01698241

Component

In this paper, the authors propose a new stochastic user equilibrium (SUE) model to capture route overlapping and route-specific perception variance, namely the Paired Combinatorial Weibit (PCW) model. The paired combinatorial model, which belongs to the Generalized Extreme Value (GEV) model, considers the similarity relationship between each pair of alternatives using a two-level tree structure. This feature is highly desirable as it does not restrict the sequence of the choice structure and allows for different covariances among all pairs of routes. In addition, the PCW model allows for non-identical perception variances for different trip lengths without the need to impose the identically distributed assumption for different routes as required by the multinomial logit (MNL) model or the paired combinatorial logit (PCL) model. More importantly, the PCW model still retain a closed-form probability solution and an equivalent mathematical programming (MP) formulation for the PCW-SUE model. Solution properties and solution algorithm are also provided along with numerical results to demonstrate the features and applicability of the PCW-SUE model.

This paper was sponsored by TRB committee ADB30 Standing Committee on Transportation Network Modeling.

19-02436

English

5p

2019

Transportation Research Board 98th Annual Meeting

Digital/other

References

Choice models; Combinatorial analysis; Route choice; Traffic assignment; Travel behavior

Stochastic user equilibrium

Highways; Planning and Forecasting

Transportation Research Board Annual Meeting 2019 Paper #19-02436

TRIS, TRB, ATRI

Dec 7 2018 9:49AM