System Design And Implementation Under Logistics Network Optimization For Electric Power Enterprises

Power industry deals with the basic energy of electricity, which affects almostevery aspect of people’s life. With the implement of power system reform andwideness of power market openness, the costs and core competitiveness areincreasingly influenced by power material management. Despite the similarity withmaterial management in other companies, power industry has more requirements suchas material choice and matching. However, in Chinese power material managementpractices, problems including the lack of regulations, geographically dispersedwarehouses and undeveloped technologies actually exist. Therefore, how to establishan effective and stable logistics network for power enterprises to support smoothmaterials flow, has become an urgent problem for China power industry,To solve this issue, this dissertation mainly focuses on three aspects of research.First, researches in the fields of power network optimization, including modeling,algorithm and case study are analyzed, With the combination of the features inChinese power industry, the logistics network topology structures, delivery methodsand material classification modes in power enterprises are analyzed and developed,thus making it more suitable for power material management practices. According tothe development features and future trends, the comparison research of materialclassifications in power enterprises and other enterprises are conducted, and the newclassification based on warehousing and delivering is developed, by introducing morepractical factors in power material management, higher control levels can be achieved.Secondly, we build an optimized network model to solve the irrationality ofwarehouse network by analysis and selection. Combining the logistics demand, thematerial classification method, and the logistics cost, this new model based on theenterprise business is different from gravity and radius method that the generalmodels used. We also discuss the penalty costs, construction costs and the operationcosts in this paper. After the completion of the model in quantitative analysis, the optimized scheme is selected by analytic hierarchy process. Referencing foreignelectric power enterprise logistics network construction, the power enterprise logisticsnetwork design is furthermore optimized. Thirdly, we choose the SA (SimulatedAnnealing algorithm) based on a mixed MCPSO (Multi-swarm Cooperative ParticleSwarm Optimizer) to solve the new model. To control the convergence, SA algorithmcombine with MCPSO. Then we design the optimization decision support system forelectric enterprise logistics network, which is expected to improve the current powerlogistics network design.This paper starts with the analysis of research background and significance,introducing the importance and necessity of power enterprise logistics networkoptimization. We find the theories and cases worth references for our research field,and thus settle the research contents and technical route. And then, we combine thefeatures of power enterprise logistics, as well as logistics network topologicalstructure and material classification management, and establish the multi-objectivelogistics network optimization model based on costs and service levels, and solve themodel with the simulated annealing algorithm based on a mixed MCPSO. Based onthis model, a logistics network optimization decision system is developed especiallyfor power enterprises, with the overall structure, use case support as well as systemimplement. The developed system proves to be satisfactory through the back-endqualitative analysis. Finally, the research process and results are reviewed andsummarized, and we make a vision of the direction of future research.