Capacity Research in Urban Rail-bounded Transportation with Special Consideration of Mixed Traffic
Final Report Abstract
In urban mixed traffic zones, the rail-bound transport is heavily influenced by road traffic. These influences cannot be ignored, as they lead to an amount of loss of URT capacity. It is meaningful to evaluate the potential loss of capacity due to RT influences. The interaction between URT and RT may result in possible economic costs not only on URT but also for RT. A standardized procedure to evaluate the possible improvement (e.g. to build new tunnels or bridges to separate URT and RT) can be further developed. The developed two approaches can be used to carry out capacity research with consideration of RT influences on URT. With two approaches, the stochastic influences caused by RT are described systematically. According to the required application conditions, it is necessary to select a suitable approach. For a modeling approach, the regulations of traffic control signaling system reflect the operation of modeled road traffic at level crossing under different time periods and day types. Moreover, shared road can be modeled as many adjacent level crossings with this method. It is suitable to use a modeling approach for relative simple mixed traffic zones. On the other hand, distribution approach is applicable for a complex investigated area with sufficient statistics data of delays to determine the perturbation parameters of RT influences. For capacity research, a key ingredient is the waiting time function, which can be derived by simulations of the stochastic influenced timetables with stepwise traffic loads. The adapted waiting time function plays a very important role to fit the discrete waiting times and then to derive the recommended area of traffic flow. The waiting time function with higher coefficient of determination can represent the trend of discrete waiting times better. The improved waiting time function has a better coefficient of determination in comparison to the existing one. The fitting curve of discrete waiting times with new adapted waiting time function shows the results much closer to reality. Accordingly the derived recommended area of traffic flow is much more plausible. The developed algorithm enables not only preliminarily derive the throughput capacity with RT influences Thp-max and the variable d for determination of adapted waiting time function, but also to assess the significant of mixed traffic zone with influences of RT, which can reduce the cost of the following work with two approaches. This research project has been fully processed within the given schedule.
