Eliminating the Shortage Factor: Modeling the Behavior of Future Donations using Forecasting Techniques to Improve the Distribution Efficacy of a Food Bank.

摘要

Food insecurity is defined as the inability to provide food for oneself. As of 2011, more than 14.9% of American households suffered from food insecurity. Many individuals suffering from food insecurity obtain assistance from governmental programs and nonprofit agencies. Food Banks are one of many non-profit organizations assisting in the fight against hunger. They serve communities by distributing food to those in need through charitable agencies. Many of the food distributed by the food bank come from donations. These donations are received from various sources in uncertain quantities at random points in time. Due to this variability, predicting the quantity of future donations is challenging which can negatively impact their ability to properly allocate food. This research utilizes several forecasting techniques to predict future donations. In particular, the effectiveness of moving average, simple exponential smoothing, Holt's and Winter's methods, and Autoregressive Integrated Moving Average (ARIMA) are applied to historical data that is segmented by donation source, type, storage, receiving branch and a combination of variables. The results show that the appropriate technique is largely dependent upon the level analyzed. The resulting forecast is then used in a Supply Level Management Assessment (SLMA) to project equitable distribution. The tool is designed to be easy to manipulate and its applications can be used for all food banks.