Study On The Location Of Return Distribution Nodes Of The Third Party Logistics Enterprises

This paper based on the the Changchun Technology Bureau research: , from the return transport's point of view, carrying out a large number of research of the location problem of the third party logistics company return distribution network building. We explorate the main factors which impact the company distribution network building deeply, analyse the feasibility of location of the return distribution network, and base on this analysis, use the self-organizing neural network to model the third party logistics enterprise distribution network; At the same time, we use the visual simulation software of Witness to simulate the model which we have built before, verify the reasonableness of the method.Along with the rapid growth of china social economy, China government encourages to develop modern logistics industry so as to induce the soaring cost of logistics activity. Thereby improve the competitive ability of our enterprise. The Eleventh Five years Programming point out clearly that logistics industry will be pivot industry to develop. In order to keep sustainable development of national economy, we must speed up the reasonable layout of third part logistics. Completing the link up of transport mode mutually, display combination efficiency and the overall superiority .As result develop an integrated transportation system which is fluent, highly effective, safe and rapid. The enterprise distribution network is composed of a number of abstract logistics network dots and the lines which connecting these dots. These networks should include suppliers, machining centers, distributors, retailers, customers, logistics centers and warehouses and all the depots which are included in the logistics activities out of enterprises. The lines between these dots express the feasible transport path, transport mode and the transformation between kinds of informations associated with logistics activities. The online logistics functions mainly achieve the transportation functions, and rests of logistics functional elements are completed in the network dots. Distribution network is the foundation of the entire enterprise network, its construction relates to the merits of the whole network, it's the most important part of the distribution network.In addition, as the improvement of people's material and cultural standard of living, efficient road distribution system has become the inevitable trend of the development of logistics enterprise, and the return transportation network is the most important element which could promote the development of vehicle distribution system. The majority of the researches of logistics network dots today are focus on the process of demand transportation, it's a analysis-location decision against single product and one-to-many form, but there are few studies have seen the serious shortage of the building of the logistics enterprise's return distribution network. The choice of return distribution network dots is a multi-factor decision against wide range of products, many-to-one form, and the level of business volume and transport economic situation. Therefore, to build a return distribution transport network is the top priority of development of logistics enterprises. The return distribution network is the key point of enhancing the efficiency of the third part logistics company's distribution system, and the choosen of the network dots directly impact the business and development of enterprise's distribution system. However, the majority of the researches agaist the third part logistics companies in our country is still very little, most of them are noly concern the location, the number, the paths and so on of the service logistics center, lack of the location and so on from the return vehicles'point of view. Therefore, based on the spirit of"perspective innovation", the research of enterprise return distribution network dots location has shown its theoretical value and practical significance.Return distribution network is one of most important subsystem of integrated transportation system. As the nodes of return distribution system, the return distribution network dots are the large-scale commodity collection and distribution center. It has the function of return transportation organization and management, transfer, storage, communication, production assistant service, which play the decisive role in linking up each part of supply chain management system together, optimizing system efficiency and service level and improving the benefit of logistics industry. The location research of the return distribution network dots is belongs to the corporate strategic level decision which is one of the important contents of the design of the return distribution network, and it also the basic condition of network operators. In this paper, combined the actual of third part logistics companies, the research agaist the location of return distribution network and the theory and methods of the simulation has been deepened. Through the deep survey to third part logistics companies, we collect and collate lots of data of corporate operational management, summarize and sum up the basic informations, and design the methodology of the enterprise distribution network that combined with self-organizing artificial neural network for enterprises to explore new theoretical approaches.The dissertation has seven chapters. The main content as following:Chapter one, introducing research purpose and the background about this article, and then simply introduces the current research status, finally state basic method and core content. After introducing the necessity of building a return distribution network, raising theoretical basis as well as weiting ideas combined with research status at home and abroad. Chapter two, briefly state the theory of status and problem of third part logistics enterprises, proposed the necessety for building enterprises'return distribution network, and analyze the self-organizing feature, including opening, non-balanced, non-linear as well as fluctuation with effectiveness characteristics.Chapter three, making the research of business traffic forecast on the basis of the proposing of the thing that the number of service traffic is directly affect the way to distribute return tranpotatiom. After analyze forecast methods, forecasting the number of service traffic by gray forecast method combined with Markov theory; and, take the actual enterprise data for example, forecast the main number of service traffic of the second year, and providing data basis for the research of enterprises return distribution network dots location.Chapter four, discussing the necessity of the usefulness of the self-organizing neural network method, analyzing its advantage and characteristics, and researching the specific method of the usefulness of SOM in the building of the third part logistics enterprise return distribution network. At the end, proposing the mathematical model.Chapter five, taking a well-known vehicle logistics company for example, using the actual data, taking the predicted data mentioned in the chapter three as sample vector, using the method of SOM, advancing the normalize of sample data, and using"VC++"iterate. Find the most suitable model for location of the enterprise return distribution network.Chapter six, introducing the data simulation software of Witness, advancing simulation modeling through the method establishing in chapter four and chapter five, and verifying the feasibility of the method of test the network by profit data.Chapter seven, summary and outlook, conclusion and expectation.Summarize mainly research work of this dissertation and presents questions that need to be further researched in the future as well.There is something new in this dissertation as the following:1 Take the third party logistics enterprise point of view, analyze the principle and methods of location of enterprise distribution network, helping to drop logistics cost and enhance the dfficiency of company.2 Use the combination forecasting method, predicting the number of service traffic through gray prediction method. This method is making up the disadvantages of traditional forecasting methods.3 Propose the SOM method for researching the location of enterprise return distribution network by the enterprise's data and appropriate sample vector for the first time.4 Use the visualize simulation software of Witness to model and analyze the return distribution network, verity the feasibility of the method.