New collaboration on Artificial Intelligence to accelerate the Agroecological Transition

WorldFish and the University of Engineering and Management have signed an MoU to develop an Artificial Intelligence/Machine Learning (AI/ML)-based decision support system

Authors:

• Dr. Ayan Samaddar, Research Associate, WorldFish, India
• Prof. Subrata Saha, Professor, University of Engineering and Management, Kolkata, India
• Dr. Arun Padiyar P., WorldFish Lead, India

Agroecological practices can effectively address the global challenges of food security, biodiversity loss, climate change, and rural livelihoods. In India, selected practices are being tested and evaluated with food system actors through the Agricultural Living Landscapes initiative established by the Agroecology Initiative team and its partners in two states: Andhra Pradesh and Madhya Pradesh, located in southern and central India, respectively.

Given the diverse motivations of stakeholders to engage with these practices and to accelerate adaptation, evaluating agroecological transitions—especially those applied at scales larger than the farm level—is imperative. To this end, partners are currently developing an Artificial Intelligence / Machine Learning (AI/ML) based decision support system for stakeholders in aquatic food production. This system will consider various dimensions necessary to achieve higher levels of social and ecological sustainability, including economic returns, soil health preservation, ecosystem integrity, and stability in natural resource utilization.

To facilitate this process, WorldFish—one of our primary collaborators in India—has signed a Memorandum of Understanding (MoU) with the University of Engineering and Management in Kolkata, India. The decision-support tool will assist relevant stakeholders in their agroecological transitions, with the first component focusing on natural method-based integrated rice-fish farming techniques in collaboration with the Andhra Pradesh Community Managed Natural Farming (APCNF) initiative. Our framework is based on the Agroecological Principles outlined by the High-Level Panel of Experts of the Food and Agriculture Organization of the United Nations (HLPE) in 2019, which will guide both quantitative and qualitative evaluations of agroecological performance. The holistic overview of the work is presented in the Figure below:

We will consider the perspectives of multiple stakeholders (e.g., governments and rural farmers) to achieve consistent benefits and enhance normative foundations. The field data generated from questionnaires will be analyzed using robust data-driven (e.g., reinforcement learning) and analytical methods (e.g., heuristic and metaheuristic optimization techniques) to identify the critical success factors. Finally, a decision support system for progressing towards SDGs will be delivered with directly actionable and understandable guidelines. 

Leveraging our experience in both fieldwork and wet laboratory research, the team will analyze mathematically manageable concepts such as production syndromes, agents, barriers, and drivers of change across three primary analytical frameworks: socio-ecological, socio-technological, and social norms and networks. These frameworks will be linked to empirical evidence gathered from wet lab experiments.

This assessment will also identify scenarios where qualitative and quantitative performance-based scoring or indexing could be feasible. Such tools may serve as valuable resources for researchers, policymakers, evaluators, and food system actors— including farmers, market players, and consumers. This will aid implementing agencies, such as government departments and non-governmental organizations, in making informed decisions regarding the essential aspects needed for an agroecological transition.

[1] Mishra, P.K., Parey, A., Saha, B., Samaddar, A., Bhowmik, T.S., Kaviraj, A. and Saha, S., 2021. Performance analysis of composite carp culture policies in drought prone district Purulia in West Bengal, India. Aquaculture, 544, p.737018.

[2] Giri, S.S., Kim, S.G., Jung, W.J., Lee, S.B., Lee, Y.M., Jo, S.J., Hwang, M.H., Park, J.H., Kim, J.H., Saha, S. and Sukumaran, V., 2023. Dietary Syzygium cumini leaf extract influences growth performance, immunological responses and gene expression in pathogen-challenged Cyprinus carpio. Fish & Shellfish Immunology, 138, p.108830.

[3] Pramanik, S., Biswas, J.K., Kaviraj, A. and Saha, S., 2023. Assessment of the Present State and Future Fate of River Saraswati, India: Water Quality Indices and Forecast Models as Diagnostic and Management Tools. CLEAN–Soil, Air, Water, 51(4), p.2200321.