Rasearch Of Logistics Distribution System Interagrated Modelling And Optimizating For Chain Enterprise

Logistics distribution is the crucial foundation and support for chain enterprise. In a certain sense, the survival and development of enterprise is close related with the operational status of logistics distribution system. Through a series of high efficient distribution operations, we can realize the reduction of logistics costs, the improvement of operational efficiency of logistics system and service level perceived by customers. So the scientific and reasonable integrated planning of logistics distribution system has wide application value and practical significance.The main component of logistics distribution system planning of chain enterprise includes such four decisions as customer service target, strategic facility location, tactical inventory control policy and operational vehicle routing. There exists widely and close trade-off relation among the above decisions-making factors. In this paper, based on the integrated view and the problems existing in the constructing and operating of the logistics distribution system, different integrated models in each decision level and the corresponding solution algorithms are systemly researched.The main research contents in this dissertation may be summarized as follows:1. Chain enterprise logistics distribution system integrated location-inventory problem researchThe distribution system design of three-echelon chain enterprise is researched, in which customer has normal distribution demand. The distribution system consists of one center warehouse, multiple candidate distribution centers and multiple candidate retailer stores.Firstly, the market share captured by the retailer outlet can be derived from a discrete competitive facility location model. Secondly, multi-echelon inventory cost is analyzed through power-of-two inventory control policy in detail. Considering the trade-off among costs, an integrated logisitics distribution system location-inventory model with the objective of maximum enterprise profit is presented. The competitive facility location, distribution network design and multi-echelon inventory control are simultaneously considered in the integetaed model. A lagrangian relaxation heuristic algorithm based on subgradient optimization is applicated to solving the practical scale problems. The solution algorithms solving sub-problems are also listed. Finally, the effective and efficient of the model and the algorithm are validated through the simulation of cases. The parameter effects on the solutions are also given by sensitivity analyse.2. Chain enterprise logistics distribution system multi-period integrated location-inventory model research considering elastic customer demand and congestion effectThe multi-period distribution system design of chain enterprise in dynamic external competitive environment and market demand is researched. Taking account of the elastic customer demand and customer's probabilistic facility choice behavior, the leader-follower relation between the designer of the logisitics distribution system and customer can be described as a bi-level integetaed location-inventory model. In the upper-level problem, multi-period facility location model with modular capacity expandance, which involves the decisions on the opening/closing and the capacity configuration of distribution centers and retailer stores, is presented. In the lower-level problem, the customer's distribution mode in the background of elastic demand and congestion effect is described by a stochastic user equilibrium assignment model with elastic customer demand. Since the bi-level integrated model is hard to solved, we first proposed an algorithm to solving the lower-level problem, and embedded it in genetic algorithm to construct the whole heuristic algorithm. Finally, a numerical example is used to illustrate the effective and efficient of the model and algorithm.3. Chain enterprise logistics distribution system long-term integrated inventory-routing model researchDistribution plan arrangement in long-term or medium-term is researched, in which the customer demand is steady over the plan horizon. The distribution system consists of one center warehouse, multiple distribution centers and multiple retailer stores which are distributed in a given region. Based on the fixed partition policy, power-of-two multi-echelon inventory control policy and vehicle muti-routing notion, the integrated long-term inventory-routing model in formulated, in which includes the decisions of distribution area partition, the distribution cycle of retailer stores, the order cycle of distribution centers, the number of vehicle and vehicle routing. A Tabu search algorithm finding the optimal partition of distribution area and the solution algorithm solving the distribution cycle and order cycle are proposed. The lower bound of the objective function value in any feasible policy is detailed analyzed. Finally, a numerical example is used to evaluate and analyzed the model and the algorithm.4. Chain enterprise logistics distribution system integrated short-term inventory-routing model research When the decision on the order cycles of center warehouse and distribution centers are already maked and the demand of retailer stores periodically change, the integrated short-term inventory-routing model with the objective function of minimal operational cost can be presented. An effective two-phase heuristic algorithm, which can solves the practical scale problem, is proposed. Set partition model, distribution plan optimization model and their solution algorithms are estabilished in two phases of heuristic algorithm respectively. Finally, a numerical example is used to evaluate and analyze the model and the algorithm.5. Chain enterprise logistics distribution system integrated location-inventory-routing model researchWhen vehicle can visits multiple retailer stores by non-full load mode and the facility location cost and vehicle routing cost are comparable in a certain sense, the distribution system design of three-echelon chain enterprise is researched, which consists of one center warehouse, multiple distribution centers and multiple retailer stores. The integrated location-inventory-routing model of distribution system is established in view of integrated decisions of strategic level, tactical level and operational level. The decisions in the integrated model involve the location decision of distribution centers, inventory control policy and vehicle routing. An external lagrangian relaxation heuristic algorithm based on sub-gradient optimization is applicated to solving the practical scale problems. The solution algorithsm for solving the distribution network design with risk pooling effect problem and the minial spanning forest with capacity and degree constraints problem are also proposed. Finally, a numerical example is used to evaluate and analyzed the model and the algorithm.The research efforts presented above can provide more sufficient scientific evidences and decision supports for the logistic distribution system design of chain enterprise...