A Study Of The Distribution Network Optimization In Supply Chain Based On Genetic Algorithm

With the incremental development of world economic integration and market globalization, supply chain management (SCM) has been the important management mode for modern enterprises. In supply chain, distribution system is the most important linkage, which is not only the sign to measure the level of supply chain management but also the most key factors to determine the competition capability of supply chain. As a result, under the environment of supply chain, logically designing and optimizing a good distribution network to reinforce the management of distribution can enable the enterprises to satisfy the customer's requirements better and eventually to win in the competition.In the traditional distribution network optimized model, such key factors in distribution system as demand volume, transportation time, etc. are supposed as the known constants, and from the integrative perspective of supply chain, synthesized analysis is seldom adopted to study the diversified factors of the gross cost of the distribution network. However, in practice, some factors which influence the optimization of the distribution network are uncertain and connect with each other. Therefore, under the condition of uncertainty, this thesis synthetically applies the inventory cost, stockout cost, transportation cost, device cost, service level and other relevant factors to the process of location, compositively optimizes the device location, inventory control and transportation management, and accordingly balances the various costs in the distribution network to achieve overall optimization of the distribution network.This thesis elaborates the optimization strategy and the diversified cost factors in the distribution network, establishes a stochastic demand optimization model of the distribution network on the basis of inventory control and extends the model as well. We adopt modified genetic algorithm to solve the model, and set an effective coding mode to decrease the memory space of the chromosome. We adopt improved adaptive crossover probability and mutation probability, intelligently design the genetic arithmetic operators, so as to effectively avoid premature algorithm and successfully advance the process efficiency. Eventually, an example is given to confirm the correctness of the optimization model and the feasibility and the effectiveness of the algorithm. This thesis designs the optimization model of distribution network, which helps to promote the harmoniously management of the production-distribution process, effectively reduces the production cost and distribution cost, and also offers determinate decision support at the layer of stratagem and tactics for the enterprises.