Do Stochastic Traffic Assignment Models Consider Differences in Road Users' Utility Functions?
NIELSEN O, Technical University of Denmark, Denmark
The early logit-based stochastic traffic assignment models (e.g. Dial, 1971) rest on the assumption that different routes are independent. Thus, they lead to problems in networks with overlapping routes (see Sheffi, 1985, pp.294-297). Daganzo & Sheffl (19
The early logit-based stochastic traffic assignment models (e.g. Dial, 1971) rest on the assumption that different routes are independent. Thus, they lead to problems in networks with overlapping routes (see Sheffi, 1985, pp.294-297). Daganzo & Sheffl (1977) suggested the use of probit-based models to overcome this problem, and Sheffi & Powell (1981) presented an operational solution algorithm in which the road users' 'perceived travel resistances' are simulated. A similar concept is used as a part of the Stochastic User Equilibrium (SUE) suggested by Daganzo and Sheffi (1977) and operationalized by Sheffi & Powell (1982).
In this paper it is discussed whether this modelling approach is sufficient to describe the road users' behaviour. It is shown that the perceived travel resistances to a certain point can make up for the road users lack of knowledge of the 'true' travel resistances. However, it is also shown that it does not fully consider variations in the road users' utility functions (e.g. the weighting of travel length versus time).
A heuristic modification of SUE is presented in which two types of stochastic components occur - the first considers road users' perception of the traffic network at link level (as in the traditional SUE) and the second considers differences in the road users' utility functions. The traditional SUE is a special case of this model, where the second stochastic component is zero. If the first stochastic component is zero, each road user is assumed to have full knowledge of the road network (the perceived travel resistances are equal to the 'true' resistances). However, the road users might still use different routes as they might have different utility functions.
To illustrate the theoretically discussions in the paper, bundles of routes between two zones in Copenhagen are presented according to the different principles and compared with the results os a comprehensive traffic survey. In addition, the methods have been tested on several full-scale traffic models.
Association for European Transport