An Improved Exponential Smoothing Model on Rental Trends Prediction of Public Bicycle Stations

摘要

We have analyzed the rental trends of public bicycle stations from the real data of PBSS (public bicycle-sharing system) and summarized 3 patterns in the rental trends of public bicycle stations using k-means. The proposed DDIES method in this paper, which has high accuracy in the prediction, is a model based double exponential smoothing model and difference-index smoothing model. The results have shown that our method has higher precision in the prediction of public bicycle.