応荷重システムデータからの列車の停車時分の推定

抄録

Train delays during morning rush hours are a serious issue in the railways in metropolitan areas. In order to solve this problem，it is essential to set appropriate dwell times at each station. We have analyzed dwell times utilizing data obtained from load compensating devices of train-sets. We have established a method to estimate dwell times at each station by applying machine learning algorithm to this data. While further improvement in the estimation accuracy is required，the proposed algorithm is promising and expected to contribute to the proper setting of dwell times，helping to prevent train delays during morning rush hours.