The Research On The Method Of City Distribution Center Selection And Application

With the continuous development of the city, the human way of life has also undergone a great change. People rely more on the transfer and distribution of materials. If the distribution center layout and location of the city logistics distribution center is reasonable, it will has great influence on the distribution and the full play its function of the distribution center. Excellent logistics distribution center location can accelerate the circulation efficiency of products, bringing convenience to both sides of supply and demand, promoting the construction of the logistics socialization, conforming to the development planning of logistics, reducing environmental pollution and reduce energy waste. It is conducive to the scientific and rational allocation of materials, and it will effectively reduce logistics costs, inventory costs and purchase and sale costs, bringing the smooth flow of funds, improving the utilization of funds. It is helpful for people to study the new technology and improve the working efficiency of workers. It is conducive to the overall economic development of the city, and it will promote the relevant industry field, increasing the number of projects, increasing the number of jobs, providing the market for many of the idle materials, creating social benefits. Therefore, the selection of city logistics distribution center has a very important significance.In view of this, after a careful study of the traditional site selection model and a lot of references to the popular intelligent algorithms in recent years, this paper proposes a new location model based on geographical grid and extreme learning machine. In case of simulation, two difficulties of affecting factors are the route problem and the determination of coverage area. This paper will try to solve these problems by using the method of geographic grid. Under the programming environment of MATLAB software, using distribution center’s information of Beijing and its periphery build the site selection model, using annual turnover of the site location as the final decision factor. The results are compared with the results of radial basis function neural network and generalized regression neural network, showing the superiority of extreme learning machine. At the same time, through this paper, it is expected that this method can develop a new method for location selection problem.