Adaptation of the paired combinatorial logit model to the route choice problem

The paired combinatorial logit (PCL) model is one of the extended logit models recently developed from the generalized extreme value theory to handle the drawbacks of the multinomial logit (MNL) model. The PCL model is considered suitable for adaptation to the route choice problem to resolve the independence from irrelevant alternatives property inherited in the MNL model while keeping the analytical tractability of the logit choice probability function. In this paper, we adapt the PCL model to handle the two drawbacks of MNL in a route choice context: (1) inability to account for overlapping (or correlation) among routes, and (2) inability to account for perception variance with respect to trips of different lengths. Specifically, the effects of congestion, scaling, and overlapping on the PCL choice probabilities are examined using a set of carefully designed experiments and graphical illustrations. The experimental designs and graphical illustrations allow for intuitive explanation and comparison of the different effects on the calculation of the route choice probabilities.