table 2. Performance of cassava harvesters used in different regions

CountryType of HarvesterPerformance MetricsReferences
NigeriaTractor-mounted mechanical harvesterEfficiency: 60–75% field efficiency, capacity: 0.25–0.4 ha/hAdekanye et al., 2017 or 2013
ThailandTwo-row automatic harvesterCapacity: 0.8 ha/h, field efficiency: 85%Wanitchang et al., 2015
BrazilAnimal-drawn cassava lifterCapacity: 0.1 ha/h, reduction in labor costs by 40%Da Silva et al., 2010
VietnamRotary digger harvesterFuel consumption: 8 L/ha, field capacity: 0.3 ha/hDo et al., 2015
GhanaTEK MCHCapacity: 0.36–0.48 ha/h, field efficiency: 60%Bobobee et al., 2019
IndiaSelf-propelled harvesterField capacity: 0.6 ha/h, labor reduction: 70%Kumar et al., 2018

Conclusions

The TEK mechanical cassava harvester has been developed, tested, and deployed in Ghana and other cassava-producing countries in Africa and beyond. The harvester is light, generates the same drawbar power as the plough, and can be pulled by existing medium-power-rated tractors owned by tractor owners and used by smallholder farmers on the continent. The implement harvests at a rate of one plant per second or less compared to 5-10 minutes per plant for manual harvesting, especially when the ground is hard. The harvester works best when the cassava is planted in line or on ridges to allow the tractor to pass between the rows.  The field capacity and efficiency of the TEK MCH compare favourably with other single-row harvesters in other regions of the world. The adoption and use of the harvester will reduce drudgery, modernise cassava production and make the agricultural sector attractive to young people to take it up as a business instead of the way of life at present on the continent. The mechanical cassava harvester will unlock the huge potential of the crop on the continent.

It is recommended that Africa mechanises cassava production to develop it from a food security crop to a multi-billion dollar industrial and export commodity on the continent. Efforts need to be intensified to train tractor operators and smallholder farmers, who are in the majority to adopt the mechanised production methodology and technology to reduce drudgery and promote the device in sub-Saharan Africa and the world. This is an emerging technology, and its use must be supported by governments, development partners, and other stakeholders to align with Sustainable Development Goals (SDGs) 1, 2, 5, 9 & 13, and Africa Agenda 2063 – The Africa we want.

References

Abass, A. B., Bokanga, M., Dixon, A., & Bramel, P. (2011). Transiting cassava urban food and industrial commodity through agro-processing and market-driven approaches: lessons from Africa. policies and institutions in support of agro-industries development, 305-352.

Adebayo, W. G. (2023). Cassava production in Africa: A panel analysis of the drivers and trends. Heliyon, 9(9). e19939.

Adekanye, T. A., Ogunjimi, S. I, & Ajala, A. O. (2013). An Assessment of Cassava Processing Plants in Irepodun Local Government Areas, Kwara State, Nigeria. World Journal of Agricultural Research, 1 (1): 14-17.

Amponsah, S. K., Bobobee, E. Y., Agyare, W. A., Okyere, J. B., Aveyire, J., King, S. R., & Sarkodie-Addo, J. (2014). Mechanical cassava harvesting as influenced by seedbed preparation and cassava variety. Applied Engineering in Agriculture, 30(3), 391-403.

Bayitse, R., Tornyie, F., & Bjerre, A. B. (2017). Cassava cultivation, processing and potential uses in Ghana. Handbook on Cassava, 313-333.

Bobobee, E. Y. H., Yakanu, P. N., Marenya, M. O., & Ochanda, J. P. O. (2019). The TEK Mechanical Cassava Harvester Development in Ghana–Challenges, Opportunities and Prospects for Cassava Production in Africa. Journal of Engineering in Agriculture and the Environment.5(1), 41-60.

Bobobee, E. Y., & Gebresenbet, G. (2007). Effect of cutting-edge thickness and state of wear of ploughshare on draught force and heart rates of Sanga oxen in Ghana. Soil and Tillage Research, 95(1-2), 298-307.

Da Silva, M. A., Filho, W.F., Santos, L. N., & Souza, F. R. (2010). Evaluation of Mechanized Cassava Harvesting Systems in Brazil. Journal of Agricultural Engineering Research, 50(4):345-353.

Do, T. V., Huang, N. M., & Nguyen, P. H. (2015). Development of Cassava Harvesting Techniques in Vietnam. International Journal of Agricultural Engineering, 67(2):215-223.

Ennin, S. A., Otoo, E., & Tetteh, F. M. (2009). Ridging, a mechanized alternative to mounding for yam and cassava production. West African Journal of Applied Ecology, 15(1).

Kumar, R., Patel, A., & Singh, V. (2018). Design and Development of Semi-Mechanised Cassava Harvester for Small-Scale Farmers. Journal of Agricultural Machinery Sciences, 14(3):120-130.

Marechera, G., & Muinga, G. (2017). Value chain approaches to mechanization in cassava cultivation and harvesting in Africa Foundation (AATF), Kenya. In Achieving sustainable cultivation of cassava Volume 1 (pp. 393-416). Burleigh Dodds Science Publishing.

Nweke, F., Spencer, D. S., & Lynam, J. K. (2002). The cassava transformation: Africa’s best kept secret. International Institute for Tropical Agriculture. Michigan State University Press, East Lansing, MI.

Onyediako, P. O., & Adiele, J. G. (2022). Enhanced Cassava Production for Food Security and Economic Development in Nigeria: A review. Nigerian Agricultural Journal. 53(3), 204-211.

Otálora A., Garces-Villegas V., Chamorro A., Palencia M., & Combatt E.M. (2024).Cassava, manioc or yuca (Manihot esculenta): An overview about its crop, economic aspects and nutritional relevance. J. Sci. Technol. Appl., 16, Art-95: 1-10. https://doi.org/10.34294/j.jsta.24.16.95

Otekunrin, O. A., & Sawicka, B. (2019). Cassava, a 21st century staple crop: How can Nigeria harness its enormous trade potentials. Acta Scientific Agriculture, 3(8), 194-202.

Parmar, A., Sturm, B. & Hensel, O. (2017). Crops that feed the world: Production and improvement of cassava for food, feed, and industrial uses. Food Sec., 9, 907–927. https://doi.org/10.1007/s12571-017-0717-8

Peipp, L., Mahnert, E., & Neumann, R. (1992). The development of a technical solution to cassava harvesting problem. Agric. Mech. Asia Afr. Lat. Amer. 23, 33–36.

Vilpoux, O. F., Guilherme, D. D. O., & Cereda, M. P. (2017). Cassava cultivation in Latin America (pp. 149-174). Burleigh Dodds Science Publishing.

Wahab, I., Hall, O., & Jirström, M. (2022). The maize is the cost of the farming, and the cassava is our profit: smallholders’ perceptions and attitudes to poor crop patches in the eastern region of Ghana. Agriculture & Food Security, 11(1), 14.

Wanitchang, J., Prasertsak, A., & Phakamas, N. (2013). Development of Cassava Harvesting Machinery for Sustainable Agriculture in Thailand. Journal of Agricultural Technology, 9(2):289-299.

Recirculating aquaculture System (RAS): The Future of Fish Farming in Developing Countries

Shadrack Kwadwo Amponsah1* Helena Asare1, Divine D. Azumah1

1CSIR-Crops Research Institute

*Corresponding author email: skamponsah@hotmail.com

KEYWORDS: aquaculture, recirculating, catfish, dissolved oxygen, Wontesty Ventures, Ghana

Introduction

With 8 billion people on the planet, demand for aquatic food is rising. Horizontal and vertical expansions of aquaculture are recommended to meet this demand. The FAO suggests increasing fishery output by 50% by 2030 to counter protein shortages. Declining fish supply from capture fisheries cannot meet the growing population's needs. Aquaculture, growing at 6% annually, already provides over half the world's fish (FAO, 2020, 2022)

Recirculating aquaculture systems (RAS) have become viable over four decades. Denmark pioneered RAS in 1940 for commercial eel aquaculture (Espinal & Matulic, 2019). Japan began RAS research in the 1950s, focusing on carp production. RAS is now used globally for various species like trout, salmon, and tilapia. Collaborative research efforts in 2014 between the CSIR-Crops Research Institute and Embrapa Mid-North led to the first successful implementation of the innovative RAS system of tilapia culture in Ghana (Amponsah, 2018). Since then, the RAS has been successfully adopted for catfish rearing across different regions of Ghana, led by Wontesty Ventures, a Ghanaian agribusiness that specializes in the local adaptation of RAS. Sustainable system intensification requires more environmentally responsible management.

RAS technology could revolutionize the aquaculture industry, unfortunately, there is a considerable lack of information, and this is especially true in developing nations. It may be helpful to break the ice by gaining an understanding of the design and operation principles underlying the RAS in order to stimulate an appetite for additional research into this innovative production system.

What is RAS?

Recirculating aquaculture systems (RAS), also known as intensive tank-based fish farming, utilize biofiltration and water exchange to minimize freshwater usage and maintain a healthy fish environment. These systems employ strict environmental control, applicable in both home aquariums and commercial fish production. Ebeling & Timmons (2012) emphasize their reliance on biofiltration due to limited water exchange, necessary for ammonia toxicity mitigation. RAS technology employs mechanical and biological filters to recycle water, removing organic waste (biosolids) and sustaining optimal conditions for fish growth.

RAS Design and Operating Principles

Unlike fishponds and cages, which have remained largely unchanged in design, recirculating aquaculture systems (RAS) are diverse and continuously evolving globally. RAS design encompasses key processes like mechanical and biological filtration, oxygenation, and water recirculation. These systems typically include mechanisms to remove solid waste, nitrifying biofilters to convert ammonia to nitrate, and gas exchange devices for carbon dioxide removal and oxygenation (Figure 1).

Figure 1. Simplified water treatment flow diagram of the RAS (Source: Schmelmer, 20161)

Some high-tech RAS designs (Figure 2) may adopt UV irradiation for water disinfection, ozonation and protein skimming for fine solids and microbial control and denitrification systems to remove nitrate.

Figure 2. High-tech RAS design with UV disinfection (Source: Bregnballe, 20222).

Benefits of RAS

Recirculating aquaculture systems (RAS) offer numerous benefits. They minimize environmental disturbance by reducing nutrient pollution discharge and lowering the risk of pathogenic infestations from external sources, thus enhancing fish health. RAS can operate in water-scarce or poor-quality areas, reusing about 90% of water through treatment loops. Compared to conventional tank culture, RAS technology replaces only 10% of total volume daily, resulting in significant water savings (Twarowska et al., 1997). Advocated as a key innovation in the food sector, RAS contributes significantly to sustainable development (Meisch & Stark, 2019), offering adaptability to various climates and proximity to population centers and markets.

Problems with the RAS

The downsides of the RAS include high initial costs and high risks due to high stocking densities. High operating costs are mostly due to electricity, and system maintenance. RAS systems have also been associated with high greenhouse gas emissions (Tilman & Clark, 2014). The need for highly trained staff to manage the systems has also been cited. Perhaps, these and other operational factors have contributed to the low adoption of RAS in most developing countries.

Harnessing the Potential of RAS: The Ghanaian Case Study

Ghana's aquaculture sector contributes nearly 5% to GDP and employs approximately 10% of the population (Statista, 2021). However, the country faces challenges due to declining fish supply from capture fisheries, exacerbated by population growth and food insecurity. Ghana experiences a significant fish production deficit, resulting in an annual fish import bill of $311 million (Damalie, 2021; USDA, 2019). Inland aquaculture meets only 13% of the total annual fish demand, utilizing various production systems such as concrete tanks, earthen ponds, and floating cages. While cage culture dominates tilapia production, tank culture systems have untapped potential to enhance Ghana's aquaculture sector.

Since the partnership with Embrapa in 2014, which led to the first known introduction of RAS into the country, a Ghanaian Agricultural Engineering Researcher has elevated the adaptation and promotion of RAS in the country. Through local materials and expertise, the initial RAS design has evolved into an effective recirculation system for cultivating freshwater species like African catfish (Clarias gariepinus) and Nile tilapia (Oreochromis niloticus). Figure 3 illustrates aspects of the RAS adaptation process from the original Brazilian design.

Figure 3. The evolution of the RAS design: (a) Initial Brazilian RAS on a concrete tank in 2014 (b) Adapted RAS design on a mobile tarpaulin tank in 2019 (c) An improved RAS design with PVC suction mat installed on a mobile tarpaulin tank (2021 – date).

The two Ghanaian RAS adaptations (Figure 3(b) and (c)) differ from their Brazilian counterpart in the suspension of the biological filter and the waste collection tank material. Figure 3(b) employs water hoses to bring water into the waste bucket from the tank's top, while Figure 3(c) utilizes PVC suction mats from the bottom via capillary action. Figure 4 depicts the evolution of the biological filter design since 2014, with Figure 4(c) serving as an upgrade to address biosolids accumulation discovered in Figure 4(b). This highlights the demand-driven approach in the evolution of the locally adapted RAS.

Figure 4. Local adaptation of the biological filter: (a) initial design in 2014 (b) 2019 design and (c) improved design in 2022

The locally adapted RAS efficiently cultivates 1000 African Catfish juveniles (=10g) to an average of 1kg weight in 4-6 months using around 500 kg of fish feed. Fish tanks, which can be permanent or mobile (Amponsah et al., 2024) and customized to any size, are constructed from various materials like wood, cement blocks, or tarpaulin canvas, and can have different shapes (circular, rectangular, ellipsoid). The RAS design is energy efficient, requiring only 100 Watts of electricity to power two submersible pumps for aerating and recirculating pond water.

The local adaptation and research approach on RAS have spurred the establishment of an agribusiness specializing in constructing RAS fishponds, aiming to pond every household in Ghana by 2040. Wontesty Ventures has installed more than 450 RAS fishponds across the country since 2017. This technology has benefited more than 120 users, both directly and indirectly, and produces an average of 300 tonnes of catfish per year. Wontesty Ventures' fish value addition campaign, initiated in 2018, has expanded the market for African Catfish in Ghana, leading to the emergence of numerous 'point and grill' restaurants in major towns and cities. These restaurants, reliant on aquapreneurs using RAS fishponds, contribute significantly to income generation and food supply, particularly for the young and elderly, making aquaculture production in Ghana highly profitable despite rising fish feed prices.

The Ghanaian adaptations to the RAS design offer several advantages. The use of locally sourced materials and simplified construction techniques significantly reduces the cost of the system, making it more accessible to small-scale farmers. Additionally, the reliance on local expertise for construction and maintenance promotes self-sufficiency and reduces dependence on external technical support. The adaptability of the system to various tank sizes and shapes further enhances its versatility, allowing farmers to customize the system to their specific needs and resources.

Conclusion

RAS represents not only a technological innovation but also a strategic tool for achieving broader development goals. By promoting the adoption of RAS, governments can enhance domestic fish production, reduce reliance on imports, create employment opportunities, and improve livelihoods, particularly in rural communities. Furthermore, RAS can play a crucial role in promoting sustainable agricultural practices and environmental conservation. Integrating RAS into national agricultural policies and development strategies can unlock its full potential to contribute to food security, economic growth, and environmental sustainability.

Ghana and other sub-Saharan African countries have favorable conditions for developing and adopting RAS as a viable aquaculture production system. The challenges in the aquaculture industry present an opportunity to innovate and reshape the narrative. Food and nutritional security are critical, especially in the face of emerging pandemics like COVID-19. Therefore, leveraging appropriate technology to revolutionize the aquaculture sector is essential. Promoting locally adapted RAS is crucial as it enables year-round fish production using readily available local resources. This is a call to action for researchers in Agricultural Engineering and related fields to develop demand-driven solutions for the aquaculture sector, particularly in developing countries.

To promote wider adoption of RAS in developing countries, it is crucial to focus on reducing costs, simplifying designs, and promoting local capacity building. The use of locally available materials and simplified construction techniques can make the system more affordable and accessible to small-scale farmers. Additionally, government support in the form of subsidies, tax incentives, and access to credit can further incentivize the adoption of RAS technology. By empowering farmers with the knowledge and resources to implement and manage RAS systems effectively, we can unlock the full potential of this technology to enhance food security and promote sustainable aquaculture development. While local adaptation is crucial, there is also immense potential for further innovation in RAS design. Integrating renewable energy sources, such as solar power, can significantly enhance the sustainability and resilience of RAS, particularly in areas with limited access to reliable electricity. Additionally, exploring other innovations such as improved biofiltration techniques, water quality monitoring systems, and automation technologies can further optimize RAS performance and resource efficiency. By embracing a culture of continuous innovation, we can ensure that RAS technology remains at the forefront of sustainable aquaculture development.

References

Amponsah, S. K. (2018). Yali Voices: Making Aquaculture Possible with Simple Fish Tanks. Young African Leaders Initiative (YALI) Network.

Amponsah, S. K., Ameyaw, D. O., & Agyemang, S. M. (2024). Design and construction of a collapsible tarpaulin-lined pond for aquaculture production. Journal of the Ghana Institution of Engineering (JGhIE), 24(2), 8–15.

Bregnballe, J. (2022). A Guide to Recirculation Aquaculture: An Introduction to the New Environmentally Friendly and Highly Productive Closed Fish Farming Systems. Food and Agriculture Organization of the United Nations. https://books.google.com.gh/books?id=jXsIkAEACAAJ

Damalie, P. E. (2021). $150m Spent on fish imports yearly - Ghana Tuna Association. Graphic Online. https://www.graphic.com.gh/news/general-news/150m-spent-on-fish-imports-yearly-ghana-tuna-association.html

Ebeling, J. M., & Timmons, M. B. (2012). Recirculating aquaculture systems. In Aquaculture production systems, (Vol. 1, pp. 245-277.).

Espinal, C. A., & Matulic, D. (2019). Recirculating Aquaculture Technologies BT - Aquaponics Food Production Systems: Combined Aquaculture and Hydroponic Production Technologies for the Future (S. Goddek, A. Joyce, B. Kotzen, & G. M. Burnell (eds.); pp. 35–76). Springer International Publishing. https://doi.org/10.1007/978-3-030-15943-6_3

FAO. (2020). The State of World Fisheries and Aquaculture 2020. In Sustainability in action. https://doi.org/10.4060/ca9229en

FAO. (2022). The State of World Fisheries and Aquaculture 2022. In Towards Blue Transformation. Food and Agriculture Organisation of the United Nations. https://doi.org/10.4060/cc0461en

Meisch, S., & Stark, M. (2019). Recirculation Aquaculture Systems: Sustainable Innovations in Organic Food Production? Food Ethics, 4(1), 67–84. https://doi.org/10.1007/s41055-019-00054-4

Schmelmer, K. (2016). The combination of aquaponics systems and grey-water recycling - Introduction of the Greyponic system- [Tongji University Shanghai]. https://doi.org/10.13140/RG.2.2.18190.36161

Statista. (2021). Ghana: annual fish production by source. https://www.statista.com/statistics/1278935/annual-fish-production-in-ghana-by-source/

Tilman, D., & Clark, M. (2014). Global diets link environmental sustainability and human health. Nature, 515(7528), 518–522. https://doi.org/10.1038/nature13959

Twarowska, J. G., Westerman, P. W., & Losordo, T. M. (1997). Water treatment and waste characterization evaluation of an intensive recirculating fish production system. Aquacultural Engineering, 16(3), 133–147. https://doi.org/https://doi.org/10.1016/S0144-8609(96)01022-9

USDA. (2019). Ghana: Fish and Seafood Report. Foreign Agriculture Services. https://www.fas.usda.gov/data/ghana-fish-and-seafood-report

PILLAR II: BUSINESS DEVELOPMENT AND ENTREPRENEURSHIP

Structural Transformation in Agrifood Value Chains and Implications for Technology Development

Satish Joshi1*, Ajit Srivastava1

1Michigan State University, East Lansing, Michigan, USA*

Corresponding author: satish@msu.edu

KEYWORDS: structural transformation, agrifood value chains, agricultural technology, business model innovation

Introduction

The focus of traditional agricultural development policy especially in the less developed countries (LDC), has been mainly on improving on-farm food production and productivity. The primary pathways supported include improving seed varietals, increasing the use of inputs such as fertilizers, irrigation water, pesticides, and herbicides, enhancing efficiency through precision agriculture, mechanization, and loss prevention, and farmer capacity building for better farm management through training and information provision. Such agricultural development has been mostly led by governments employing policy tools such as input subsidies, input supply chain development, irrigation infrastructure development, government procurement with price supports, government-run agricultural extension services, and agricultural Research & Development (R&D).

However, the macroeconomic development process from primarily a subsistence agrarian economy towards a diversified industrial economy unleashes transformative forces characterized by growing urbanization and rural-to-urban migration, higher marginal wages in industrial/service sectors leading to growing urban demand for diverse food attributes and convenience, and the emergence of global and regional agrifood value chains (AVCs) that link rural producers to the diverse set of affluent and demanding consumers. Barrett et al (2022) identify three revolutionary changes occurring in AVCs: (a) the supermarket revolution where large-scale retailers and agribusinesses have modernized food retailing and supermarket chains have steadily increased their market share of food dollars spent; (b) the food service revolution characterized by a dramatic increase in the consumption of food away from home (FAFH) and the corresponding growth of the food service sector consisting of fast-food chains, restaurants, street vendors and institutional food service providers; and (c) revolution in the agrifood value chain intermediation evidenced by the emergence of specialized third-party logistics service providers for procurement, storage, and transportation, large scale food processing, and the evolution of global integrated food giants such as ADM, Bunge, and Cargill. These AVC transformations, mostly led by market forces and private enterprise, have received relatively less attention from policymakers and researchers. However, understanding the key drivers and accelerating AVC modernization is the key to development efforts in low- and medium-income countries. These transformations also have implications for future agricultural technology R&D.

Technologies Enabling AVC Modernization

Technologies that facilitate such modernization of AVCs are different from on-farm production technologies such as seed varietals, input use and farm mechanization that have been the focus previously. Modern AVCs aim to link rural and often global producers to sophisticated urban consumers intermediated by supermarkets and food service providers that often have very different and stringent requirements in terms of quantity, quality, consistency, timeliness, convenience, cost effectiveness, shelf life and packaging. Technologies that enable AVC modernization include: grading and sorting technologies; storage and preservation technologies including freezing, pickling, drying and use of preservatives; pre-processing and final product processing technologies; packaging technologies that meet various size, shelf life, convenience, aesthetic and recyclability specifications; transportation and logistics services; technologies that facilitate food safety, traceability and sustainability attributes and tracking technologies such as RFID and block-chains; mobile and cloud based technologies for information, monitoring, pricing and transaction/order processing; and innovative insurance and financial products for value chain risk management. Developments in information and telecommunications have also spawned business model innovations and two-sided platform technologies for shared value creation, e.g. Uber, Door Dash, and similar sharing/contracting platforms for agricultural equipment, commodity transportation and on-farm services such as plowing, spraying and harvesting. Most of the technology R&D has occurred in the developed countries where AVCs were modernized earlier. However, AVCs in low- to middle- income countries (LMIC) with their larger but poorer customer pools, larger number of diverse and smallholder farmer producers, and general resource scarcity need to develop appropriate and cost-effective AVC technologies. The ‘Jugaad’ innovation approach whose principles include, do not wait for a panacea, think frugal, be flexible and generate breakthrough growth is seen to be more effective in the LMIC contexts (Radjou, Prabhu & Ahuja 2022).

Pathways For AVC Modernization

While many of these modern agricultural value chains linking rural producers to urban consumers have evolved organically, public/private interventions have often accelerated modernization. These interventions can either be aimed at specific selected commodity value chains, or targeted at improving the general business environment, institutions and infrastructure to facilitate modern AVCs. Such interventions can be top down, e.g. supply chain development efforts by private companies like McDonalds, Walmart and Nestle during their entry into emerging markets, or bottom-up efforts where rural producer firms develop value chains and processing capabilities to link with urban/global demand.

Examples Of Successful AVC Development

One institutionally innovative and successful example of an organic evolution of a modernized AVC for a specific commodity is AMUL milk cooperative in India. AMUL employs a tiered producer owned cooperative structure consisting of village level cooperatives which collect milk from small-scale producers, a district level cooperative that handles transportation, milk processing and cattle feed operations, and a state level cooperative that centrally manages marketing and distribution operations for all district unions. AMUL currently has 3.6 million milk producers organized into 18,505 village cooperatives, and 18 district unions, and a state milk marketing federation, that process an average of 27 million liters of milk per day in 95 processing plants and is ranked the 8th largest dairy processor in the World. AMUL has a fully digitized integrated value chain that handles over 7.6 million milk procurement transactions per day and has a product portfolio of over 1000 SKUs sold through 1100 distributors resulting in annual sales of $6.7B. Its innovations such as buffalo milk processing, mobile Milk-o-bikes for milking for small farmers, balloon based localized bio-gas distribution, and industrial production of traditional Indian confectioneries demonstrate its jugaad approach.

Motivated by the success of AMUL, the Government of India has attempted to replicate the AMUL model of dairy cooperatives throughout the country is what is dubbed as the White revolution. These efforts have resulted in over 16.9 million milk producers being organized into 190,000 village cooperatives, 245-member unions and 28 state level milk marketing federations. Some state federations have emerged as serious competitors for AMUL in both size and profitability. Government of India has also launched similar programs aimed as revolutionizing other AVCs, e.g. Blue revolution (fishery), Golden revolution (horticulture and honey) and Silver revolution (poultry). Specialized educational institutions, aimed at management capacity development for such developmental organizations have been launched, e.g., Institute for Rural Management Anand (IRMA). Barrett et al (2022) provide other international examples of AVCs in various stages of development in LMICs.

Concluding Remarks

For agricultural development in LMICs simply focusing on improving farm productivity is not enough, but modernizing AVCs to link rural production to urban demand is the key. Pathways and technologies for AVC modernization need to be creatively tailored to specific value chains considering spillovers and economies of scale/scope. Institutional and business model innovations are equally important. A Jugaad innovation approach is warranted in low-/medium- income countries. Medium-income countries like India provide some examples of successful pathways for AVC modernization. Providing youth with high education/skills is important for unleashing opportunities for creative solutions to challenges.

Acknowledgments

Funding for Satish Joshi’s work was provided by USDA Hatch Grant MICL12075

References

Barrett, C. B., Reardon, T., Swinnen, J., & Zilberman, D. (2022). Agri-food value chain revolutions in low- and middle-income countries. Journal of Economic Literature, 60(4), 1316-1377.

Radjou, N., Prabhu, J., & Ahuja, S. (2012). Jugaad innovation: Think frugal, be flexible, generate breakthrough growth. John Wiley & Sons.

Hybrid Optimization of Coccidiosis Chicken Disease Prediction, Detection and Prevention Using Deep Learning Frameworks

Wairimu David Waititu1*, John Kagira1, Joseph Morake2

1Jomo Kenyatta University of Agriculture and Technology,2Botswana International University of Science and Technology

*Corresponding author email: waititudavid570@gmail.com

KEYWORDS: coccidiosis, computer vision, deep learning, edge computing, Convolutions Neural Networks

Abstract

Poultry farming is one of the thriving businesses in Kenya, therefore, playing a crucial role in the economy and the food value chain. Most of the farmers are small-scale scale while a good number practice large-scale farming. Egg- laying birds are the most preferred as a result of the high profits gained from egg sales. However, various stresses including disease outbreaks have greatly caused loss due to late detection and lack of systems to predict the diseases before the infections. Coccidiosis has been one of the most prevalent and highly contagious poultry diseases. As such, there is a need to address the challenges, by employing emerging technologies in a built environment. In this study, deep learning models were deployed in the TensorFlow framework to detect the onset of the disease. Due to the fast spread of the disease, an automated vaccination system was used to protect healthy birds from attaining the disease and to adopt a robust prevention system, a disease prediction framework was deployed to alert farmers to adapt to better mechanisms. Different deep learning models were deployed and tested and their accuracies were compared to get a fully efficient model. A Convolutional Neural Network (CNN) ResNet50 model showed the highest accuracy of 96% through the transfer learning technique. The deployed automatic vaccination system revealed high efficiency in releasing the right dose amounts after the disease is detected. Therefore, the integration of engineering technologies to foster automated systems will not only ensure food security in poultry farming but also open up more industries to build these systems for the long-term benefit of farmers.

Introduction

Coccidiosis poultry disease is one of the most important diseases caused by protozoans of the genus Eimeria which is mainly transmitted via the fecal-oral route. Due to its fast spread, poultry production is affected in a great way leading to losses and a threat to food security. In the event of an outbreak, severe losses are incurred attributed to the high cost of treatment, isolation of affected birds, and low-quality produce from the birds (Wang et al., 2019). This, overall, negatively affects the economy, export markets, loss of labor when the birds die massively, and spiking in eggs and chicken meat prices resulting from reduced production. As is the case with all other diseases, early diagnosis of coccidiosis disease would greatly help to offset the cost burden to farmers and also minimize the occurrence of zoonotic diseases to the consumers. However, the available conventional techniques of diagnosis which are mainly sound distinctions and visual observations, are time-consuming, labor- intensive, and inaccurate diagnoses by the farmers (Sadeghi et al., 2015). Additionally, veterinarians may not be flexible enough to cater to the increasing poultry farming activities. Therefore, these conventional methods would be ineffective and unreliable for early detection of coccidiosis.

This study presents a deep learning machine vision-based early detection, prediction and auto- vaccination technique to aid poultry farmers tackle the disease. Deep learning is a subset of machine learning inspired by the anatomy of the human brain and has recently gained application in agriculture. Its algorithms use multi-layered neural networks which are complex and whose abstraction levels improve step by step through non-linear transforms of the data inputted (Mbelwa et al., 2021). The detection was aided by Convolutional Neural Networks (CNNs) which are the most effective and versatile deep learning architectures. They work by the principle of mimicking the human brain thus being able to interpret things the same as humans would, and even better. They have gained much attention and popularity because of their multi-later processing ability, less computation power needs, and ability to permit extracted features to be optimized. Therefore, they allow the computer system to capture only the required data, classify and recognize automatically, and hence trigger an action to be done as per the data captured.

Training of a deep learning model requires a large amount of data and this may increase the cost of the system. A transfer learning approach was adopted which refers to the application of a known CNN model used in another classification task and applying it to be used in the new task (Machuve et al., 2022). Some of the pre-trained models include ResNet50, XceptionNet, VGG19, and InceptionNet. This allows training of the model with little data and thus reduces computation power and development time. Transfer learning in a great way, helps to reduce overfitting or underfitting of the model.

Embedded systems are computer hardware systems embedded with software to perform specific tasks with the aid of a microcontroller (Saddik et al., 2022). They help to integrate and deploy the trained Deep Learning model to be used for detection and autovaccination. These systems are mainly managed by digital signal processors (DSP), microcontrollers, field programmable gate arrays, and gate arrays. These systems are incorporated together with components dedicated to handling mechanical and electrical interphase.

Several studies have been carried out to explore the use of image datasets for either recognition, classification, or segmentation. Mbelwa et al. (2021) demonstrated a poultry disease detection model using Transfer Learning. The model gained an accuracy of 94%, outperforming Resnet, VGG, and MobileNet CNN architectures. However, this research did not show ways to tackle the disease after detection unless under human intervention. In another study, Support Vector Machine was used to demonstrate an early diagnosis and warning algorithm for broilers who were sick (Zhuang et al., 2018). The technique employed was to extract the posture of healthy and sick birds following the establishment of eigenvectors. Predictions were made after analysis of the bird’s posture. The average accuracy obtained from this study was 91.5% but on the incorporation of all features, an accuracy of 99.5% could be achieved. However, the study focused primarily on poster-based techniques which may give false positive error for instance when the birds are just resting. In a similar study, was carried out research to demonstrate chicken disease detection using a CNN model developed to detect and diagnose three types of chicken diseases using fecal images (Machuve et al., 2022). Different models were trained which include VGG16, Xception, InceptionV3, and MobilenetV2. The model was deployed in smartphones as a diagnostic tool with MobilenetV2 attaining the highest accuracy. However, this research was only limited to classification alone.

The objective of this study, therefore, is to develop a novel system to predict, detect and automatically give vaccinations to the birds to protect them from contracting the disease with minimal or no human intervention. This will be enabled by fecal images of birds suffering from coccidiosis.

Materials and Methods

Dataset

For purposes of this study, fecal images of coccidiosis were obtained from Kaggle Datasets (https://www.kaggle.com) and processed for purposes of training the model. Kaggle is a public online community that offers open-source datasets for deep learning. The dataset contained 2476 colored images of coccidiosis each of size 224 by 224 pixels and 2404 images of healthy fecal matter with similar properties as those of coccidiosis, which were all labeled and annotated for training. This dataset from Kaggle was obtained from farmers of poultry from both indigenous and cross-breeds of chicken. In its establishment, fecal samples were collected and analyzed in the laboratory to ascertain the accuracy of the data. Animal health professionals were also involved in verifying the dataset before it was released for use to the public. A sample is expressed by Figure 1.

To develop a fully versatile and accurate model, different architectures designed for image classification were used, both in training and testing. These were Resnet50 (He et al., 2016), VGG19 (Simonyan and Zisserman, 2014), Xception (Chollet, 2016), and MobilenetV2 (Sandler et al., 2018). VGG19 is an image classification model which is an improvement of its predecessor VGG16. This convolutional Neural Network (CNN) contains up to 19 layers and it is trained on the ImageNet which contains 1000 classes. Due to diminishing gradient which results in the decrease of the model’s depth, VGG19 is difficult to train. However, it has the best feature extraction characteristics. To address the issue of diminishing gradient, the Resnet 50 model was proposed. This allows CNN to go much deeper. It also consumes less training memory although it has many layers compared to VGG19. MobilenetV2 is also trained on ImageNet which is considered a lightweight model due to its high performance and efficiency in deployment on mobile devices. It is mainly made up of superficially separable neural network layers which are made from depthwise differentiable convolution filters. One convolution filter with one-to-one convolutions serves as the mechanism for each input network. In a study conducted, Resnet 50, MobileNet, and VGG 19 pre-trained models were compared for the identification of pneumonia disease (Kavya et al., 2022). The accuracy of the architectures was 87%, 92%, and 90%, respectively. This explains that not all architecture performs well in every problem. Mbelwa et al. (2021) proposed the leverage on the use of the transfer learning approach and finetuning the model weights instead of training the model from scratch since this not only improves the accuracy of the model but also reduces the amount of time used for training the model.

Figure 1: Sample images from Kaggle dataset (https://www.kaggle.com/datasets/allandclive/chicken-disease-1).

Model’s Architecture

The transfer learning approach was used in the training of the models. Images of size 224 by 224 pixels were fed into the convolution layers as inputs for ResNet50, MobilenetV2, and VGG19. An image size of 299 by 299 pixels was used for the Xception model architecture. The batch size of the input images was set to 32. After the convolutional layer, which operates over a 2 by 2-pixel window, is the max-pooling layer, which has filters with narrow receptive fields of 3 by 3. A combination of these layers takes place leading to the formation of a single block which is applied iteratively while raising the network’s filler depth to integer values such as 32, 64, 64, 128, 128, 256, 256, and 512 which makes it a full block convolution. Similar padding is applied throughout the different model training to maintain the width and height shape of the output. Adam optimizer was used for VGG19, MobilenetV2, and Resnet50 with a learning rate of 0.001 to minimize the error. The Sparse Categorical Crossentropy probabilistic class was used to compute the loss between the predictions and the labels and SoftMax activation function in the output layer.

Autovaccination

Embedded systems offer a novel way of deploying deep learning models in real life and using them to make inferences. In this setup, Raspberry Pi 4 model B, 4Gb RAM was used to deploy the TensorFlow Lite quantized model. The micro-submersible pump was connected to a 5V relay module to offer protection to the Raspberry Pi board. The VCC port for the relay module was connected to General Input and Output (GPIO) pin 17 which was configured to give a signal whenever coccidiosis was detected. To get the image data, a USB webcam was connected to the Pi USB 3.0 port.

    CNN Model

    The dataset obtained from Kaggle with four classes was used to develop the Convolutional Neural Network to detect and classify coccidiosis disease from infected fecal images. From the different architectures that were used, ResNet50 produced the top performance of 96.32% after finetuning the model. Before finetuning was done, the model produced an overall test accuracy of 94% using the transfer learning approach. To increase this accuracy, finetuning was done for an additional 10 epochs and a learning rate of 0.0001 to minimize the overfitting of the model. The layers of the base model were also unfrozen by enabling base_model.trainable= true. Accuracy scores for the other models were MobilenetV2 model 93.95%, VGG19 model 89.15%, and Xception model 96.14%. However, it is worth noting that the Xception model performed closely to ResNet50. Table 1 shows the performance of the models after finetuning. The parameters for the best-performing model were as shown in Table 2. The training and loss curves on Resnet50 model are shown in Figure 2 and 3 respectively.

    Table 1:Model’s Performance

    ModelTest Accuracy (%)Validation Accuracy (%)
    VGG1989.1590.63
    Resnet5096.3295.96
    MobileNetV293.9593.75
    Xception96.1496.75

    Table 2: ResNet50 training parameters

    ParameterValue
    Learning Rate0.0001
    Number of Epochs10
    Hidden Layers256
    Drop-out0.2

    Autovaccination

    The main goal of this research is to develop an autovaccination dispensation system, the best-performing model was saved and converted to a TensorFlow lite (tf lite) version which was deployed on a Raspberry Pi. Tf lite is a lighter version of the TensorFlow model that is ideal for resource-constrained devices such as Raspberry Pi. To obtain the tf lite file, tensorflow.lite.TFLiteConverter instance was used to convert the saved model. The resulting tf lite file was optimized for size and yielded 21 MB of data. The file was also normalized and the class labels were added to the metadata. Deployment done and tested on the Python OpenCV library showed greater performance in the detection of coccidiosis in real time. Confidence scores obtained were 97% in a bright-lit environment and 67% in a partially dark room. The embedded system was also able to identify multiple instances of the disease from the same image input. With the instances of the disease detected, the micro-submersible pump through the relay module dispensed the vaccine portions which were calibrated. Calibration was done to match the time taken for the pump to operate once the disease instance was detected. This approach was best in that the flow rate is known per unit time and the speed of the pump motor was constant. This alludes that the required dosage of the vaccine is equivalent to the dispensing time. Farmers, however, have to prepare the vaccine and store it carefully in a tank where the pump can be installed. Reapplication time of the vaccine was set to one week after which any recorded instance of the disease would trigger another application of the disease.

    Figure 2: Training curve

      Figure 3: Loss curve

    Conclusion and Future Scope

    As a result of using deep learning technology to identify coccidiosis disease in chicken houses, vaccination systems can be used to counter the onset of the disease and hence minimize losses incurred by the mass death of chickens. Instances of the disease occurrence can also be recorded and reported to the farmer to enable them to make informed measures and decisions on their rearing practice. Leveraging artificial intelligence comes in handy in that farmers can boost and expand their production with minimal reduction in losses anticipated as a result of a disease outbreak. In combination with other systems, this paper proposes a framework where the integration of other systems can be combined to build intelligent chicken farming systems while leveraging deep learning and computer vision. These include; live streaming of the detections made, integration with feeding systems, and recommendations to farmers based on data collected.

    References

    François Chollet. Xception: Deep learning with depthwise separable convolutions. CoRR, abs/1610.02357, 2016. URL http://arxiv.org/abs/1610.02357

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.

    Nallamothu Sri Kavya, N Veeranjaneyulu, D Divya Priya, et al. Detecting covid19 and pneumonia from chest x-ray images using deep convolutional neural networks. Materials Today: Proceedings, 64:737–743, 2022.

    Dina Machuve, Ezinne Nwankwo, Neema Mduma, and Jimmy Mbelwa. Poultry diseases diagnostics models using deep learning. Frontiers in Artificial Intelligence, 5:733345, 2022.

    Hope Mbelwa, Dina Machuve, and Jimmy Mbelwa. Deep convolutional neural network for chicken diseases detection. 2021.

    Amine Saddik, Rachid Latif, Abdelhafid El Ouardi, Mohamed Elhoseny, and Adel Khelifi. Computer development based embedded systems in precision agriculture: Tools and application. Acta Agriculturae Scandinavica, Section B—Soil & Plant Science, 72(1):589–611, 2022.

    Mohammad Sadeghi, Ahmad Banakar, Meghdad Khazaee, and MR Soleimani. An intelligent procedure for the detection and classification of chickens infected by clostridium perfringens based on their vocalization. Brazilian Journal of Poultry Science, 17:537–544, 2015.

    Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. Mo- bilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4510–4520, 2018.

    Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.

    Jintao Wang, Mingxia Shen, Longshen Liu, Yi Xu, and Cedric Okinda. Recognition and classification of broiler droppings based on deep convolutional neural network. Journal of Sensors, 2019:1–10, 2019.

    Xiaolin Zhuang, Minna Bi, Jilei Guo, Siyu Wu, and Tiemin Zhang. Development of an early warning algorithm to detect sick broilers. Computers and Electronics in Agriculture, 144:102–113, 2018. ISSN 0168-1699. doi: https://doi.org/10.1016/j.compag.2017.11.032

    Scaling to Achieve Impact from Technology and Improvement

    Ranjani Krishnan1*, Satish Joshi1, Ajit Srivastava1

    1Michigan State University, USA

    *Corresponding author email: krishnan@bus.msu.edu

    KEYWORDS: agricultural technology, scaling, technology readiness, innovation, diffusion, gap analysis

    Introduction

    Scaling is critical to achieving impact from any technology and innovation. A variety of factors influence successful scaling of agricultural technology, including the nature of the innovation and its ability to meld with existing agricultural practices. Importantly, scaling requires behavioral, financial, social, political, and institutional commitment. Scaling is not merely the automatic diffusion of proven technology, but rather it is a component of a broader dynamic social process, which can disrupt prevailing social norms and power structures. Consequently, many promising innovations do not progress beyond research and prototype testing/ demonstration. Numerous problems arise in the process of linking locally distributed and highly fragmented individual or small-scale operations into a coordinated system along the entire value chain of farmers, artisans, manufacturers, consumers, and finance providers, which is essential for effective scaling.

    The key barriers to successful scaling include gaps in technology, human capital, institutional, and local environmental facilitators. The first step in achieving scaling goals is to understand the extent and the nature of gaps to scaling. With this goal in mind, two new instruments were developed: (a) Technology Readiness Assessment (TRA) which seeks to understand perceptions of the technology and provide a quantitative assessment of the factors that may undermine scaling, and (b) Scaling Gap Analysis (GAP), a qualitative assessment of factors that can facilitate or undermine efforts to scale the technology to a larger group of institutions and beneficiaries across the entire ecosystem. Primary data collected through these instruments, can shed light on the local institutional context, incentives, and programs that influence scaling.

    Design of the Tra Instrument

    We designed a TRA instrument to elicit and collect systematic data about the demographics of the user group for the specific technology (choppers or planters), their current level of knowledge about the technology and its techno-economic appeal. The TRA, which is a 15 question, five-item instrument, also allows an assessment of potential demand for the technologies. The TRA questions covered demographics; farming system descriptions; extent of engagement with the technology by various family members; relative effectiveness of the technology in terms of time, quality, reliability, and effort exertion; ease of use and training needs; potential for local manufacture, repair and service; affordability, cost effectiveness, sources of financing, and openness to renting.

    Design of the Gap Instrument

    We designed a qualitative tool to assess three types of gaps in scaling. namely technology gaps, human capital gaps, and institutional gaps. To construct the Gap analysis instrument, we first conducted a review of the existing literature and the available measures related to diffusion of agricultural technology. We adapted some measures from the Agricultural Scalability Assessment Toolkit (ASAT) (Kohl & Foy, 2018), and the PPP Lab Scaling Scan (Jacobs & Woltering, 2019) and developed new measures, based on the team’s assessment of the capabilities required for scaling at our sites. Our final Gap analysis instrument contains items that characterize specific aspects of scaling capability. These items are aimed at generating in-depth dialogues to elicit the respondents’ perceptions about factors that can facilitate or hinder scaling.

    Examples of technology gap assessment questions included:

    Examples of human capital gap assessment questions included:

    Example Institutional gap assessment questions included:

    The initial drafts of the survey instruments were translated into local languages and pilot tested with representative respondents. The feedback received during pilot testing was used to revise and refine the drafts leading to the final instruments. An illustrative extract of the instrument is shown in Figure 1, and the complete instrument is available on request from the corresponding author.

    QuestionScaling Up Is EasyScaling Up Is NeutralScaling Up Is Harder
    1. Is the technology relevant to farmers?
    Is farmers group well-defined in terms of potential users?
    Ask the respondents to name a few users by name, ask if they can use the technology    		Farmer group members can name 3-4 users, they themselves can be usersWith prompting, respondents can name one potential userRespondents cannot name potential users
    Does the planter directly affect the farmer group in terms of crop yield (loss prevented) or change productivity?
    Ask the respondents if the technology can increase yield or prevent losses. Ask how much of each     		Respondents can name the quantity of increase in productivity and/or reduced lossesRespondents are unsure but believe there will be productivity gains and/or loss preventionRespondents don’t believe that there will be productivity gains or reduction in losses. They don’t believe they can get any benefit
    Is the benefit of using the planter clearly visible?
    Ask if planter alone can increase productivity by itself    		Respondents believe that there will be an increase in yield from just using the planterRespondents believe that benefits are there but depend on other factors in the farming systems such as crop variety or seed quality Respondents do not feel that users can observe an increase in yield from using the planter

    Figure 1: An illustrative extract from the GAP analysis instrument.

    Implementation of Tra and Gap Analysis

    The TRA and GAP analyses were implemented in Burkina Faso, specifically for an improved planter that was developed in collaboration with local farmers, and was explicitly redesigned for planting in minimally tilled soil. Fifteen graduate students of the Nazi Boni University were selected to implement the TRA and GAP analyses. These staff were trained first in the administration and recording of the TRA and GAP instruments in online training sessions. The target respondents were farmers and value chain participants e.g. artisans, and service providers. Focus group discussions were conducted in 2022 with a total of 45 respondents in six municipal locations in Burkina Faso. Each respondent group included 5-6 participants and had adequate representation of women and other marginalized groups. Each of the group discussions were led by two facilitators who ensured that views from all participants were sought, and discussions generated some degree of agreement on each question. Each group discussion lasted around 3 hours. The insights gathered from the TRA and GAP analysis were synthesized by the project team and used in its scaling efforts.

    Project Outcomes and Concluding Remarks

    Thirty-five planters are currently in use in the project area and 40 more planters are being manufactured locally. Anecdotal evidence suggests that a family of three with a planter can now carry out the work of a hand planting crew of 15-20. Reported yield gains ranged from 50-150%. Women reported time and effort savings and increased engagement in other activities. Similarly, local artisans reported benefits of increased economic activity.

    Acknowledgments

    Funding to support Satish Joshi’s work was provided by USDA Hatch Grant MICL12075

    References

    Kohl, R., & Foy, C. (2018). Guide to the Agricultural Scalability Assessment Tool for Assessing and Improving the Scaling Potential of Agricultural Technologies. Management Systems International, a Tetra Tech Company, for the E3 Analytics and Evaluation Project. United States Agency for International Development. https://pdf.usaid.gov/pdf_docs/PA00T6KX.pdf

    Jacobs, F., & Woltering, L. (2019). The Scaling Scan: A Practical Tool to Determine the Potential to Scale.https://repository.cimmyt.org/entities/publication/6a22f251-16f4-4334-afec-6e587b6a7d98

    Exploring the Value Proposition of IF4MAAS: Pioneering Sustainable Development Finance and Impact Investment in Agriculture

    Girma B. Abel1*, Ajit Srivastava2, Ranjani Krishnan2

    1XSyn Corporation &IF4MAAS

    2Michigan State University, USA

    *Corresponding author: abel.b.girma@gmail.com

    KEYWORDS. modernizing African agriculture, business development, blended financing, sustainable development, agrifood systems, impact investment

    Abstract

    This article examines the Impact Fund for Modernizing African Agri-Food Systems (IF4MAAS), which proposes a novel development finance and impact investment model. IF4MAAS provides an efficient value proposition for blended financing, bridging financial gaps, promoting gender-inclusive growth, and empowering youth in the agricultural sector. IF4MAAS aims to leverage support from foundations, multinational donors, and corporate entities for impact investment that are aligned with the United Nations Sustainable Development Goals (SDGs). Acting as an efficient intermediary for blended financing, IF4MAAS connects project owners with development finance institutions, impact investors, donors, and grant providers. This collaboration facilitates value chain integration and scalable agricultural infrastructure investments, aligning with the mission of the Alliance for Modernizing African Agrifood Systems (AMAA) to transform Africa's agrifood landscape.

    Introduction

    African agrifood system is facing many challenges including population growth, rapid urbanization, diet transformation and climate change. Agricultural development is vital for economic growth, food security, and poverty reduction in African countries. Promoting an ecosystem that supports agribusiness development is crucial to modernizing African agrifood systems (Srivastava, 2020). The efficient blending of development finance, private capital, and impact investment has become increasingly crucial in addressing global challenges, particularly in emerging economies in Africa (Thompson, 2020). These financial models aim to generate both social impact and financial returns, contributing to sustainable development. Investments in this sector can lead to significant social and economic benefits. IF4MAAS is a non-profit public benefit corporation dedicated to transforming the agricultural landscape in Africa by actively co-creating impactful projects financed through blended financing, leading to full private capital uptake. Its vision is to create a sustainable and prosperous future for Africa's farmers, businesses, and communities by modernizing agrifood systems through innovative and impactful initiatives. IF4MAAS is committed to taking a value-chain approach to enhancing productivity, reducing postharvest losses, promoting sustainability, and fostering inclusive growth through strategic investments and partnerships in the African agrifood sector.

    Supported Project Initial Findings

    Evidence suggests that donor-supported projects often fail to provide lasting change because intermediaries impose unrealistic and uninformed Key Performance Indicators (KPIs). The pursuit of such KPIs leads to misaligned priorities and a failure to address real challenges. In some cases, donor-funded projects have caused market distortions, posing challenges for private sector-led agricultural service providers. To ensure lasting impact, it is crucial to co-create projects with realistic expectations, focusing on sustainable growth and the creation of decent job opportunities, with less emphasis on measuring dollars spent per job created.

    IF4MAAS Model: an Overview

    IF4MAAS unites investors, innovators, and stakeholders to drive impactful change in Africa's agrifood sector, focusing on food security, sustainable growth, and environmental resilience. Acting as a bridge, it fosters collaboration and leverages each contributor's strengths and resources to achieve UNSDGs. By using blended financing, IF4MAAS reduces risk and attracts investment to projects with high impacts, and structures the financing to promote sustainability and effective risk management. It enhances productivity and value chains by implementing modern agricultural practices and technologies, supporting value- chain infrastructure, and enabling investment in the entire agrifood value chain to increase yield and minimize food loss. IF4MAAS empowers women and youth by providing access to finance, training, and markets, promoting inclusive growth and gender equality. It also advocates for renewable energy and sustainable farming practices to protect the environment and make high-quality carbon credits available. Additionally, IF4MAAS offers digital platforms for market linkage, product traceability, technology transfer, and aggregating demand across countries to create scale for local production. Importantly, IF4MAAS deliberately designs and structures projects with relevance across multiple countries to create the scale needed to advance local production of critical components.

    Alignment With AMAA

    The collaborative approach between IF4MAAS and AMAA is driven by unified goals to enhance productivity, promote sustainability, and achieve the UNSDGs. Both entities share a vision of creating a modern, productive, and profitable agrifood system in sub-Saharan Africa, ensuring food security, economic growth, and improved quality of life sustainably and equitably. Their mission is to modernize African agrifood value chains from production to consumption through advanced technologies and methodologies. By leveraging their respective strengths, AMAA’s platform for stakeholders complements IF4MAAS’s innovative financing model, effectively attracting partners and investors to drive impactful change in Africa's agrifood sector

    IF4MAAS and AMAA share complementary objectives to enhance technology and innovation as well as entrepreneurship and business development in Africa's agrifood sector. IF4MAAS conducts technology readiness assessments, assesses demand for local production, and promotes renewable energy, aligning with AMAA’s goals. Additionally, IF4MAAS leverages blended financing, empowers women and youth through finance and training, offers capacity-building programs, and uses digital platforms for market access, supporting AMAA’s mission of fostering inclusive growth, workforce development, and improved market linkage through technology.

    Conclusion

    IF4MAAS will strategically align with the Alliance for Modernizing African Agrifood Systems (AMAA). The goal of AMAA is to develop a framework that unites entrepreneurs, innovators, investors, financial institutions, academic institutions, policymakers, and farmers to establish an ecosystem aimed at modernizing African agrifood systems (Gitau et al. 2021). IF4MAAS, with its focus on blended financing and inclusive growth, serves as a critical intermediary that addresses the financial gaps hindering agricultural development. By uniting diverse stakeholders, conducting thorough due diligence, and leveraging innovative technologies, IF4MAAS enhances the efficiency and sustainability of investments in the agrifood sector. It is crucial for IF4MAAS to effectively present itself to stakeholders as a trustworthy and highly efficient intermediary for sustainable development. Building trust involves demonstrating transparency, accountability, and a strong track record of successful interventions. As IF4MAAS continues to develop, understanding the best practices for stakeholder engagement and communication will be key to attracting the necessary support and investment.

    Together, IF4MAAS and AMAA can accelerate progress towards a sustainable, productive, and inclusive agrifood sector in Africa (Srivastava et al. 2024). This collaboration ensures food security, economic growth, and an improved quality of life for all stakeholders, ultimately contributing to the achievement of UNSDGs across the continent.

    References

    Gitau, M., Hiablie, S., Ileleji, K. & Srivastava A. 2021. The Alliance for Modernizing African Agrifood Systems. Resource, Nov/Dec 2021, ASABE, St. Joseph, MI. https://bt.e-ditionsbyfry.com/publication/?i=727239&p=4&view=issueViewer

    Srivastava, A. 2020. The Power for Small-Scale Mechanization. Resource, March/April 2020. ASABE, St. Joseph, MI. https://bt.e-ditionsbyfry.com/publication/?i=651804&p=4&view=issueViewer

    Srivastava, A., Gitau, M., Hiablie, S. & Ileleji, K. 2024. Imagining the Future at the 2023 AMAA Conference. Resource, May/June 2024, ASABE, St. Joseph, MI. https://bt.e-ditionsbyfry.com/publication/?i=821591&p=16&view=issueViewer

    Thompson, M. 2020, Unlocking Access to Alternative and Innovative Finance. Sahel Quarterly, October 2020, Volume 25.

    PILLAR III: CAPACITY BUILDING AND WORKFORCE DEVELOPMENT

    Future Workforce Development through Hands-on Remote Training

    Judith Nkechinyere Njoku1, Senorpe Hiablie2, and Daniel Dooyum Uyeh1*

    1Department of Biosystems and Agricultural Engineering, Michigan State University, East Lansing, MI, United States

    2 Shell International Exploration and Production Inc., Houston, TX, United States

    *Corresponding author: uyehdani@msu.edu

    KEYWORDS: e-learning, internship, workforce development, smart agriculture, AI and machine learning

    Introduction

    The rapid advancement of technology has transformed various sectors, including workforce development. The growth of remote training solutions has facilitated a paradigm shift in professional education, enabling the development of a skilled workforce irrespective of geographical constraints (Ashmarina et al., 2022). Remote training offers unmatched flexibility, allowing learners to engage in educational activities from any location. This accessibility is particularly beneficial in fields like AI and ML, where continuous learning and up-to-date knowledge are essential. This paper explores the potential of remote training, particularly focusing on an innovative artificial intelligence and machine learning (AI/ML) internship program established in 2020, aimed at nurturing future data scientists and machine learning experts.

    The Impact of COVID-19 On Remote Work

    The COVID-19 pandemic significantly accelerated the global shift towards remote work, which was essential for the survival of many industries during the crisis. This transformation introduced two primary work methods: remote and hybrid. The hybrid approach allowed employees to work on-site or remotely, depending on the circumstances.

    Figure 1 Estimated remote workforce by industry before and during the lockdown (Great Place to Work, 2020)

    The pandemic led to substantial increases in the percentage of the remote workforce across various industries as illustrated in Figure 1. In the professional services sector, there was a 77% increase in remote work, while the information technology industry saw a 64% rise. The healthcare sector experienced the least significant increase at 12%, primarily due to the necessity of on-site presence for many healthcare roles (McKinsey & Company, 2021). A survey conducted by Gartner in 2020 showed that 88% of organizations mandated their employees to work from home as the virus spread globally (Gartner, 2020). The effect of the pandemic was also reinforced in a study conducted by Pew Research Center, which showed that before the omicron variant began to spread in the U.S., 83% of the workers were already working from home all or most of the time (Pew Research Center, 2020). Additionally, a study by the U.S. Bureau of Labor Statistics (2021) found that 80% of employees at a particular company continued to work remotely, even after COVID-19, highlighting the widespread adoption of remote work practices. Many companies today operate fully remotely, raising concerns about learners' preparedness for remote work environments. These shifts underline the importance of remote training programs to equip future professionals with the necessary skills to thrive in such settings.

    Challenges and Benefits of Remote Education

    The United Nations Educational, Scientific and Cultural Organization (UNESCO) reported that the pandemic caused a global increase in school closures, affecting approximately 1.38 billion learners, including those at the tertiary education level (UNESCO, 2020a, b). Despite the successful implementation of remote working arrangements in the workforce, educators did not adopt similar approaches effectively, leaving many students idle and stagnating in their learning. Remote study platforms such as Coursera, Udemy, and EdX saw a significant increase in attendance during the pandemic, highlighting the need for more remote learning opportunities (World Economic Forum, 2020). Preparing learners worldwide for remote work scenarios is crucial to ensure their productivity and engagement despite remote working arrangements. Some of the benefits of remote education include the following:

    The AI/ML Internship Program

        The AI/ML internship program, initiated in 2020, exemplifies the potential of remote practical technical training in shaping the future workforce. This program is designed to provide a rigorous and immersive learning experience, combining theoretical knowledge with practical applications in smart agriculture. This section discusses the selection process, program structure, and case studies. Figure 2 illustrates the internship selection process and workflow.

    Selection Process

    The program employs a thorough screening process to ensure the selection of highly qualified candidates. This includes:

    Figure 2: The AI/ML Internship selection process and workflow.

    Program Structure

    Once selected, interns collaborate for six months on various projects to solve real-world problems using advanced machine learning techniques. The structure of the program includes:

    Impact Measures on Interns’ Professional Development and Project Case Studies

    The AI/ML internship has spanned three cohorts, as represented in Figure 3. Over 95% of the interns remained committed to the internship until the end, which highlights the program's effectiveness and the interns' engagement during the internship. The effectiveness of the AI/ML internship program is also demonstrated through successful case studies, which show its tangible impact on interns' professional development. These case studies highlight how interns have applied their skills to real-world challenges in smart agriculture, leading to innovative solutions and advancements in the field.

    Impact Measures on Interns’ Professional Development

    The impact of the AI/ML internship program is evident across various metrics, reflecting its success in enhancing the professional trajectories of the interns over the three cohorts.

    A significant achievement for the nineteen interns who completed the program is their involvement in academic publications. Approximately 85% of these interns are co-authors of completed or published papers. This high rate of co-authorship highlights interns' active role in contributing to scholarly research, showcasing their ability to engage deeply with complex projects and produce publishable work.

    The program has also proven effective in preparing interns for their future careers. Approximately 53% of the interns have advanced to graduate school or secured technical roles post-internship. Specifically, four interns from the 2022 cohort have advanced to graduate school, while one has joined the industry. For the 2023 cohort, two interns have advanced to graduate school and secured technical roles post-internship. The 2024 cohort has not yet concluded and has already seen two interns accepted into graduate school. Most of the remaining interns are still wrapping up their undergraduate studies. This statistic reflects the program's strength in equipping participants with the necessary skills and experience to succeed in advanced academic pursuits and professional environments.

    Overall, these impact measures demonstrate the effectiveness of the AI/ML internship program in fostering professional growth and preparing interns for future academic and technical careers.

    Figure 3: The statistics of Internship participation before and after the program.

    Project case studies

    The following case studies include projects undertaken in the 2022 and 2023 cohorts of the internship program.

    Case Study 1: Optimal Sensor Placement in Controlled Greenhouses

    The first cohort of interns focused on optimal sensor placement in controlled greenhouses. This project employed reinforcement learning (Uyeh et al., 2021), supervised learning (Uyeh et al., 2022a), unsupervised learning (Uyeh et al., 2022b), and evolutionary algorithm (Ajani et al., 2023) techniques. The result was a publication of papers that showed the innovative use of these machine learning methods to enhance agricultural efficiency and productivity.

    Case Study 2: Multimodal Livestock Feed Quality Prediction

    The second cohort undertook multiple projects, including the development of a multimodal livestock feed quality prediction system using AI. This project demonstrated the interns' ability to integrate various data sources and machine learning techniques to predict and improve feed quality, which is important for livestock management.

    Case Study 3: Soil Quality Prediction Using AI

    Another project from the second cohort focused on soil quality prediction using AI. This project highlighted the application of machine learning algorithms to analyze soil data, providing valuable insights for improving soil management practices and agricultural productivity.

    Case Study 4: An Ensemble Approach for Optimal Sensor Placement

    Building on the work of the previous cohort, the second cohort also developed an ensemble approach for optimal sensor placement in controlled environments. This approach combined different machine learning techniques to improve the accuracy and efficiency of sensor placement strategies, further advancing the field of smart agriculture. The results of all these works have been documented and submitted for publication in various reputable journals.

    Concluding Remarks

    The AI/ML internship program represented a significant step forward in remote workforce development. By combining rigorous training with practical, real-world applications, the program prepares interns to become leaders in the dynamic and ever-evolving fields of AI and ML. As the global workforce continues to adapt to new technological advancements, remote training programs like this will play a crucial role in developing the next generation of skilled professionals. The success of this remote training methodology highlights its potential to revolutionize workforce development across various industries. By fostering a culture of continuous learning and innovation, the AI/ML internship program not only contributes to the advancement of the AI and ML domains but also sets a benchmark for excellence in professional education and training.

    References

    Ajani, O. S., Aboyeji, E., Mallipeddi, R., Uyeh, D. D., Ha, Y., & Park, T. (2023). A genetic programming-based optimal sensor placement for greenhouse monitoring and control. Frontiers in Plant Science, 14. https://doi.org/10.3389/fpls.2023.1152036

    Ashmarina, S. I., Mantulenko, V. V., & Gryaznov, S. A. (2022). Reskilling and upskilling the future-ready workforce for Industry 4.0 and beyond. Information Systems Frontiers, 24, 1517–1533. https://doi.org/10.1007/s10796-022-10308-y

    Gartner (2020). Gartner HR Survey Reveals 88% of Organizations Have Encouraged or Required Employees to Work From Home Due to Coronavirus. Retrived from: Encourage Employees to Work From Home Due to COVID | Gartner

    McKinsey & Company. (2021). The future of remote work: An analysis of 2,000 tasks, 800 jobs, and 9 countries. Retrieved from: https://www.mckinsey.com/featured-insights/future-of-work/whats-next-for-remote-work-an-analysis-of-2000-tasks-800-jobs-and-nine-countries

    Pew Research Center. (2020). COVID-19 Pandemic Continues To Reshape Work in America. Retrieved from: https://www.pewresearch.org/social-trends/2022/02/16/covid-19-pandemic-continues-to-reshape-work-in-america/

    U.S. Bureau of Labor Statistics. (2021). Telework during the COVID-19 pandemic: estimates using the 2021 Business Response Survey: https://www.bls.gov/opub/mlr/2022/article/telework-during-the-covid-19-pandemic.htm

    UNESCO. (2020). Education: From COVID-19 school closures to recovery. Retrieved from https://www.unesco.org/en/covid-19/education-response.

    UNESCO. (2020). UNESCO responds to the global crisis in education due to COVID-19. Retrieved from https://www.unesco.org/en/articles/unesco-responds-global-crisis-education-due-covid-19

    Uyeh, D. D., Akinsoji, A., Asem-Hiablie, S., Bassey, B. I., Osinuga, A., Mallipeddi, R., Amaizu, M., Ha, Y., & Park, T. (2022b). An online machine learning-based sensors clustering system for efficient and cost-effective environmental monitoring in controlled environment agriculture. Computers and Electronics in Agriculture, 199, 107139. https://doi.org/10.1016/j.compag.2022.107139

    Uyeh, D. D., Akinsoji, A., Mallipeddi, R., Asem-Hiablie, S., Amaizu, M., Ha, Y., & Park, T. (2021). A reinforcement learning approach for optimal placement of sensors in protected cultivation systems. IEEE Access, 9, 100781-100800. https://doi.org/10.1109/ACCESS.2021.3096828

    Uyeh, D. D., Olayinka, I., Mallipeddi, R., Asem-Hiablie, S., Amaizu, M., Ha, Y., & Park, T. (2022a). Grid search for lowest root mean squared error in predicting optimal sensor location in protected cultivation systems. Frontiers in Plant Science, 13. https://doi.org/10.3389/fpls.2022.920284

    World Economic Forum. (2020). The rise of online learning during the COVID-19 pandemic. Retrieved from https://www.weforum.org/agenda/2020/04/coronavirus-education-global-covid19-online-digital-learning/

    Agricultural Tools: From School to the Market

    Eric F. Zama1,2*, Richard A. Cooke1, Martin N. Ngwabie2

    1Department of Agricultural and Biological Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA

    2Department of Agricultural and Environmental Engineering, College of Technology, University of Bamenda, Bambili, Cameroon

    *Corresponding author: ezama@illinois.edu

    KEYWORDS: agricultural tools, universities, engineering schools, student projects, markets and industry

    Introduction

    Engineering schools and colleges across the globe play a crucial role in innovations that shape our environment (Abdulwahed, 2017). In principle, this is their role and purpose. Most of the knowledge to invent and build the tools that shape our environment is borne and tested in these institutions prior to deployment to the market (Nair-Bedouelle, 2021). Through classroom and end of course design projects, student engineers craft innovative designs for environmental and agricultural use. In ideal situations, these innovative prototypes are sent out and fine-tuned in the industry to make them marketable. However, in some developing countries, notably in Africa, these projects, often with great potential, are rarely transitioned to the industry and typically end up as mere academic exercises with no clear path to commercialization or practical implementation. In developed countries on the other hand, the majority of engineering schools often contribute directly to industrial development of tools and technologies that address societal challenges, thereby validating their existence and purpose. For example, there is a close tie between the Silicon Valley and Stanford University where mentorship and collaboration help turn student projects into market-ready products and technologies. The University of California, Davis once developed the vermifiltration technology that is today commercialized by the BioFiltro company (Dore et al., 2022; Agribusiness Review, 2024). Learning from these examples and doing more to move major scholarly innovations in developing countries from the conceptual stage to practical application or market deployment is crucial for sustainable development. Doing this, will promote the effective harnessing of local innovations that can revolutionize agriculture and environmental management in these regions and reduce the cost of importing tools and technologies (Wormley, 2013).

    Motivation

    There is need to bridge the gap between innovative engineering designs from academic institutions and their practical application in the market, particularly in the context of agricultural tools in developing countries. Specifically, there is need to:

    Case Study

    This article considered the case of the Department of Agricultural and Environmental Engineering (AEE) of the College of Technology, University of Bamenda, Cameroon. This department is host to two engineering streams (agricultural engineering and environmental engineering), graduating an average of 35 undergraduate and 15 graduate students yearly. Like many other engineering departments in developing countries, this department creates innovative tools and technologies to address agricultural and environmental challenges, primarily through student projects. However, the lack of pathways to move these tools beyond academic boundaries into the real world means that the prototypes accumulate each year with their fate unknown, perhaps waiting to be discarded (Figure 1). The industry is usually not integrated into the academic process, leading to the gap between what is taught and designed and what is needed in the market (Meyer & Norman, 2020). Departments like AEE in developing countries often lack access to infrastructure and resources needed to support the commercial development of students’ innovations. This leads to wasted potential. Figures 2 to 4 showcase some specific innovative tools developed within the AEE department which if further developed, can have the potential to change the agricultural landscape of the region. The animal feed mixer shown in Figure 4 for example, has the potential to help livestock farmers in cutting down importation costs, saving time and labor, and improving livestock health and productivity. If produced locally, farmers will be encouraged to mix their own feed, and demand for locally grown feed ingredients such as grains, legumes, and by-products will increase, thereby supporting local economies.

    Figure 1:Accumulation of student projects left abandoned after defense.

    Recommendations

    Industry-Academia Collaboration

    There is need to establish formal partnerships between engineering schools and local industries to align academic projects with market demands. Although industry professionals rarely have time and may often have differing priorities and resource constraints, they should be involved in mentoring students and guiding project development to ensure relevance and feasibility. Joint ventures can also be created within universities where students and industry professionals can collaborate on transforming projects into market-ready products (Ankrah & Omar, 2015). A good example is the Stanford University’s StartX Accelerator Program (https://web.startx.com/) which brings together students, industry professionals, and entrepreneurs to mentor student startups and help transform them into market-ready products.

    Establish Commercialization Pathways

    There is need to invest in prototype development and testing facilities within universities to allow students to advance their designs beyond the conceptual stage. Market research should be integrated into the curriculum to educate students on how to assess the commercial viability of their projects and this should be linked up with funding mechanisms, to support the transition of student projects from concept to market. It could be challenging to translate research into marketable products but institutes like the Massachusetts Institute of Technology (MITMECHE, 2024) have succeeded in running industry immersion programs where students work directly with industry mentors to solve real-world problems. Within this program, a number of projects have been commercialized with students gaining access to capital and industry networks.

    Curriculum and Policy Reforms

    Incorporating industry-driven projects into the curriculum is of essence, to ensure that student projects are aligned with real-world agricultural and environmental needs. Institutions may advocate for policies that facilitate (e.g. through incentives) industry investment in academic innovations and provide support for the commercialization of student projects. In addition, there’s need to enhance the curriculum to include skills in entrepreneurship, project management, and business development, which ultimately prepares students for the practical challenges of bringing innovations to market. A typical example here is the Finland’s education reforms that focus on personalized learning, critical thinking, and collaboration with industry (NCEE, 2024). Schools partner with local businesses, and the curriculum is continuously updated to reflect changes in technology and society.

    Infrastructure Development

    We need to develop innovation centers of excellence within engineering schools that focus on agricultural and environmental tool development and encourage research in agricultural engineering by vigorously investing in state-of-the-art engineering laboratories that help students in polishing their local designs to make them competitive and marketable. This involves heavy investment cost but the ripple effect in the long term is significant. Universities need to establish technology transfer offices to manage intellectual property and facilitate the commercialization process. These offices must liaise with counterpart institutions in the developed world for mentorship and joint funding programs. For example, the Georgia Institute of Technology has established an Advanced Technology Development Center (ATDC, https://atdc.org/), a start-up incubator and innovation center that focuses on supporting technological innovation in engineering and science. This center is also known to provide resources like lab spaces, mentorship, business development support, and access to capital. It also connects students and faculty with industry partners to drive commercialization.

    Conclusions

    Transforming innovative engineering designs from academic institutions into commercially viable agricultural tools requires a multifaceted approach involving industry collaboration, commercialization pathways, curriculum reforms, and infrastructure development. Although often hampered by limited funding, resistance to change and difficulties in finding industry partners, it is crucial to bridge the gap between academia and industry, where developing economies can harness local innovations to revolutionize agriculture, drive economic growth and enhance environmental sustainability. The case of the University of Bamenda serves as a compelling case for the potential impact of such initiatives in realizing the untapped potential of student innovations.

    References

    Abdulwahed, M. (2017). Technology Innovation and Engineering Education and Entrepreneurship (TIEE) in engineering schools: Novel model for elevating national knowledge based economy and socio-economic sustainable development. Sustainability, 9(2), 171. DOI: https://doi.org/10.3390/su9020171

    Agribusiness Review, 2024. BioFiltro Collaborates with UC-Davis to Study Biodynamic Aerobic System. https://www.agribusinessreview.com/news/biofiltro-collaborates-with-ucdavis-to-study-biodynamic-aerobic-system-nwid-952.html

    Ankrah, S., & Omar, A. T. (2015). Universities–industry collaboration: A systematic review. Scandinavian journal of management, 31(3), 387-408. https://doi.org/10.1016/j.scaman.2015.02.003

    Dore, S., Deverel, S. J., & Christen, N. (2022). A vermifiltration system for low methane emissions and high nutrient removal at a California dairy. Bioresource Technology Reports, 18, 101044. https://doi.org/10.1016/j.biteb.2022.101044

    Meyer, M. W., & Norman, D. (2020). Changing design education for the 21st century. She Ji: The Journal of Design, Economics, and Innovation, 6(1), 13-49. https://doi.org/10.1016/j.sheji.2019.12.002

    MITMECHE (Massachusetts Institute of Technology, Department of Mechanical Engineering), 2024. Industry Immersion Projects. https://meche.mit.edu/alliance-i2p-program#:~:text=Students%20get%20real%2Dworld%20experience,MIT%20during%20the%20industry%20project. Accessed, December 2024.

    Nair-Bedouelle. (2021). Engineering for sustainable development: Delivering on the Sustainable Development Goals. (S. Nair-Bedouelle, Ed.) London: Royal Academy of Engineering.

    NCEE (National Center on Education and the Economy) 2024. Top Performing Countries: Finland. https://ncee.org/country/finland/#:~:text=The%202021%20curriculum%20shifts%20from,VET%20schools%20are%20discussed%20below. Accessed, December 2024.

    Wormley, D. (2003). Engineering education and the science and engineering workforce. In Fox, M.A. (2003). Pan-Organizational Summit on the U.S. Science and Engineering Workforce: Meeting Summary. National Academy of Sciences, National Academy of Engineering, Institute of Medicine. ISBN: 0-309-52530-6. (pp. 40-46).

    Training Future Professionals in the Field of Smart Agriculture

    Ajit Srivastava1* and Daniel Uyeh1

    1Biosystems and Agricultural Engineering, Michigan State University

    *Corresponding author: srivasta@msu.edu

    KEYWORDS: smart agriculture, modern agriculture, African agriculture, digital agriculture, precision agriculture

    Abstract

    Smart Agriculture includes application of emerging technologies such as sensors systems, automation, robotics and drones, big data, internet of things (IoT), and modeling and decision-support systems to make agriculture productive, sustainable, resilient and profitable. These technologies are deployed at every phase of production agriculture including planning and documentation, planting, growing and harvesting, and even in post-harvesting operations3. Areas of application include yield, soil, and nutrient mapping, variable rate applications, controlled traffic and guidance systems, and soil and crop health to name a few. The field of Smart Ag is a growing field as the use of technologies such as yield monitors, GPS guidance systems, GPS enabled sprayers and GPS enabled planters has steadily increased since 2005. The use of GPS guidance systems grew from 4% in 2005 to 65% in 2018. It is estimated that IoT has the potential of increasing agriculture productivity by 20% by the year 2025. About 70-80% of farm equipment sold today has an element of precision agriculture. It is also estimated that there will be over 225 million connected devices used in agriculture in 2024. Predictive weather modeling and precision agriculture can reduce crop losses by 25%.

    African agriculture faces many challenges such as small farm size, low productivity, lack of mechanization, and degraded soils. Modernizing African agriculture is a way to address many of these challenges. Smart Ag should be an integral part of modernizing African agriculture as it holds much potential to address many challenges it faces today.

    Smart Ag is a is a rapidly growing field as evidenced by commercially available Smart Ag technologies, product catalogs, trade shows and professional conferences4. Smart Ag is an interdisciplinary field that includes disciplines such as agricultural/biological/biosystems, mechanical, and electrical engineering, ag. technology and systems management, and computer and data science. To sustain this growth, it is necessary to provide a path for students from these various related fields to build on their disciplinary technical expertise to engage/contribute to this growing field. This paper describes the concept of a Minor in Smart Ag at Michigan State University.

    Introduction

    Global agrifood systems face many challenges including population growth, climate change, rapid urbanization, and diet transformation.  It is expected that by the year 2100 the global population will reach 10.9 billion with Africa and Asia comprising more than 80% of the population (United Nations, 2022).  It is also projected that by 2050 the global middle class will increase to 70% resulting in a significant increase in demand for animal protein.  According to the Global Harvest Initiative 2014 Gap Report (GHI, 2014), by 2030 demand for poultry will increase by 63%, milk by 55% and meat by 44%.  Producing this extra amount will require land, water, and energy resources that are already limited.  For example, agricultural land will shrink to 0.17 ha/capita in 2025 from 0.44 ha/capita in 1960.  It is also expected that by 2030 energy demand will increase by 50% (IEA, 2023) and water by 30% (IFPRI, 2016).  To meet the growing demand, global agriculture will need to be productive, efficient, resilient, and sustainable.  Emerging engineering technologies and innovations such as artificial intelligence, data analytics, sensors, and sensing (including remote sensing), Internet of Things (IoT), automation, robotics and drone technologies hold much promise in meeting these challenges. 

    Smart Ag technologies are based on many engineering disciplinary expertise. The rapidly growing field of Smart Ag provides professional opportunities for many engineering majors. In general, engineering students are unaware of these opportunities and as such there is no clear pathway for them to learn of these and be trained for these opportunities. However, this need has been well recognized as the University of Florida has created a graduate certificate program in Smart Ag (Burk et al., 2024). The College of Engineering and College of Agriculture and Natural Resources at Michigan State University are taking a slightly different approach to creating a pathway to train undergraduate engineering students in the field of Smart Ag. They are offering a Minor in Smart Agriculture. This paper highlights the key element of the MSU Smart Ag Minor.

    Goal and Objectives

    The long-term goal of this project is to train engineering students to serve the growing need for qualified professionals in the field of smart agriculture. The specific objective was to develop an engineering minor that can be completed by engineering majors that will train them to apply their disciplinary expertise to the broad field of smart agriculture.

    Minor in Smart Ag

    To complete a Minor at Michigan State University, students must complete 16-17 course credits that are not required in their major. The main principle in designing this minor was to create a pathway for students to apply their disciplinary expertise in the broad field of Smart Ag while gaining sufficient knowledge of the many areas of application in agriculture. We first offer an introductory one credit course targeting Sophomores in Engineering entitled, Introduction to Smart Agriculture (BE 221). This course is designed to expose students in various disciplines to opportunities in the field of Smart Ag. If they find this field interesting and are willing to pursue this minor, they will then take a 3-credit course in principles of precision agriculture (BE 321) to get in-depth exposure to the various aspects of production agriculture through the lens of precision agriculture – a subset of Smart Ag. Students then take two courses (6-7 credits) from a group of carefully selected interdisciplinary courses, based on their own major, that give them a solid technical background based on their area of interest. After completing this course requirement, students will take two additional courses (6 credits) in ag. sensors and robotics and in crop modeling and optimization. Table 1 includes the course requirement for the Smart Ag minor at MSU.

    Conclusion

    The Smart Ag. minor has been approved by the University and is scheduled to be offered in Fall 2024 semester. The Biosystems and Agricultural Engineering (BAE) department has provided leadership in developing this program and will manage the program. However, this is a college-wide effort in collaboration with Mechanical Engineering, Electrical and Computer Engineering, Computer Science and Engineering in recognition of the fact that the field of Smart Ag must take an interdisciplinary approach.

    Table 1.Course Requirements for the Smart Ag Minor at Michigan State University

    Students must complete a minimum of 16 credits from the following:credits
    1. All of the following courses (10 credits)
      1. BE 221 Introduction to Smart Agriculture
      2. BE 321 Principles of Precision Agriculture
      3. BE 421 Sensors and Robotics for Agricultural Systems
      4. BE 422 Crop Modeling and Optimization
    2. Two of the following courses (6-7 credits)
      1. BE 449 Human Health Risk Analysis for Egr. Controls    
    3. BE 456 Electrical Power and Control
    4. BE 481 Water Resources Sys Anlys. & Modeling
    5. BE 482 Engineering Ecological Treatment Systems    
    6. CSE 404 Introduction to Machine Learning    
    7. CSE 440 Introduction to Artificial Intelligence    
    8. CSE 480 Database Systems        
    9. CSE 482 Big Data Analysis
    10. CSS 467 Bioenergy Feedstock Production     
    11. ECE 416 Digital Controls        
    12. ECE 417 Robotics        
    13. ECE 431 Smart Sensor Systems        
    14. ECE 434 Autonomous Vehicles        
    15. ECE 477 Microelectronic Fabrication        
    16. ME 417 Design of Alternative Energy Systems    
    17. ME 451 Control Systems        
    18. ME 456 Mechatronic Systems Design    

    1
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    3
    4
    3
    Total Credits16-17 credits

    References

    Burks, T., Watson, A., Fredrick, Q., Migliaccio, K., & Lu, R. 2024. Creating Parallel Smart Ag Systems Certificate Programs for Engineering and Applied Science Graduate Students. J. ASABE (in press). http://doi.org/10.13031/ja.15358

    Global Harvest Initiative (GHI), 2014 GAP Report. www.globalharvestinitiative.org

    IFPRI (International Food Policy Research Institute). 2016. Blog: Markets, Trade, and Institutions. Climate, water, and the economy. https://www.ifpri.org/blog/climate-water-and-economy/

    IEA (International Energy Agency), 2023. World Energy Outlook 2023. www.iea.org

    United Nations. 2022. World Population Prospects. United Nations Department of Economic and Social Affairs, Population Division, New York, NY.

    Entrepreneurship and Business Model Development

    Ranjani Krishnan1*, Satish Joshi1, Terry Howell2

    1Michigan State University, USA,

    2University of Arkansas, USA,

    *Corresponding author: krishnan@msu.edu

    KEYWORDS: entrepreneurship, agribusiness, business model, business model canvas, value generation, training

    Introduction

    The one-day training workshop on entrepreneurship and business model development was aimed at developing an understanding and the ability to articulate the purpose, role and importance of business models, including the key elements of a business model and the interactions and interdependencies among the elements. Additionally, the exercises enabled participants to critically evaluate, design, and recommend business models for an entrepreneurial new venture. The workshop consisted of seven sessions that walked the participants through the 12 building blocks of a business model canvas (BMC) shown in Figure 1, which extends the Osterwalder & Pignuer (2010) BMC to cover social enterprises that seek public/donor funds and need policy support. At the end of the workshop, the participants developed preliminary business models for their potential product ideas.

    Structure of Sessions

    The first session introduced the concept of business model, which describes the rationale of how an organization creates, delivers, and captures value, and provides a blueprint for a strategy to be implemented through organizational structures, processes, and systems. The twelve elements of a BMC were presented that address feasibility (Key partners, Key activities, Key resources and Governance), desirability (Customers, Extended beneficiaries, Value propositions, Channels, and Customer segments), and economic viability (Cost structure, Revenue streams and Impact). A few examples of famous start-up companies were described. This was followed up with a brainstorming session for product ideas around which teams could be formed. Team formation exercises discussed the team mission, objectives, commitments, resources, and risks using a team alignment map.

    The next two sessions focused on the customers. The goal was to identify the direct users of the product or service that have a direct relation to the value delivered, and their specific and detailed characteristics in terms of demographics, lifestyles, behavior, etc., as well as the extent of knowledge (concrete and perceptual) about the customers. A customer insights template was used to clearly describe the customer jobs and the associated pains (e.g. costs, effort, time, access, barriers, risks, perceptions) and how the proposed business idea ameliorates these pains to create gains. Discussions covered customer segmentation based on distinctions in offers, distribution channels, relationships, and profitability. Finally, the role of extended beneficiaries in a social enterprise setting was presented. A value proposition canvas was employed in a group exercise to clearly identify not only the value created for direct and extended beneficiaries, but also the environmental, public, and societal impacts.

    The fourth session discussed customer channels and customer relationships. Customer channels describe how a company communicates with and reaches its customer segments to deliver the value proposition. The channels help raise awareness about products and services, help customers evaluate the value proposition, enable purchases that deliver the value proposition, and provide post-purchase support. The economic benefits of maintaining customer relationships including customer lifetime value, acquisition of new customers through word-of mouth networks and the relatively high costs of acquiring new customers were discussed. Ways to maintain extraordinary customers relationships through deeper knowledge of customers, keeping in touch, providing customized service, and prompt redressal of problems were touched upon. Discussions also covered opportunities and challenges in maintaining customer relationships in rural contexts. The importance of listening to the customers, and helpful rules for skillful listening were presented. The group breakout exercise applied the lessons from the discussions to their specific product ideas.

    Mapping and quantifying various revenue streams including sales revenues from different customer segments and donor/public grants was the spotlight of the fifth session. Diverse and dynamic pricing mechanisms tailored to customer segments, product categories, payment methods, subscription services, leasing etc. were touched on. Spreadsheet exercises were implemented for quantitatively mapping future revenue growth.

    The sixth session covered key activities and key resources that are needed to: produce and deliver the value proposition, operate the value chain, maintain customer relationships, and manage revenue streams. Key activities across business functional areas such as procurement, production, marketing, financing, and stakeholder management were outlined. The importance of comprehensively assessing needed tangible and intangible resources, including physical, intellectual, human and financial capital was stressed upon.

    The cost structure is driven by the key activities and key resources. Cost accounting issues such as cost accumulation, assignment, tracing and allocation of direct and indirect costs, and activity-based costing were examined and elaborated using examples. Discussions touched on the generation of projected financial statements namely income statement, balance sheet and cashflow statements, drawing on the cost and revenue flows.

    The final session aimed at identifying the key partners needed for delivering the value proposition, the roles they are expected to play in terms of activities and resources, and the governance mechanisms that facilitate synergistic coordination. The importance of leadership, power dynamics and management processes for decision making, communication, operational control, risk management, grievances etc. were briefly discussed. The final wrap-up exercise involved an integrated analysis of the BMC with all the completed twelve building blocks.

    Concluding Remarks

    Capacity building for entrepreneurship and agri-business management is critical for agricultural development in low- and medium countries in Africa. This training workshop was aimed at introducing BMC for social enterprises and its building blocks, to potential entrepreneurs with business ideas. The initial design was for in-person interactive workshop. However, despite the limitations, the online format worked reasonably well.

    Acknowledgments

    Funding support for Satish Joshi’s work was provided by USDA Hatch Grant MICL12075.

    References

    Osterwalder, A., & Pigneur, Y. (2010). Business model generation: a handbook for visionaries, game changers, and challengers (Vol. 1). John Wiley & Sons.

    PILLAR IV: INFRASTRUCTURE AND POLICY FRAMEWORKS

    Pursuing Scale in Digital Tools for Data-Driven African Agrifood Systems through a Reference Architecture

    R. Andres Ferreyra1*, Johannes Lehmann2

    1Syngenta Crop Protection, Murray, Kentucky, USA.

    2DIN, Berlin, Germany

    *Corresponding author email: andres.ferreyra@syngenta.com

    KEYWORDS. data, standards, interoperability, ISO, AI, scale, digital agriculture.

    Introduction

    Agricultural and food production is a complex, adaptive process involving hundreds of management decisions per crop season. In the past these decisions were often driven by traditional local customs; in our present rapidly changing world, they must increasingly be made based on data, scientific principles, and a variety of models, statistical and otherwise. While data-driven decision making is complex at any operation size, smaller producers are more vulnerable: they have less access to resources such as crop inputs, agronomic advice, financial services, risk management, equipment, digital literacy, and affordable internet access. They stand to gain much from easier access to accessible decision-support technologies.

    This idea of data-driven, principled decision-making is foundational to the concept of smart farming, but it requires data that is correct, complete, timely, available, and understandable. A major impediment to reaching this state is the lack of data interoperability in the industry: Hardware and software systems from different manufacturers just do not “talk to one another” due to a proliferation of proprietary data formats, code lists, and different ways of representing the meaning of data. The inevitable consequence is that practitioners must spend excessive amounts of time reformatting and translating data before it can be used to create value.

    African technologists seek to innovate in this context, developing data-driven agrifood system tools with the prospect of both helping solve regional problems (including those of smallholders) cost-effectively and also generating revenue from international users. The lack of data interoperability standards, however, makes these efforts difficult to scale to being cost-effective and sustainable.

    Standards provide a powerful way to enable data interoperability in an industry. Some good examples of such enablement include web standards (W3C, 2024) and business process modeling standards (OMG, 2024). Unfortunately, the agriculture and food industries are lacking in data standardization beyond agricultural machinery and supply chain traceability.

    Recent ISO Activities on Agrifood Systems Data Standardization

    In mid-2021 the International Organization for Standardization (ISO) chartered a Strategic Advisory Group for Smart Farming (SAG-SF) with a triple mandate:

    The final report (ISO, 2023) containing the strategic roadmap was approved by ISO’s Technical Management Board and motivated several subsequent efforts. Chief among them:

    ISO Technical Committee 347, Data-driven agrifood systems

    One of the impediments to data interoperability identified by the SAG-SF was the lack of a permanent home for relevant standards development. This led to the convening, in early 2024, of a new technical committee for this purpose, ISO/TC 347, Data-driven agrifood systems (“TC 347”). This committee met for the first time in an online plenary on April 10, 2024. As of this writing, its membership (Figure 1) extends to 43 national standards bodies or NSBs (e.g., ANSI for the USA, BOBS for Botswana, DIN for Germany, SON for Nigeria). Unfortunately, and similar to what happened with the SAG-SF, the Global South is under-represented in the committee.

    Figure 1:Membership of ISO/TC 347. The darkest shade of blue represents the secretariat, held by DIN (Germany). Participating members (24) are shown in a lighter blue. Observing members (19) are shown in the lightest shade. Note, as indicated by red ovals, that as of this writing the Global South, which could stand to benefit greatly from ISO/TC 347’s standards, is under-represented therein.

    This under-representation was expected and is the object of frequent discussion. The primary reasons given by NSBs are:

    International Workshop Agreement IWA 47

    A lack of representation by the Global South poses a responsible innovation (Meijer et al., 2023) dilemma for TC 347. On one hand, participants in TC 347 are eager to begin development on standards that can quickly make a positive impact on the industry’s ability to interoperate. Examples of this perceived “need for speed” include:

    On the other hand, data standards development, seen as a form of potentially disruptive technological innovation, could conceivably bring unintended harm to actors within agrifood systems, hence the responsible innovation component. While TC 347 is, as of this writing, preparing to ballot the creation of a standing advisory group on responsible innovation in its midst, the SAG-SF also made recommendations for short-term measures that can help balance the aforementioned need for speed, with the need for caution implicit in a responsible innovation framework. Specifically, the SAG-SF recommended working on a reference architecture (Garcés et al, 2021) as a tool that can provide quick, useful guidance in the direction of greater interoperability. Moreover, the recommendation was to conduct the early stages of work on a reference architecture using an ISO mechanism called an International Workshop Agreement, or IWA.

    The dimensions of responsible innovation include anticipation, inclusion, reflexivity, responsiveness, and equity (Hurst & Spiegal, 2023). While ISO’s protocols and guiding principles emphasize equity and decision making through consensus, as mentioned earlier, the practical challenges faced by NSBs in the Global South make participation from African experts difficult. The IWA mechanism targets this problem, enabling inclusion though minimal administrative overhead, thus making it possible for experts to join the IWA directly, without the need for identification by, and mediation from, an NSB.

    International Workshop Agreement 47, Reference architecture for data-driven agrifood systems, ensued, planned as a series of approximately four monthly workshops starting in July 2024 that would allow all kinds of agrifood systems stakeholders worldwide to have a seat at the table. While this is an incomplete solution to the responsible innovation needs of a global industry, it does represent a significant step toward greater inclusion. The challenge: ensuring participation.

    AMAA and the Mini-Workshop on Data Interoperability Infrastructure

    One of the authors (R.A. Ferreyra) had been exploring opportunities for collaboration between the Alliance for Modernizing African Agrifood Systems (AMAA) and ISO’s smart farming work since 2022. The Imagining African Agrifood Systems Looking Forward summit organized by AMAA in November 2023 enabled taking concrete steps in this direction, by making the AMAA membership aware of progress on the ISO front. This was followed on 31 May 2024 by a mini-workshop to raise awareness of IWA 47 and the opportunity for it to help advance AMAA’s agenda (e.g., regarding AMAA’s infrastructure pillar), and to provide prospective IWA 47 participants with some level-setting advance information to help ensure their successful participation in IWA 47. During the 120-minute workshop, participants were asked to respond to the questions shown in Table 1, meant to elicit attitude / engagement towards standards, and the effectiveness of the mini-workshop itself.

    Results

    AMAA Mini-Workshop Results

    The mini-workshop was held on May 31, 2024, with approximately 20 participants. The program consisted of three parts:

    Table 1.Questions asked of participants in the AMAA Mini-workshop on Data Interoperability Infrastructure

    Item
        Can you share some examples of things that should interoperate but do not?
        Regarding data and African agrifood systems, where do you feel there is a need for speed? For caution? Why?
    Do you think the topic of data standards is important for Africa?
        Did the presentations provide the information you need to decide about IWA 47 participation?
    Will you participate in IWA 47?
    Why will / won’t you participate?

    Responses to the questions presented in Table 1 follow in Tables 2-4. There was limited engagement with the more domain-oriented questions (Tables 2,3), possibly influenced by a lack of familiarity with Zoom’s polling feature, but the opinions expressed were interesting and reflective of participants’ understanding of problems posed by poor interoperability.

    Table 2. Participant responses to “Can you share some examples of things that should interoperate but don’t?”

    Item
    Climate data with modeling software
    Cranes
    Market pricing
    Marketing data reaching local farmer and vice versa
    Security

    Table 3. Responses to “Regarding data and African agrifood systems, where do you feel there is a need for speed? For caution? Why?”

    Item
    Speed up training farmers on using technology and methodology. Caution not to go over their head
    Standards related to mechanization
    We need speed for postharvest loss reduction data
    Policy frameworks, data protections
    Caution does not enable change

    Reponses to the workshop evaluation questions shown in Table 4 were encouraging: most participants described the topic of data standards in Africa as “Very important”, and generally indicated an interest in participating.

    Table 4. Participant responses to four mini-workshop evaluation questions. Each row represents the answers of one participant.

    Do you think the topic of data standards is important for Africa?Did the presentations provide the information you need to decide about IWA 47 participation?Will you participate in IWA 47?Why will / won’t you participate?
    ImportantUndecidedNot likelyTime limitation
    Very importantAgreeVery likelyIt is important to be part of the frontier of Ag-Food data standardization in Nigeria
    Very importantAgreeSomewhat likely
    ImportantAgreeSomewhat likelyIt will depend on my availability
    Very importantAgreeVery likelyI will want to participate because I am interested in enhancing Nigeria's agrifood system
    Very importantStrongly AgreeVery likelyKnowledge sharing: I will be informed on information and insights on the latest developments in agriculture technology, data analysis and digital innovations
    Very importantAgreeVery likelyTo discover more about data architecture
    Very importantAgreeVery likelyWill definitely participate
    Very importantAgreeVery likelyStandardization is very important to manage data in the agrifood systems
    Very importantAgreeVery likelyInterest in data standards
    Very importantStrongly AgreeVery likelyMost likely

    IWA 47

    The first session of IWA 47 was held on July 9, 2024. Registrants included multiple experts each from Botswana, Egypt, Indonesia, Jamaica, Malawi, and Sudan (total: 30 experts from these six countries). These are Global South countries that had not been represented in the SAG-SF and are either not members, or are only observing members of TC 347 as of this writing. This represents about 18% of the 168 total registrants to date in IWA 47, and is a welcome step in the direction of inclusion. Investigation of whether the AMAA Mini-workshop influenced participation is ongoing.

    Concluding Remarks

    The Global South in general, and Africa in particular, must take a seat at the table of international agrifood systems data standards. This is necessary in order for those standards to have a lasting, positive impact that can advance the SDGs. Ensuring this inclusion through participation while enabling rapid progress on standards development and the multiple dimensions of responsible innovation will be an ongoing challenge that will require careful strategy. AMAA can benefit from and influence this process.

    Acknowledgments

    The authors gratefully acknowledge the assistance provided by the leadership of the AMAA in making the mini-workshop possible, as well as to Dr. Michael Ngadi and Dr. Daniel Uyeh for their valuable contributions planning and delivering it. We also thank the participants, and sincerely hope that the experience will be of lasting value to them.

    References

    Garcés, L., Martínez-Fernández, S., Oliveira, L., Valle, P., Ayala, C., Franch, X. & Nakagawa, E.Y. (2021) Three decades of software reference architectures: A systematic mapping study. J. Systems and Software, 179, 111004. https://doi.org/10.1016/j.jss.2021.111004 8

    Hurst, Z.M. & Spiegal, S. (2023). Design thinking for responsible Agriculture 4.0 innovations in rangelands. Rangelands 45:68–78 https://doi.org/10.1016/j.rala.2023.03.003

    ISO (2023). Final report of the Strategic Advisory Group for Smart Farming. Geneva, ISO. Retrieved from https://bit.ly/3MP0SXf

    Meijer, A., Wiarda, M., Doorn, N., & van de Kaa, G. (2023). Towards responsible standardisation: Investigating the importance of responsible innovation for standards development. Technology Analysis & Strategic Management, 1–15. https://doi.org/10.1080/09537325.2023.2225108

    OMG (2024). The OMG Specifications Catalog. Retrieved from https://www.omg.org/spec/#iso-specs

    W3C (2024). Web Standards. Retrieved from https://www.w3.org/standards/

    Climate and Water Data Policy Landscape and Impacts: an example from East Africa

    Margaret W. Gitau1*, Victoria Garibay2, Nicholas Kiggundu3, James Kisekka4, Victor Kongo5, Bancy Mati6, Daniel Moriasi7, Subira Munishi8

    1Purdue University, 2University of Amsterdam, 3Makerere University, 4Aidenvironment, 5Global Water Partnership Tanzania, 6Resource Plan Ltd and Jomo Kenyatta University of Agriculture and Technology, 7USDA-ARS, 8University of Dar es Salaam.

    *Corresponding author: mgitau@purdue.edu

    KEYWORDS. climate, water, data policy, Africa, agriculture

    Background

    Agriculture is by far the biggest user of water accounting for about 70% of freshwater use globally (World Bank Group, 2024). Meanwhile, agriculture continues to pose a formidable threat to water quality, presenting a variety of associated challenges to the environment, human and animal health, and overall societal well-being. The economic impact of water quality impairments can run into billions of dollars including: cost of water treatment; clean-up costs; costs to human health; and, losses in revenues from tourism and fishing (USEPA, 2012). Ironically, agriculture bears a substantial portion of related costs due to loss of productive land (through soil erosion) and the need for additional inputs (as inputs are lost to water bodies through runoff, leaching, or attached to eroded sediment). Therefore, in the move towards modernizing African agriculture and African agrifood systems in general, it is important to keep in mind the increased pressures this could exert on water resource systems considering both quantity and quality. Furthermore, it is important to account for current and future climate dynamics—which, for Africa, include rapid warming, decreasing rainfall totals, and pronounced incidences of floods and droughts (WMO, 2022)—and the impact these have and/or will have on water resource systems.

    Photos:Agriculture is by far the biggest user of water. Pollutants from agricultural lands pose substantial threats to the quality of water resources. Source: Adobe stock photos.

    In East Africa, water availability varies substantially with time, region, and climate (Table 1). Water deficits are common and inter-basin transfers are often used to alleviate the deficits. The integrity of water resources in the region (both quantity and quality of water) is affected by changes in land use, land management, and climate. For example: water quantity and quality in Simiyu River, Tanzania, are threatened by extensive land use changes comprising progressive increases in cultivated and built up areas driven by population growth in the contributing watershed and exacerbated by climate change (Kongo et al., 2023); in the Murchison Bay Watershed, Uganda, anthropogenic perturbations—primarily rapid urbanization—present substantial water quantity and quality challenges (Kisekka et al., 2023); in the Sasumua River Watershed in Kenya, land fragmentation with intensive agriculture and increased use of agricultural inputs along with rapid urban growth pose a threat to water quality in the Sasumua reservoir (Mati, 2023).

    Climate, water quantity, and water quality data play a critical role in water resources decision making and management in the face of such changes; yet, enormous challenges are often encountered when it comes to data availability and the ability to freely access available climate and water data in this region (Garibay et al., 2021; Gebrechorkos et al., 2018). This greatly impacts the ability to undertake relevant analytical studies needed to make informed water resources management decisions in the region. In this Executive Summary, we provide an example of analysis done in Kenya, Tanzania, and Uganda to discern key elements in water and climate data policies that are effective and uncover obstacles to the effective application of data in decision making, where data have been collected. We then explore what this might mean in the context of modernizing African agrifood systems, given a changing climate.

    Table 1. Estimates of precipitation and water availability in East Africa. Source: (World Bank Group, 2024).

    CountryAverage Annual Precipitation[a]
    (mm/year)    		Total Renewable Freshwater[a]
    (billion cubic meters)
    Burundi127410.06
    Djibouti2200.3
    Eritrea3842.8
    Ethiopia848122
    Kenya63020.7
    Rwanda12129.5
    Sudan2504
    Somalia2826
    South Sudan90026
    Tanzania107184
    Uganda118039
    [a] Based on 2001-2020 data.

    Study Details

    Premise

    Despite the acknowledged importance of climate and water data for water resources decision making and management in the study region, existing data were often incomplete, insufficient, sometimes very dated. In some cases, data were not at all available. Where data were available, they were not necessarily accessible. Sometimes the data were accessible, but were: 1) held by different entities; 2) in varied extents, forms, and formats; and/or, 3) lacking metadata among other concerns, all of which made their effective use a challenge. In general, climate and water information was limited whether due to availability, accessibility or both. Furthermore, tools to enable effective data use were often lacking. This pointed to the possibility that existing policy frameworks surrounding (climate and water) data and the management of these data were either lacking or ineffective, in essence, hindering availability, accessibility, and effective use of data for decision making and management.

    Approach

    The study comprised an analysis of different policy documents, government websites, and other official documents that addressed water and climate data. In addition to documents from Kenya, Tanzania, and Uganda, we looked at documents from: surrounding east African countries and other countries in Africa, as available; and, countries outside Africa in which data were readily available and accessible, and routinely used for water resources decision making and management—for a total of 10 countries. A survey, similar to what might be used for interviews, was developed and used to gather information from the aforementioned documents. The information gathered was then analyzed considering: 1) Meteorological, water quantity, and water quality data (level 1); 2) Different elements including: Format, instrumentation, destination etc (level 2); and, 3) Functionality and accessibility (level 3). A score of 1 (present) or 0 (not present) was assigned to each element as appropriate and a representation metric computed by theme (level 1) to indicate the proportion of all possible elements on a country-by-country basis. More details on this study are presented in Garibay et al. (2022).

    Results

    Figure 1 shows the scores for the different elements presented as a comparison between where data streams were functional and where they were not. Highlighted at the bottom are the elements that were found key to successful data streams—meaning that the data were readily available and accessible. Elements that were found to be key included: Destination (a database or storage system was available and specified as a home to the collected data); Service Provision (there is a commitment to making data accessible to the public); Responsibility (an organization or agency was specified as being in charge of collecting that data type); and, Format: (format of variables that will be collected is indicated directly or indirectly). Average values of the representation metric ranged from 0.15 (fewest elements) to 0.81 (most elements). The more the elements, the more likely it was that the data stream was successful. None of the countries had all of the elements, indicating that it was not necessary to have all the elements to ensure a successful data stream. In general, meteorological data were more readily available than water quantity and quality data attributable to the more involved and expensive process of collecting the latter. This situation is not unique to East Africa.

    Figure 1:Comparison between functional data streams and non-functional data streams considering key elements. Source: Garibay et al. (2022) CCBY 4.0.

    Concluding Remarks

    In general, the processes surrounding water quantity and quality data were not as well documented as for climate data. Water data were also much harder to get than climate data. Sometimes a better way to look at impact is by looking at what is lost due to the lack of a product or a resource, as opposed to what is gained by its presence. Without climate and water data that are accessible and reliable, water resources managers are forced to work with limited information which, in turn, limits the decisions or quality of decisions that can be made as well as the ability to manage decisions effectively. That coupled with the challenges of climate and land use/land management changes can lead to a cycle that is difficult to break out of, compromising the integrity of water resource systems and ultimately, agrifood systems that are very much dependent on the water resource. This is similar to trying to solve a puzzle for which pieces are missing. Without reliable and accessible data, valuable insights are likely to be missed and key patterns will not be discernable. Management response to changing conditions is impacted and it is difficult to plan ahead. The cost of wrong decisions can be high, as can be the whole process of trying to work around missing data, for example: imputing data, and/or making and verifying assumptions. To support the modernizing of African agrifood systems, climate and water data need to be available and accessible for use for the common good. Recognizing that there might be challenges with climate and water data as related to transboundary waters, we recommend that these data be at least available to and accessible by researchers and development practitioners.

    Acknowledgments

    This Executive Summary was made possible in part by the LASER-PULSE program through USAID’s Innovation, Technology, and Research Hub.

    References

    Garibay, V., M.W. Gitau, N. Kiggundu, D. Moriasi, and F. Mishili. 2021. Evaluation of reanalysis precipitation data and potential bias correction methods for use in data-scarce areas. Water Resources Management. DOI: https://doi.org/10.1007/s11269-021-02804-8

    Garibay, V.M., M.W. Gitau, V. Kongo, J. Kisekka, and D. Moriasi. 2022. Comparative Evaluation of Water Resource Data Policy Inventories Towards the Improvement of East African Climate and Water Data Infrastructure. Water Resources Management. DOI: https://doi.org/10.1007/s11269-022-03231-z

    Gebrechorkos S.H., S. Hülsmann, and C. Bernhofer 2018. Evaluation of multiple climate data sources for managing environmental resources in East Africa. Hydrol Earth Syst Sci 22:4547–4564

    Kisekka, J.W., N. Kiggundu, and D. Mugenyi. 2023. Land use management to reduce flooding in the Greater Kampala Metropolitan Area, Uganda. Policy Brief. https://bit.ly/policybrief_Uganda

    Kongo, V., F. Anderson, C. Kibugu, S. Munishi. 2023. Data for Decision: Enhancing Water Security in Tanzania. https://bit.ly/policybrief_Tanzania

    Mati, B.M. 2023. Data Driven Decision Support for Securing Kenya’s Water Towers. Policy Brief. https://bit.ly/policybrief_Kenya

    U.S. Environmental Protection Agency, USEPA. 2012. The Facts about Nutrient Pollution. Fact Sheet. U.S. Environmental Protection Agency.

    World Bank Group. 2024. World Development Indicators. Available at: https://datatopics.worldbank.org/world-development-indicators/. Last accessed, July 2024.

    World Meteorological Organization, WMO. 2022. State of the Climate in Africa. Available at: https://library.wmo.int/viewer/67761/download?file=1330_State-of-the-Climate-in-Africa-2022_en.pdf&type=pdf&navigator=1

    1. The image is reused under the Creative Commons Attribution-Non-Commercial-No Derivatives 4.0 International License.

    2. Reproduced with permission from FAO and EUROFISH under their copyright policy. This material is used for research purposes in compliance with the terms stated in the FAO copyright statement.

    3. McFadden, J., Njuki, E., & Griffin, T. 2023. Precision Agriculture in the Digital Era: Recent Adoption on U.S. Farms, EIB-248, U.S. Department of Agriculture, Economic Research Service.

    4. McFadden, J., F. Casalini, T. Griffin, & J. Antón. 2022. The Digitalization of Agriculture: A Literature Review and Emerging Issues, OECD Food, Agriculture, and Fisheries Papers Number 176, Organization for Economic Co-operation and Development, Paris, France.