Experimental study of boarding and alighting times

Document Type

Conference Paper

Publication Date


Subject Area

mode - bus, place - south america


buses, boarding, alighting, dwell time, passenger numbers


During the past three years, we have done, as part of a project funded by the Chilean Fund of Science and Technology (FONDECYT), laboratory experiments aimed at reducing delays to buses at their stops. We were motivated by the fact that buses remain almost half of their travel time transferring passengers at bus stops. Therefore, a reduction of time in bus stops should have a major impact on journey times for users. This work has been performed in collaboration with the Pedestrian and Movement Environment Laboratory (PAMELA) of University College London and the Digital Imaging Research Centre (DIRC) of the Kingston University in London. The experiments were performed in PAMELA and used a real-scale model of Chilean buses. Normal people got on and off the mock-up simulating the conditions of daily use of the new Santiago's transport system (Transantiago). This type of experience has no precedent in Chilean transport research and, as far as we know form literature review, there are few in the world. As in any laboratory experience, we obtained unexpected results. The first finding is that there is an optimal vertical gap between the platform of the bus stops and the bus chassis. This gap is about five to ten centimetres, which would still allow the rise of a wheelchair or a pushchair. If the bus stop provides this vertical gap, the boarding time of passengers could be reduced by more than a third in normal stops and even more in prepayment stops; i.e., bus stops in which passengers pay when entering the platform. Conversely, the engineers thought that it was best to have platforms at the same level of vehicles. Another unexpected finding is that although the boarding time decreases when the fare is paid before boarding the bus, as in prepayment stops, the decrease is significant only if buses have wide doors; i.e., like the new buses of Transantiago which doors are 1.10-m wide. In such a case stop delays can be reduced in more than 20 percent. However if buses have narrow doors, paying before getting on is irrelevant. Therefore, the installation of prepayment stops in zones of the city where buses have narrow doors would not have a significant effect on passengers’ travel time. Finally, we found that in wide-door buses, regardless of the height of the platform, the alighting time is reduced by 40 percent and the boarding time could be halved. Thus, if bus stops have an optimum platform height, buses have wide doors and the fare is paid before entering the vehicle, a journey time of 45 minutes by bus could be reduced to only 35 minutes. Not bad when you consider that this does not require any expensive technology. In a second run of experiments we studied the effect of the density of passengers in the vehicle on the boarding and alighting times per person. Experience and common sense indicate that the longer a bus is full, the harder it to get on and off. It does not say, however, that the average boarding time increases proportionally as the bus goes full, but the alighting time increases more than proportionately. For example, for a density of six passengers per square meter – the design density of Transantiago – the alighting time is three times longer than the boarding one. Given these results we are building at Universidad de los Andes a Human Dynamic Laboratory in which we will study from how a person put an object in their mouths to the behaviour of a mass of people in confined spaces. This will help from the construction of robotic prostheses to the interior design of public transport vehicles. In this way we want that engineering go back to its origins as experimental science to address the problems of contemporary society.


Permission to publish the abstract has been given by AET, copyright remains with them.