This article shows the microscopic modelling of the interactions between vehicles and passengers on public transport stops and stations, and its relation with the rest of the traffic, as a way to contribute to an appropriate physical and operational design of bus rapid transits (BRT) systems. Its aim is to contribute to the design of bus infrastructures such as busways, interchange stations, bus stops, as well as conventional bus priorities. The text reports how to improve the description of the public transportin conventional traffic microsimulators by means of API (Application Programming Interface). Previous advances reported elsewhere on an API for PARAMICS are considered in this work. The API named MISTRANSIT for MIcroscopic Simulation TRANSIT combines models of stop interactions considering passengers as modelling entities such as the PASSION model with traditional car-following and lane-changing models of PARAMICS. Firstly, two issues on the design of divided BRT stations to increase their capacity are analysed (a divided station is made of m stopping points with n berths each one). The first issue is the suitable separation between two adjacent stop points. In order to do this the evolution of the queue length at the downstream stop point depending on combinations of bus flows and passenger demands was studied. The second element was the study of the impact of weaving manoeuvres on the capacity of the upstream stop point; combinations of flows and demands at the downstream stop point are considered for this issue. Bothissues made possible to obtain the global capacity of a divided station, being observed that the interaction between stop points can reduce capacity by between 7 to 10 %. Having defined the appropriate designs of stops tocope with the demand of buses and passengers, its interaction with the rest of the traffic was analysed. First, the interaction among a stop or station with a downstream traffic signal was studied; in particular, the effect of the distance to the signal and its time settings in the capacity anddelay at the stop. Previous results obtained with the bus stop simulationsoftware IRENE were compared with those coming from the MISTRANSIT approach. The performance of a fixed-time plan versus active priority at trafficsignals was also analysed. The combination of MISTRANSIT and PARAMICS allowed the measurement of benefits and costs for all users as well as those of public and private transport users. Preliminary results indicate that, as the flow of buses increases, the strategy of signals operated by buses lose rapidly its efficiency. In summary, the article contains directions to improve the design of BRT such as those to be implemented in cities as Santiago (Transantiago) or London (East London Transit, Greenwinch Waterfront Transit). The concept and capacity of divided stations, for example, has not been reported in the literature. This has effect in the design of other BRT facilities, as signal priorities, segregated right of ways, and special cross-sections. Also, the suitability, costs and global benefits of other strategies can be analysed as a result of this work; for example, short bus lanes to avoid traffic queues or pre-signals to facilitate manoeuvres of public transport vehicles. For the covering abstract see ITRD E137145.
Abstract