Firdaus Ahmed Choudhury, a doctoral student in the College of Information Sciences and Technology at Penn State, will deliver “From Manual Planning to Automated Routing: Scalable Sapling Distribution for Smallholder Farmers in India” as part of CSRAI's Young Achievers Symposium. This lecture is free and open to the Penn State community.
About the Talk
Artificial intelligence and optimization offer significant opportunities to improve how nonprofit organizations allocate limited resources and deliver services to communities. However, deploying these methods in real-world settings requires accounting for operational constraints that are often absent from standard problem formulations. This work explores these challenges through a collaboration with the Naandi Foundation, a nonprofit organization that distributes saplings free of charge to smallholder farmers across rural India. Naandi's distribution planning involves selecting trucks, allocating saplings, and constructing routes across a large rural road network, a process that is currently performed manually. To automate this process, we develop an AI-based decision-support system that models Naandi's distribution problem as a variant of the Vehicle Routing Problem (VRP). A Mixed-Integer Linear Program (MILP) captures real-world requirements including heterogeneous truck capacities, road-capacity restrictions, split deliveries, and routing precedence constraints. As exact optimization does not scale to Naandi's operational setting, we further develop a scalable heuristic framework combining cost-aware solution construction with metaheuristic optimization. Evaluated on real-world data covering a network of more than 700 locations, the resulting system generates complete distribution plans in approximately 19 minutes. Compared with a historical manually planned distribution run, these plans reduce operational costs by approximately 16% and travel distance by approximately 23%. This work highlights the potential of AI-based decision support to improve nonprofit operations while emphasizing the practical challenges of translating optimization research into systems that domain partners can deploy and use.
About the Speaker
Firdaus Ahmed Choudhury is a doctoral student in Informatics in the College of Information Sciences and Technology at Penn State, under the supervision of Dr. Amulya Yadav, and a member of the RAISE Lab. He received his bachelor's degree in Computer Science from Penn State. His research focuses on AI for social good, with an emphasis on explainability and the development of human-centered AI and decision-support systems for nonprofit and public-sector organizations, particularly those serving marginalized communities. His work combines artificial intelligence, optimization, and data science with close collaboration with domain partners. He has worked with organizations including the Naandi Foundation in India on improving agricultural resource distribution for smallholder farmers and the Paterson, New Jersey Fire Department on data-driven approaches to support fire inspection prioritization. He is a recipient of the Data Science for Social Good Fellowship and Penn State’s AI for Societal Impact Award for his work with the Naandi Foundation.
About the Young Achievers Symposium
The Young Achievers Symposium highlights early career researchers in diverse fields of AI for social impact. The symposium series seeks to focus on emerging research, stimulate discussions, and initiate collaborations that can advance research in artificial intelligence for societal benefit. All events in the series are free and open to the public unless otherwise noted. Penn State students, postdoctoral scholars, and faculty with an interest in socially responsible AI applications are encouraged to attend.