Modelling of accidents for four lane non-urban highways using artificial neural networks technique

the purpose of road accidents forecasting is to analyze the tendency of road accidents under existing road traffic conditions, evaluate the feasibility and practical effectiveness of traffic safety measures reasonably, control the factors affecting road accidents, and reduce the traffic accidents. The characteristics such as non-linearity, randomness and uncertainty in traffic system make it difficult to forecast road accidents, the behavioral feature of traffic system. The objectives of this study are to determine the critical accident variables for accident prediction purposes, to utilize Artificial Neural Network (ANN) as a tool for accident prediction analysis, and to develop an Accident Prediction Model. This study uses a series of artificial neural networks to model and estimate accident by the applied software that is NeuroSolutions 6.0. An important implication of this paper after sensitivity analysis is that the variables like mixed-use developments, side-drain condition, guard rail, marker-post, chevron markings and curve warnings, cross-drainage works, over-head structures, Fly-over Bridge, side roads/access roads, under-pass, and volume of traffic are most significant factors which increase accident.