A stochastic model for developing speculation-postponement strategies and modularization concepts in the global supply chain with demand uncertainty

Abstract

Modern business traits have resorted companies to the global scale and complex supply chain operations to fulfill the customer demand around the world. In doing so, companies have to anticipate volatile and uncertain global demand, while adjusting their supply chain operations to various regional difference risks. Addressing these issues has motivated our study to propose a two-stage stochastic model of a supply chain network (SCN) design integrated to postponement and modularization concepts. The proposed research considers multi-period planning, involving procurement, processing, and sales of a multi-variant product with modular product architecture to model the operations in the supply chain network. The configuration of this comprehensive SCN is integrated to postponement strategy that is adjusted to speculation strategy, accommodating both forecast and demand-driven operations. The speculation-postponement strategy allows the optimization of the forecast-driven speculation strategy operations planning and service level of multiple items with varying value and variances, as well as the postponement strategy operations to be all adjusted to both global demand uncertainty and regional differences factors. The complexity of the model leads to the usage of the sample average approximation method to yield a reliable solution efficiently. The results of configuring a global notebook computer SCN case study show that speculation-postponement (SP) strategies with early postponement can reduce excessive production processes, lost sales, deadstock, and increase the inventory turnover rate, allowing better uncertainty risks mitigation in optimizing the performance of the global SCN. Under various demand uncertainty levels, SP strategies can improve the supply chain operations efficiency while also being responsive in handling the unexpected changes to the uncertain demand. The results also highlight the importance to configure the SCN and SP strategy altogether with optimizing speculation strategy operations planning and production service level decisions to prevent sub-optimal solutions, such as the significant offset of the expected SP performance when the service level is not adjusted to demand uncertainty, showing how forecast-driven planning inaccuracy can be costly. Configuring SCN with the right SP strategy has consistent performance, potentially increasing SCN robustness and resiliency against global demand uncertainty risks.