Research On Logistic Allocation-Send Optimization Based On Genetic Algorithm

With the quick development of information technology, electronic commerce is gaining more and more attention worldwide and has become the new increasing point of economy in the 21 century. Along with the further expansion and application of the electronic commerce, the importance and influence of logistics become more and more obvious. The long time and high cost of logistic allocation have become the "bottle-neck" of the expansion of the electronic commerce in our country. The research target of the paper is to explore the theories and methods to optimize logistics, lower the logistic cost, optimize the social stock installation, improve the service quantity and sequentially increase economic and social benefit of the business enterprise and guarantee the healthy development of logistic system with appropriate scale and pace.The logistic allocation is a complicated systems engineering. Many models and algorithms established to optimize problems are complex and most have the nature of NP. Firstly, the dissertation introduced the background and meaning of selecting the topic, elaborated the relationship between electronic commerce and logistics and proposed the conception of building the logistics allocation system of electronic commerce in our country; Afterwards, it theoretically analyzed the two typical problems in the optimization of logistic allocation (the selection of sites and optimization of vehicle routes), established the mathematical model under certain assumption prerequisite and briefly introduced the tool (genetic algorithm) with whole random search capability to solve problems; Finally, it made a series of improvements to the fundamental genetic algorithm in operating aspects of selection, crossover and mutation, applied it to the problem of optimization of logistic allocation and carried out beneficial experiment and analysis. In order to check the validity of the proposed improved genetic algorithm, the paper worked out the source programs of CSGA.C and VRPGA.C enclosed in the annex and made substantial analysis of practical problems by the two programs.Through the result of substantial analysis, it can be seen that the convergence rate of the whole algorithm can be greatly improved by adopting the improved genetic algorithm and thereby the article supplied an effective method to solve the relatedproblems of optimization.