Dynamic control using knowledge-based systems and genetic algorithm technology can consider current flow demand information, site-dependent information, accident occurrence data, flow predictions and meteorological conditions, and use a wide range of weight-adjustable performance parameters as goals of the optimisation process. Genetic algorithms are used for the optimisation. This approach is not goal-oriented towards `the optimum' for any given situation. The advantage of the method is that the most optimal arrangement need not be found due to the stochastic nature of the input parameters. The inter-relationships between the measures of intersection performance are not linear therefore, linear therefore, depending on sensed or perceived conditions, each is assigned a weight that is varied by a knowledge-based system. (A)
Abstract