The Transportation Demand Analysis Model Research And Its Application For The Third Party Logistics Enterprise

Abstract:Logistics market demand forecasting is the important premise of transportation planning and management decision for logistics enterprises. The existing relative literatures are averagely based on a macro or meso point of view and mainly focus on the research on collective logistics demand, but in terms of the micro level, the literatures on logistics enterprise’business forecasting are relatively small. At present, the third party logistics industry is developing rapidly and its competition has become increasingly fierce, in china. To study the main business of third party logistics enterprise of transport type, namely the problem of transportation market demand forecasting concerns the scientization of business decision.According to the double characteristics of seasonal fluctuation and tendency which the volume of the third party logistics enterprise’s transportation business presents, this thesis uses additive and multiplicative model theory of seasonal time series to analyze the transportation demand respectively. Using GM(1,1) model to analyze the trend component of the time series of transportation demand, then, using periodic extensional model to judge the seasonal fluctuation and mark off the optimal periodic boundaries, on the basis of them, this thesis addresses GM(1,1)-periodic extensional model with residual modification based on the theory of additive model. Using seasonal and unseasonal difference operator for smooth handling, the SARIMA model is established for the time series of transportation demand by multiplicative model theory, and the precision of the SARIMA model is verified using the auto-correlation and partial auto-correlation coefficient function figure of the residuals. From an empirical perspective, the fitting models are established in three samples selected based on the background of X logistics company, and the contrastive analysis of the fitting results leads to the conclusion that goodness-of-fit of GM(1,1)-periodic extensional model is better than that of SARIMA model. Furthermore, this thesis uses above-mentioned fitting models to forecast the next three month’s volume of the transportation business and the analysis indicates that predictive value of SARIMA model is low compared with the measured value, which that of GM(1,1)-periodic extensional model is more approach.In terms of the influence factors of transportation demand, this thesis views the third party logistics enterprises’internal factors as a part of the whole influence factor system, and use the method of grey relation entropy analysis to quantitatively study the interdependence between the influence factors and transportation demand. In addition, this thesis takes X logistics company for example and analyzes the grey relation entropy coefficient between the influence factors and its volume of the transportation business. The results prove that logistics service node number has the strongest relation with the volume of transportation business.