Research on forecasting model and its algorithm for urban public transit passenger volume

In order to improve the accuracy of the forecasting of public transit passenger volume, the model of random gray forecasting and recurrent neural network based on ant colony algorithm have been built respectively according to the characteristics of randomness and nonlinear of passenger volume forecasting of public transit by using the uncertainty that random gray variable describes forecasting system, and on that basis, a model of passenger volume forecasting of public transit and corresponding algorithm have also been put forward based on random gray ant colony neural network. Finally, a case study has been carried out to take Tongling city as an example in order to testify validity, objectivity and applicability of this model has been made. The results show that the forecasting accuracy of recurrent neural network based on random gray ant colony algorithm is not only greater than any other single forecasting models, but also superior to other traditional combinational forecasting models. Therefore, it can well reflect the laws of development of things and can be of practical and effective value in actual use of construction project. © 2011 by Binary Information Press.