基于SARIMA模型的上海市中心城区共享单车需求预测

Demand Prediction of Bike-sharing in Urban Center of Shanghai Based on SARIMA Model

无桩式共享单车的出现与推广在减少碳排放的同时,带来了道路拥堵问题。如何高效、准确地进行交通流量预测已经成为人们关注的热点。利用2016年8月上海市中心城区摩拜共享单车数据,利用季节性差分自回归移动平均模型(Seasonal Autoregressive Integrated Moving Average Model,SARIMA模型)进行模拟和预测,再通过折线图的方式揭示共享单车需求量与时间之间的变化关系。研究发现,SARIMA(0,1,3)×(0,1,0)_(84)模型能够有效预测上海市中心城区共享单车的需求量。预测交通流量,可以缓解城市主干道的拥堵状况,提高市民的生活质量。同时,预测通勤需求可以平衡供需关系,为运营企业和用户提供更高效的服务,为政府规划提供决策依据。

The emergence and promotion of dockless bike-sharing has also brought about road congestion problems while reducing carbon emissions.How to efficiently and accurately predict traffic flow has become a hot topic of concern.Based on the data of Mobike shared bicycle in the central city of Shanghai in August 2016,Seasonal Autoregressive Integrated Moving Average model is used to simulate and forecast,and then the changing relationship between the demand of shared bicycle and time is revealed by line graph.It is found that the SARIMA(0,1,3)×(0,1,0)_(84) model can effectively predict the demand for shared bicycles in the central city of Shanghai.Predicting the traffic flow can alleviate the congestion of the city’s main roads and improve the quality of life of citizens.At the same time,predicting commuting demand can balance the supply and demand relationship,provide more efficient services for operating companies and users,and provide a decision-making basis for government planning.