Research On Dynamic Optimal Configuration Methods Of Cluster Supply Chain Based On Augmented Lagrangian Coordination

A new form of network organization is formed by the coupling of industrial cluster and supply chains. It has the characteristics of regional proximity, dynamic and so on. Industrial cluster is a collection of enterprises and institutions which are connected with each other in a particular field.In our country, medium and small-sized enterprises (SMEs) not only have a huge quantity, but also are the main body of the national economy.However, because the majority of SMEs, backward production technology, production scale is small, the shortage of funds, the ability of acquiring information is very limited, it is difficult that they rely on their own ability to adapt to the ever-changing market environment. In industrial clusters, many SEMs of some specific industires, which are interrelated and relatively concentrated in geographical location, are able to obtain cost and time advangtages through mutal cooperation, cooperation with institutions, efficient resource sharing and cooperation and division of labor, thus forming the regional competitiveness in order to get a better industrial brand and market opportunities. Therefore, the research on the operation of the industrial cluster, especially the research on the operation of cluster supply chain (CSC) is becoming an important trend.Under the above background, this dissertation focuses on process cluster supply chain and assembly cluster supply chain in our country.In view of the very complex coupling relationships between participants of two types of CSC, the highly dynamic supply chain structure changes with the change of customer’s demand, especially, participants’privacy protection and independent decision-making operation in the pursuit of self profit, it is very difficult to sovle such configuration problems of two types of CSC by using traditional methods. The thorough research and analysis of two types of CSC configuation is done using augmented Lagrangian coordination (ALC) method with open structure and support independent decision-making in the field of complex system optimization, and considering the order production collaboration between single supply chains, crosss-chains private suppliers sourcing, independent suppliers sourcing.The main research work in this dissertation is as follows:(1) ALC, which supports open architecture, is applied into the field of supply chain management. A complete and feasible solution procedure for cluster supply chain is proposed.It includes target AIO model establishment, target decomposition model, solving sub-problems coordination mode, the coupling establishment of consistency constraints between sub-problems, the sub-problem local target model establishment, sub-problems solving algorithm selection and so on.(2) For the optimal configuraton problem of process CSC, for single supply chain order production cooperation, process manufacturers of single supply chains outsourcing process to cross-chain private suppliers or to indepentdent suppliers,three sub problems, which are the optimal configuration problems considering independent suppliers sourcing (ISS) process CSC, cross-chain suppliers sourcing (CSS) process CSC, and ISS&CSS process CSC, are studied respectively. By analyzing the operation model of each problem, formulating the relevant assumptions and making configuration strategies such as alliance strategy, production outsourcing strategies and outsourcing strategy, the distributed ALC configuration model of each problem is established. The ALC method is applied to each model to solve the problem, and the results are analyzed, and a series of important management implications are obtained.(3) For the optimal configuraton problem of assembly CSC, ccording to the different operation strategies of assembled cluster supply chain operation, three sub problems, which are the optimal configuration problems considering different components of product assembly independent suppliers sourcing (ISS) assembly CSC, different components of product assembly cross-chain suppliers sourcing (CSS) assembly CSC, and different components of product assembly ISS&CSS assembly CSC, are studied respectively. By analyzing the operation model of each problem model, formulating the assumptions and configuration strategies, the configuration model of each problem is established. At the same time, the distributed ALC model of each configuration model is set up through the decision-making power division. ALC two coordination methods are used to solve each model, and the solving results are compared and analyzed at last.(4) Due to the difference between centralized and distributed ALC method to solve the optimization configuration model of CSC in speed, computational parallelism and the accuracy of results, a dominant ALC (DALC) method is proposed according to the coordination process characteristics and model formutations of two coordination ALC methods.At the same time, the introduction mechanism, application conditions and selection principle of the main problem in the DALC method are pointed out, and the rigorous mathematical equivalence proof and convergence analysis of the DALC model are given.Finally, the proposed DALC method is applied to the typical instances and the configuration model based on ISS&CSS-A CSC respectively. The simulation results show that the proposed method is effective and accurate.