Cost and risk aggregation in multi-objective route planning for hazardous materials transportation - A neuro-fuzzy and artificial bee colony approach

Abstract This paper proposes a new approach for cost and risk assessment in the multi-objective selection of routes for the transport of hazardous materials (hazmat) on a network of city roads. The model is based on the application of an Adaptive Neuro Fuzzy Inference System (ANFIS). The values of the cost and risk criteria are, using an adaptive neuro-fuzzy network trained with an Artificial Bee Colony (ABC) algorithm, integrated into a single CR value by means of which the worthiness of each branch in the network is expressed, and after which the selection of the route is made using Dijkstra's algorithm. The ANFIS adequately treats a number of uncertainties and ambiguities in the input data and enables the inclusion of the knowledge of experts and the preferences of the decision makers. The procedure is also applicable in cases in which the decision maker does not have high quality data available. The proposed model is tested in a real urban route planning problem, in a case study of the distribution of oil and oil derivatives in Belgrade, Serbia.