Comprehensive Application Research On Methods Used In Dealing With Traffic Flow Information In Logistics Intelligent Dispatching System

With the fast development of domestic economy and process of civilization, traffic jam and the increasing traffic accidents have inhibited the development of economy. Intelligent logistics dispatching system which is the core component of intelligent logistics is the best way to solve the transportation problems. With the progress of modern science and the development of traffic, in allusion to the problems that exist in route network, it's necessary to gain traffic flow information of most intersections to realize the management of route network. Usually the traffic flow information is known according to detectors of intersections, it's difficult to get the information from non-detector intersecton due to scarcity of detectors.Due to the issues above, this paper is finished in combination with the project of National Science Foundation of China(NSFC)"Research on port logistics intelligent dispatching system model and key technology based on the information of route network". In this paper, traffic flow information is divided into the normal situation and the special situation. Based on SPSS 13.0 and statistics methods, it's possible to predict the traffic flow information of non-detector intersections according to the co-relationship between the intersections in different situations.This paper consists of five chapters. In the first chapter, the background and purpose, current research, the basic idea and content of this paper are provided. In the second chapter, the method of obtaining and dealing with traffic flow information in logistics intelligent dispatching system is introduced. Chapter 3 introduce clustering analysis and stepwise regression. The clustering analysis method can amalgamate different kinds of variables into one kind based on the co-relationship between these different kinds of variables. The stepwise regression method is a statistic method that can select several important variables to construct the regression equation. Chapter 4 introduce main-components analysis method. In these methods, The method of main-component analysis is a statistic method that can convert a lot of quotas to few synthetical quotas and these quotas have less co-relationship. Chapter 5 is the summary. The main contribution of this paper is summarized. The shortcoming and expectation of this paper appear in this chapter.