Travel time variability (TTV) is increasingly acknowledged to be an important concern for both the users and the providers of transport services. The correct measurement of variability and its value for the users are important in the design of transport policies.

Travel time variability is associated with risk, hence risk and its perception should be reflected in the valuation of variability. The focus of this paper is on the perception of variability and its value for the users by application of a Cumulative Prospect Theory (CPT) approach. CPT is a descriptive theory of decision making under risk that departs from Expected Value Maximization in two essential ways; the transformation of outcomes using value functions where the carriers of value are gains and losses rather than final levels, with diminishing sensitivity for both gains and losses; and the rank-dependent transformation of probabilities using probability weighting functions.

The current paper applies data from two Stated Choice experiments from a recent Norwegian value of time study (Ramjerdi, et al, 2010), which addresses variability of travel time for all modes of travel and for both long and short distance travel. In both experiments, travel alternatives differ with respect to travel time and cost, but in the first experiment, travel time is known with certainty, while in the second it is subject to variability, represented by a five-point distribution.

We model behaviour using a logit model with separate value functions for travel time and cost and a rank-dependent transformation of probabilities. We hypothesize that the resulting estimates of the value of travel time differ between the two experiments, because the value from the second experiment encompasses additional disutility due to risk. We argue that the difference between the values of time provides a measure of the value of travel time variability.