Supply chain production planning: Modeling dynamic lead times and efficient inter-cell inventory policies

A practical and effective framework for decentralized, yet coordinated, management and control of supply chains is proposed. It exploits the natural time-scale-based decomposition of decisions with different scope and functionality while capturing salient dynamic and stochastic characteristics of production systems. Coordination of planning and operational decisions of manufacturing cells interacting either under common ownership or under a multiple-tier-supplier relationship is achieved through optimal and consistent production targets and safety stock levels for each cell and part type during each time period, determined by a master problem over an appropriately selected planning horizon. The master problem interacts with sub problems to learn and model the impact of production target and safety stock level decisions on the required lead time. Appropriate algorithms involving different information exchange requirements are proposed and developed for the iterative interaction of a centralized master problem with decentralized sub problems. The objective of these algorithms is to induce a stable dynamic interaction that converges to a high Quality of Service at minimal WIP levels and production delays.; Computational experience is provided to elaborate the value of variable lead time information and the convergence behavior of the proposed master-problem-solving algorithms in determining production targets. Networks of manufacturing cells producing multiple part types are used as representative examples of a complex supply chain to illustrate the efficiency gains made possible relative to currently used state of the art approaches. (Abstract shortened by UMI.)...