基于ARIMA模型的城市公共自行车需求量短期预测方法研究

预测在城市公共自行车的研究中占重要地位，对站点未来需求量进行分析和预测，可为管理者提前分配自行车和用户合理制定出行方案提供依据. 本文采用自回归求积移动平均（ARIMA）模型，对公共自行车高峰时段的需求量时间序列进行拟合和预测，并与基线法（Baseline）预测误差比较，结果显示对于不同站点类型的预测，此模型的预测值与实际值的平均相对误差均低于Baseline预测方法. ARIMA模型的预测精度相对较高，且预测结果可信，可为城市公共自行车管理和使用提供预测的理论与方法.

Research on Short-Term Prediction Method of Demand Numberin Urban Public Bicycle Based on the ARIMA Model

Prediction occupies an important position in study of urban public bicycle. Analyzing and predicting the demand numbers at every station in future can provide a basis，which managers allocate bicycles and the users make travel plan in advance. It is necessary to use the Autoregressive Integrated Moving Average（ARIMA）model，which models the demand number time series of public bicycle during peak hours of the week. Comparing with prediction error of the Baseline method，the results show that the average relative error of the value of the prediction and the actual are both lower than the Baseline prediction method for different stations. The prediction precision of the ARIMA model is relatively high，and the prediction result is credible. It provides theory and method of the prediction for management and use of the urban public bicycle.