Optimize The Design, Supply Chain Based On The Atc

ATC (Analytical Target Cascading) has recently emerged as a new approach to solving hierarchical decentralized coordination problem. In ATC, each element in the hierarchy makes plan autonomously and father element coordinates the plans from its child elements to achieve target optimization. ATC is used to solve the coordination problem in supply chain in this dissertation.First, introduce the basic concept of ATC, design the coordination mechanism for ATC model, whose efficiency is demonstrated by the numerical studies in this dissertation..Secondly, use ATC to solve distributed production task allocation problem, ATC is compared with AAO (All At Once) method and the cooperative strategy for this problem proposed by Nigro et al. to test its computation speed and efficiency in solving this problem.Thirdly, introduce oddment tactic, does research on its impact on the benefit of supply chain and design a reasonable benefit allocation mechanism; prove the equivalence between backorder-optimal oddment allocation and inventory-optimal one when inventory transfer among retailers is permitted or not; use ATC to solve the application problem of oddment tactic in distributed scenario.Fourthly, present a method for calculating the global optimal solution to (R,Q) ordering policy with BOM (Bill of Material) and random demand in centralized scenario; use ATC to coordinate the BOM based production plans of production entities in distributed scenario.Fifthly, build ATC model for a distributed production-distribution problem, design a coordination mechanism for the model.Finally, a distributed production planning and supplier selection problem is formulated as an ATC model, in which there are sub-models that are responsible for negotiating supply plans with suppliers and selecting suppliers according to production plan.The numerical studies in this dissertation shows that ATC is efficient in solving the distributed decision making problem. Nevertheless, there are still some problems in this dissertation to be solved in the future study.