SAFETY EVALUATION OF ALTERNATIVE HIGHWAY DESIGN IN A COMPUTER AIDED DESIGN ENVIRONMENT

Abstract

Modeling traffic crashes is a complex undertaking. Previous research studies have used a variety of techniques to analyze crashes. Conventionally, traffic crashes have been modeled using regression models. Recently, intelligent systems have been applied in highway safety modeling. Such methods include artificial neural networks, decision trees, nearest-neighbor rule, Bayesian methods, and clustering algorithms. This research investigated the use of Artificial Neural Network technique in highway safety modeling. A prediction model using Artificial Neural Networks technique is proposed as part of the efforts to enhance traffic safety data analysis. The technique takes advantage of the knowledge of causal relationships or statistical dependencies (or independencies) among the model variables. A simple hypothetical Belief Network that comprised of seven variables i.e., lane width, shoulder width, shoulder type, Number of access, Number of curves, Road side development was constructed. The model allows for the prediction of number of crashes per year per kilometer at a roadway segment given a set of values of each of the model variables. Using a number of ANN architecture the model which is giving the least value of root mean square error is suggested as the best model also in this model the average difference between observed data and the Network output is least. Therefore using the model the sensitivity of all the variables was determined and using AUTO CAD, ADOBE PHOTOSOFT AND 3D MAX softwares the possible remedies for the entire section is suggested by visualization. The area of study for entire work is state highway 57 which connects Delhi to Saharanpur.