Unified Framework for Efficient, Effective and Fair Resource Allocation at Food Banks: Approximate Dynamic Programming Approach

Abstract

The evidence linking food insecurity, poor nutrition, and increased risk of chronic health problems combined with the high cost of health-care system to treat food insecurity, pose significant health threats and present challenges to the food banks system. Food banks personnel and policy makers must proactively seek new policies and practices that combat food insecurity and the diseases associated with it (diabetes and malnutrition for instance). We develop a framework for optimizing resource allocation for food banks. Our framework explicitly considers the effectiveness and efficiency measures of performance of the resource allocation problem faced by food banks and implicitly considers the equity performance measure. We measure effectiveness based on the nutritional quality of the allocation decisions, efficiency as the utility of served agencies and equity as fairness in share allocation among served agencies. To this end we develop a dynamic programming model where primary decision is how much to allocate/distribute of each product. To deal with the high-dimensional state space in the dynamic program, we construct approximations to the value function that are parameterized by a small number of parameters. Computational experiments using real-data obtained by one of the food banks in New York State, serving around 19,000 individuals per week, demonstrate the performance of the approach. More specifically, when compared against the policy implemented in practice, our algorithm demonstrates 7:73% improvement in total utility. Furthermore, when compared against the o_ine model, where randomness is revealed upfront, the gap between our algorithm and the offline is less than 9:50%. On the effectiveness side, our framework demonstrates 3.0% improvement in the nutrition of served population.