Agricultural supply chain optimization and complexity: A comparison of analytic vs simulated solutions and policies

Abstract

With worldwide food security emerging as a major policy issue moving forward, the structure and optimization of key agricultural supply chains is of growing importance. In turn, while many working models of supply chain optimization have been developed to ensure analytic tractability, others are building more precise characterizations of a supply chain as a complex system that may not be amenable to analytic solution. This research examines an important agricultural supply chain from the perspective of developing effective solutions to complex internal optimization issues that could ultimately affect food security. To this end, the Canadian wheat handling system is a complex export oriented supply chain that is currently undergoing extensive changes with respect to quality control. We develop both analytic and simulation models of this supply chain with the ultimate goal of identifying effective wheat quality testing strategies in a complex operational and regulatory environment. While the analytic model is founded on limited assumptions about individual behavior, agent-based simulation allows us to model farmers and handlers as rational and learning individuals who make decisions based on their own experiences as well as the experiences of others around them. We then make explicit comparisons between solutions and policies generated using the simulation approach against those generated by the analytically tractable model of the wheat supply chain. While the two approaches generate somewhat different solutions, in many respects they lead to similar conclusions regarding the overall testing and quality control issue in wheat handling.