Abstract
We carried out a prospective study on train stop position errors at stations. This study aims to clarify the factors of occurrence of delays in braking manipulations. Driving performance data over a period of one month and characteristics of 81 railway stations were analyzed. After that we conducted follow-up survey for two years to learn whether delays in braking manipulations occurred or not. In the results of the follow-up survey, the number of stations with occurrence of delays in braking manipulations (stations with errors) was 14, the number of stations without occurrence of delays in braking manipulations (stations without errors) was 67. t-test and logistic regression analysis were conducted to extract factors causing delays in brake manipulations. The objective variable was the occurrence of delays in braking manipulations. The explanatory variables included driving performance data, such as the mean of train velocity at 200 m before stopping, and characteristics of stations, including slopes. The results of the t-test revealed that the mean of train velocity at 200 m before stopping at the stations with errors was significantly faster than that of the stations without errors. Similarly, the number of velocity patterns by signal aspects at the stations with errors was significantly more than that of the stations without errors. The results of logistic regression analysis revealed that delays in braking manipulations were significantly associated with the mean of train velocity at 200 m before stopping, and the number of velocity patterns by signal aspects.
https://ror.org/036vt3264 
