Food Banks play an important role in assuaging hunger and improving food security in many nations worldwide. These organizations provide food and services to people in need. Food banks rely on food and cash donations that occur infrequently, to meet their objectives. In a highly uncertain environment such as this, balancing the supply and demand of food is challenging considering the limited availability of resources and the complex system. This research addresses these challenges and presents and analyses several statistical and mathematical models to facilitate the distribution of food to the food insecure in a sustainable and effective manner. The objective of this research is to develop data-driven models and analytical techniques and developing decision support frameworks to assist the food bank administrators in understanding the dynamics of supply and demand of food donations and improve the prediction accuracies of the food supply and demand behavior at various levels of planning to ensure equitable and efficient distribution of food to the food insecure.   

 

First, a systematic review was conducted to research the evolving literature in the field of food bank logistics. Perusal of the literature shows that research in the field of food bank logistics is in evolving phase and issues pertaining to fairness, sustainability, cost reduction, food quality and nutrition, data uncertainty, and food waste study have not been reviewed extensively. Second, for understanding the food supply behavior, a novel hybrid model combining ARIMA and neural network autoregressive (NNAR) model was proposed for univariate analysis and the work was extended to conduct a multi-variate numerical analysis implementing machine learning algorithms with Random Forests (RF) best capturing the complex structure of the data. Thirdly, to understand the dynamics of the food demand behavior, a Gaussian Mixture Model (GMM) clustering method is implemented to identify the possible causes of food insecurity in a given region by means of understanding the characteristics and structure of the food assistance network in a particular region, and the clustering result is further utilized to explore the patterns of uncertain food demand behavior and its significant importance in inventory management and redistribution of surplus food thereby developing a two-stage hybrid food demand estimation model with the proposed method significantly improving the prediction accuracies.  

 

Finally, the results of the analytical methods implemented and developed to study the supply and demand of the food donations is extracted and used to develop a conceptual framework for designing a decision support system to apply visual analytics to a food bank’s decision-making process.