Research On Theory And Methods Of Forecasting & Planning Based On L-OD In Modern Logistics

With the development of globalization and information technology, modern logistics is developing to a very different extent not only in developed countries, such as America, Japan and Europe etc. but also in china. The government, society and experts in china have created heightened awareness of the special area. Modem logistics has been becoming main industry, even supporting industry in some areas and cities. In recent ten years, the investment to logistical foundational facilities in China has grown greatly, the total investment and new investment of logistics fixed assets belonging to nation increase by degrees of 24.3% and 27% per year.However, the industry needs to set up a set of theory system and solutions which solve the problems about how to confirm logistics pattern, how to hold the development rule of logistics demand, and logistics facilities layout. Nowadays, many provinces and cities in China are making logistics plan, and it is exigent for them to study the theory and methods, and so do the enterprises which is developing the third party logistics. Experts overseas as well as home have been studying the logistical demand forecast and planning, especially several models and solutions have been got on demand analysis. But unfortunately, there are disparity between the scientific system and operable approaches. On the basic of analyzing ready-made logistical theory, this paper set up a forecast and planning system based on L-OD, develop a corresponding forecast system, put forward the basic principles and major methods of modern logistics plan and put them into practice with the beginning of studying logistical demand feature, affected factors, forecasts. Then, the application improve the theory is effective and reliable. The paper is divided into three parts including six chapters.Part 1 we introduce connotation, functions and characteristics of modern logistics, form and evolution of logistics industry and knowledge, the significance and characteristics of logistics plan, the functions of demand forecast. Then after looking up all the relative fruits, I analyze their features and problems, this paper refer to logistical object main part, object characteristics, contents and targets. I discuss five stages underlie logistics evolution according to logistical study area, study content, study depth, application and socialrecognition. The five stages are conception formation, military logistics, logistical naissance, physical distribution, and modern logistics. In this paper, I portray the functions of modern logistics, the feature of successional logistics services, integrated supply chain process, sociable logistical activities, informational logistical management. The differences between modern and traditional logistics are pointed out. The meaning, characteristics and functions of modern logistics system plan and functions of logistics demand and forecast are discussed in the paper. Under consideration of the ready-made research wholly, we analyzed the pulse, characteristics and disadvantages of logistics demand forecast and planning. These mentioned previously assures the paper's direction and scientific emphases.Part 2 introduces connotation of logistics demand, defines demand function, supply function, discusses what affect the requirement and its characters, analyzes supply-demand balance trend. I define the following concepts: L-OD, logistics zone, L-OD table and logistical generation, distribution, logistical chain and assignment underlie aggregate analysis. Different models of different phase are presented and their respective implications on forecast based on affection mechanism are discussed. The first forecast stage models include: regress analysis, elasticity factor, growth factor, exponential smoothness; the second stage models include: mean growth factor, modified growth factor, single restricted gravity model, double restricted gravity model; the third stage models include: proportion model, probit model, multiple logarithm model; the forth stage models include: minimum path model, limited...