For city bus users, in-vehicle crowding is a criterion to assess the quality of the service such as comfort and convenience. City bus managers and operators should consider in-vehicle crowding as a key quantitative measure when considering the service...
For city bus users, in-vehicle crowding is a criterion to assess the quality of the service such as comfort and convenience. City bus managers and operators should consider in-vehicle crowding as a key quantitative measure when considering the service quality improvement. However, there is no quantitative indicators available for city bus as existing studies focused on Stated-Preference(SP) methodology which is more applicable to subway crowdedness quantification.
According to the Smart-Card(Transportation Card) of Seoul in 2019, 4,541,000 traffic per day occured, of which 512,000 traffic experienced in-vehicle crowding. In particular, 271,000 traffic is experiencing inconvenience due to the crowded standing space of the vehicle. An average crowding rate per bus when in a congested state is about 116%, similar to the average crowding rate(119%) of subway lines 1 to 8 as of 2019,which requires research on the value of crowding time of city buses.
This study estimated the Value of Crowding Time(VoCT) in vehicle by constructing a Logistic Regression Model, one of the Discrete Choice Model, using Smart-Card’s big data. The method of Smart-Card's big data analysis is developed and applied to calculate the travel time of each passerby in-vehicle movement conditions. As a result, travel conditions, such as sitting, standing, and crowded standing, were subdivided to estimate the VoCT. Through the SP method, the travel conditions for crowding avoidance of Seoul City-bus users and the amount of time to pay for each travel time and crowding duration were estimated and the impact of crowding in-vehicle on the total travel time was analyzed. In addition, the service quality in-vehicle was restructured and reflected in consideration of the changed size of the Seoul City-bus, the latest Korean body size, and the spare distance between people.
According to the study, Seoul City-bus users waiting or moving to avoiding crowding to no-crowding vehicle or route, even if the total travel time increases. Bus users were found willing to pay extra time to avoid crowding in vehicle. At this time, the time for Willingness to pay(WTP) increases proportionally to the time for travelling and crowding, but the ratio of time for WTP is inversely proportional to the time for crowding.
Initially, it tried to estimate VoCT in the car by matching smart card data with respondents, but there was a limit to matching because there was no personal information on the smart card. Therefore, No, 702 and 7022 runing at the same origin and destination were analyzed at depth. 702 occurs Crowding but travel time is short, and 7022 not occur Crowding but travel time is long.
The VoCT In-vehicle was estimated in the form of time multiplier, such as standing time versus sitting time and crowded standing time versus sitting time.
As a result of estimating, the value of standing time per unit hour in-vehicle was 1.26 times the sitting time, and the value of crowding and standing time was 1.99 times the sitting time.
This study is meaningful as the first study to estimate the VoCT in-vehicle of city buses using Smart-Card data from reliable city buses. It is judged that it can be used to review the feasibility of adjusting the number of proper operations and headway between routes and to evaluate the management and service of city buses using the estimated the VoCT in vehicles. In addition, the method of Smart-Card big data analysis presented in this study is available as a new method to analyze crowding and characteristics of city buses and is expected to contribute to various research in the future.