Two aspects of public transport management are addressed: real time control and passenger information; the latter comprising pre-trip information as well as en-route information. For each aspect, after a recap of the main requirements, the state of the art is presented (especially Europe and North America). Artificial Intelligence (AI) techniques and methods are viewed as a subset of computer tools which, along with more conventional computer tools, can help the public transport operators as well as improve traveller information. Benefits expected from such tools are listed. More details are provided for those specific tasks or problems which are amenable to AI techniques. As a result of all previous experiences and current initiatives, emphasis is placed on three conclusions: (i) Real time management of public transport (e.g. bus network) must be carried out in connection with the management and control of urban traffic; (ii) AI techniques cannot by themselves address such a complex situation. Real advanced management of urban public transport is primarily based on the development of complementary techniques such as real time vehicle localisation, data bases, geographical information systems, communication links; and (iii) Knowledge Based Systems and Constraint Reasoning are likely to be the most popular AI techniques to be implemented in the near future in the urban public transport domain. (A)
Samenvatting