Bike sharing has become popular in London, as it has the advantages of easing traffic pressure and connecting to the public transit. However, the relationship between the bike sharing and public transit systems has not yet been fully studied. This paper aims to measure the impacts of abnormal tube service status on the London Cycle Hire (LCH) usage via using 1 million bicycle travel records data from April 20 to May 22, 2019. Over the same period time, London tube service status data were also continuously collected from Transport for London (TFL). Covariates such as tube station volume, socio-economic information and road network characteristics are also included in the data set. In order to eliminate the error caused by the randomness, LCH stations within a certain distance of subway stations were aggregated. A buffer-area based regression was conducted to find the effect distance of the abnormal tube status. Our results suggest that the effect area of severe delay and part closure is within 400m from tube station. Regarding the effects of the abnormal tube service status on the public bike usage, the results shows that the part closure of tube stations decreases the number of LCH bike departures and arrivals, while the severe delay has the opposite impacts. Furthermore, the reduced bicycle travels are mainly concentrated on travels which travel time is less than 1200s when part closure occurs. In terms of the effects of abnormal tube status on duration, a significant increase is observed in evening peak time when tube station is in part closure status, while the effects of the other situations are not significant. In addition, the effects of abnormal tube service status also vary across different periods of the day. Keywords: bike sharing; abnormal tube service status; transaction records data

CICTP