Decision support for bus operations under uncertainty: A fuzzy expert system approach

This paper presents a novel approach to handle uncertainty in the daily operation of urban bus transit systems using a straightforward framework. A fuzzy expert system is developed to properly formulate the current bus operation problem in Taiwan. To facilitate the construction of the fuzzy knowledge base and to allow feasible action alternatives of multiple action types to be generated for multiple abnormal conditions on multiple bus routes, the system is composed of three decision support modules: demand, supply, and decision. The demand module determines the extra trip requirement to cope with multiple abnormal conditions. The supply module works out the capacity available for providing extra trips. The decision module generates action alternatives to meet the extra trip requirement, and ranks them based on the degree of feasibility modelled by the willingness/ability of drivers available for providing extra trips. As a result, the best feasible action alternatives can always be generated if supply exceeds demand. In cases where supply is less than demand, the best possible action alternatives based on the extra available capacity are suggested. Examples are given to illustrate how the system can support decision making under different situations.