Research On Logistics Demand Forecasting Of Beijing

As the basis of city logistics system planning, logistics demand forecasting is the important premise to make logistics resources effective configuration and build efficient logistics system. In this context, analysis of various factors such as social and economic activities to meet the demand for logistics supply, and the effective logistics demand forecasting can help to grasp the regional logistics demand intensity and realize the regional logistics service demand and supply of relative balance. It also has important theoretical and practical significance to improve the quality of regional logistics planning and regional logistics operation efficiency.Based on the regional logistics demand forecasting research status at home and abroad and the related theoretical knowledge as the foundation, combined with the characteristics of logistics demand, for Beijing logistics demand forecasting and the related application mainly done the following several aspects of the research:(1) Combined with the influencing factors of logistics demand, this part established the index system of Beijing logistics demand forecasting. Using grey relational model to the quantitative analysis of correlation index system, select the relevant factors having a strong correlation for the establishment of logistics demand forecasting model.(2) In view of the characteristics of complexity, nonlinearity, randomness of the city logistics system, combined with the present situation of logistics statistical data, the grey system model and neural network model used for problem solving. On this basis, the gray neural network model is easy to fall into local optimum and slow convergence speed, using particle swarm optimization algorithm to optimize the initial parameters for grey neural network. Based on particle swarm optimization algorithm of grey neural network forecast model, it can overcome the uncertainty of grey neural network structure selection and improve the soundness and precision of the model.(3) Based on regional logistics demand forecasting, the method to calculate the formula of the land use scale of parameters has been revised. The calculation formula of parameter method is improved on the basis of comparing advantages and the disadvantages of different methods. Accordingly, a more general method of calculating the scale of the logistics park is proposed by introducing other land area ratio coefficient, reserved land coefficient, comprehensive correction coefficient.(4) Empirical study on logistics demand forecasting in Beijing shows that based on particle swarm optimization algorithm of grey neural network prediction model is displayed on the precision than single prediction model and grey neural network model.The prediction effect is better so as to verify the height of the model fitting and stability.