Estimation of critical gap on a roundabout by minimizing the sum of absolute difference in accepted gap data

Estimation of critical gap for a vehicle type under mixed traffic conditions prevailing in developing countries has been always a challenging task. This is due to the poor lane discipline and very limited priority being followed by the vehicles at priority intersections like roundabouts. A simple procedure, which is based on minimization of the sum of absolute difference between a gap value and accepted or rejected gap, is proposed in this paper. The iterative procedure provides a value of gap that is termed as the critical gap under mixed traffic conditions. The method is different from maximum likelihood method (MLM) in two aspects. First, it does not assume any predefined distribution for the critical gap and second, it does not fail even if there is no rejection of gaps which is very common under limited (or no) priority conditions. Field data are collected at a roundabout in India. Prominent methods available in literature to estimate critical gap are compared for different categories of vehicles. Based on the results of consistency test, the MLM and the proposed method are found to be the most acceptable estimation methods. It has been further observed that the proposed method is better than MLM when working with low sample size, as well as, in no-priority conditions, which arise due to heterogeneous traffic flow prevailing in developing countries. The results are validated on another roundabout having similar physical features.