Resilient Models and Strategies for Agribusiness Supply Chain Problems under Supply and Demand Disruptions

Abstract

Agribusiness supply chains (ASCs) involve more sources of uncertainty than typical manufacturing supply chains due to characteristics such as long supply lead-times, seasonality, and perishability. Such characteristics make ASCs especially vulnerable to supply and demand disruptions. Thus, a novel approach to risk management (RM) is required. This thesis focuses on RM strategies for managing disruptions ASCs and investigates how different RM strategies may be used to design resilient ASCs. First, we investigate the usefulness of different RM strategies in a real-life case study of Zespri's kiwifruit supply chain under supply- and demand-side disruptions. We use optimisation models in order to find the best sets of RM strategies for managing key supply- and demand-side disruption risks. For supply-side disruptions, we investigate the effectiveness of a number of robust and resilient strategies. We develop a two-stage stochastic programming model, which integrates perishability, to conduct our analysis. The model maximises the expected profit by selecting optimal RM strategies and by making optimal tactical supply chain planning decisions. The results suggest that a mixed combination of robust and resilient strategies are most effective for mitigating supply-side disruptions; and as perishability increases, RM provides a greater relative improvement in the expected profit. Then for demand-side disruptions, we focus on RM strategies to ensure that ASCs are flexible enough to respond to market disruptions quickly and efficiently. Hence, we investigate the effectiveness of multiple RM strategies for building flexibility. A robust optimisation formulation is used to obtain solutions for an objective combining expected profit and risk. The results show that the proposed demand-based RM strategies improve both expected profit and risk measures. Although market-related RM strategies are important for both risk-neutral and risk-averse decision makers, only risk-averse decision makers are interested in transshipment related RM strategy. Next, we simplify these supply and demand disruptions down to a single type, namely port disruptions (either in the supply- or demand-side of the supply chain) in order to study the problem analytically. We also define a variety of resilience measures and investigate the impact of RM on these measures as well as profit. For this problem, the strategy for managing risk is either a risk acceptance (RA) strategy or a backup transportation strategy, which has variable backup capacity during port closure disruptions (i.e., different proportions of backup capacity can be replaced in each disruption period). We derive optimal decisions for supply chain planning and for when to use each strategy and how that strategy should be used. These optimal decisions on RM strategy depend on the fixed and variable costs of the strategy as well as the supply chain's costs/profit and the level of perishability. Hence, this study provides a RM strategy selection framework and suggests that we should not select the strategy based solely on the cost and benefit of the strategy but also based on the specific supply chain context. Overall, this thesis provides investigations on RM strategies to build resilience in ASCs under key supply- and demand-side disruption risks. The thesis particularly focuses on RM strategies that improve the recovery capacity in ASCs with perishable products.