Tactical planning of the production and distribution of fresh agricultural products under uncertainty

Abstract

We present a stochastic tactical planning model for the production and distribution of fresh agricultural products. The model incorporates the uncertainties encountered in the fresh produce industry when developing growing and distribution plans due to the variability of weather and demand. The main motivation for building this model is to make tools available for producers to develop robust growing plans, while allowing the flexibility to choose different levels of exposure to risk.The modeling approach selected is a two-stage stochastic program in which the decisions in a first stage are designed to meet the uncertain outcomes in a second stage. The model developed is applied to a case study of growers of fresh produce in Mexico and in a simulation of various scenarios to test the robustness of the planning decisions. The results show that significant improvements are obtained in the planning recommendations when using the proposed stochastic approach as compared to those rendered by deterministic models. For instance, for the same level of risk experimented by the producer, planning based on the proposed stochastic models rendered increases of expected profit of over 50%. At the same time when risk aversion policies were implemented, the expected losses decreased significantly over those recommended by deterministic planning models.