Heuristic neuro-fuzzy model for evaluation of urban transportation projects

Summary form only given. Urban transportation planning projects are largely concerned with providing two benefits: trip time reduction and operating cost reduction. However, neglect of human factors in transportation projects and failure to recognise urban transportation as a basic need of a society, have resulted in systems that do not execute their primary objective: to move human beings quickly, comfortably, economically and effectively. The work described here develops a methodology to evaluate urban transportation systems based on qualitative descriptions, wherein humans and their environment are the main factors. The method is based on a neuro-fuzzy approach with heuristic rules organised in a hierarchy designed for a participatory environment of experts, current users and future users. All inputs are transformed into linguistic variables by ascribing degrees of belief to arrive at a final conclusion to assign a humanistic grade to the overall analysis. Neuro-fuzzy technology has been chosen because of its ability to mimic human decision-making and because of the ease of model-validation.