Resumen
Delay is an important parameter that is used in the performance evaluation of signalised intersections. Delay is influenced by many variables and hence its determination is a complex task. Webster's classical delay formula is the oldest and the most popular one among the models developed to estimate average delay per vehicle at signalised intersections for homogeneous and strict lane disciplined traffic. Direct application of these models to non lane based heterogeneous traffic condition will result in erroneous estimation of delay. In this study Webster's delay model is modified to suit the road traffic condition existing in India. Semi-empirical adjustment term in the Webster's model has been modified and calibrated based on field observation of delay for different control conditions at signalised intersections. Data regarding traffic stream parameters, signal timing details and delay to vehicles are collected from different signalised intersections having varying control conditions using videographic survey and Global Positioning System (GPS). Based on field observed delay, adjustment term is estimated and modelled using Artificial Neural Network (ANN) approach. The input variables selected for modelling are approach width, effective green time, degree of saturation, vehicle arrival rate, proportion of heavy vehicles and proportion of right turning vehicles. The ultimate network architecture is determined based on Root mean Squared Error (RMSE) values and coefficient of determination (R2) values. Comparison of the results obtained using the developed model with conventional methods and field observations shows that the delay estimated by the developed model is in good agreement with observed delay.