A Belief Rule based expert system to control traffic signals under uncertainty

Traffic congestion is a condition on road network that occurs as vehicle increases, and is characterized by slower speeds, longer trip times, and increased vehicular queuing. It effects the economic growth of a country, increases accidents, resource cost and environment pollution. One of the most cost-effective ways, to deal with this problem is by employing traffic control signals at the road intersections. Now-a-days, most signal controls are implemented with either fixed cycle time control or dynamic control. These conventional methods for traffic signal control fails to deal efficiently because they are unable to taking account of the uncertainty associated with traffic flow. Therefore, this paper presents the design, development and application of a belief rule based expert system (BRBES) with the capability of handling uncertainty. The system uses Belief Rule Base (BRB) as the knowledge representation schema and the evidential reasoning as the inference engine. The results generated by the system have been compared with a fuzzy logic based expert system (FLBES). The BRBES´s reliability is better than that of FLBS. The system applied in a number of road intersections of the Chittagong City of Bangladesh.