Study Of Railway Passenger Volume Forecast Method Based On Optimization Of Gray Markov Chain Model

Along with our country’s economy continued rapid growth, China’s railway enterprise also obtained the leap-forward development.There are still under constructions and the railway construction project in succession. How to determine the construction scale of the railway project, making its can effective evacuation of passengers and cause no waste of capacity, this is the focus of the railway construction project prophase work.Traffic forecast is one of the core contents of railway project construction preparation, the choice of prediction method is the key to decide the prediction level. It necessary to find a scientific and effective method for railway passenger volume forecasting.This paper first briefly introduces the development of Chinese railway history and the current development status, and the development trend in the future. Then, the analysis shows that it is importance for railway passenger traffic forecasting as for the meaning of the whole railway system, Then it simply elaborate the main forecast method of traffic volume forecasting, and make a review and summary for various methods.Although there are many railway passenger volume forecast method, but most of the methods have limitations. So it is necessary to find a scientific and relatively simple method for railway passenger volume prediction.By analysis of the characteristics of railway passenger traffic volume and its data,founding the railway system is affected by many factors, such as natural and social conditions, and greater volatility of the data. According to these characteristics, the paper chose to use a relatively simple calculation model gray Markov chain model with a combination of railway passenger volume to forecast.But now using gray markov chain model for railway passenger volume forecast, Without considering the shortcomings of the models.In this paper, Putting forward the traditional gray markov chain model two inadequate:First,when gray model to fit index sequences having deviation;Second,the markov chain model in the gray interval during bleaching, having defects on the choosing of an albino coefficient.In this article, the above two inadequate, first by establishing unbiased gray model to correct the inherent bias of traditional gray model, then use particle swarm algorithm to calculate the optimal coefficients albino. The formation of a higher predictive accuracy of the model, a new particle swarm--- unbiased gray Markov chain model.Calculations show that PSO unbiased gray Markov chain model compared with the traditional gray Markov chain model, the prediction accuracy is improved.