需求响应式公交到站时间预测研究

Research on Time-of-Arrival Prediction of Demand-responsive Bus

针对城市绵延情况下城市新区公交低密度地区的交通出行问题,开展对实时响应模式下的需求响应式公交的到站时间预测研究。以时间序列法和BP神经网络模型为基础,提出综合预测模型,并用最小二乘法对模型求解。证实了在城市新区开通需求响应式公交服务模式的可行性,以及为居民提供出行便捷、合理约束到站时间的公共交通服务的能力,为需求响应式公交服务模式的推行起到积极作用。

In order to solve the problem of bus arrival time in low-density areas of new urban districts,this paper carries out research on the arrival time prediction of demand-responsive bus under real-time response mode.In the demand-responsive bus-to-station forecasting stage,this paper proposes a comprehensive forecasting model based on two classical forecasting models,the time series method and the BP neural network model,and uses the least squares method to solve the model.The final results of this paper confirm the feasibility of opening a demand-responsive bus service model in the new urban area,and provide residents with the ability to travel public transportation services that are convenient and reasonably constrained for arrival time.It has played an active role in implementing the demand-responsive bus service model.