This paper deals with a case study where the objective is to identify the optimal subset of routes for real-time traveler information in a highway network. It is assumed that the benefit from providing this information is directly related to the uncertainty of route travel times. The variance of travel times within a time period over consecutive days is employed as the indicator of this uncertainty. New Jersey Turnpike is used as the study network due to the availability of vehicle-by-vehicle network specific data. The data set covers travel times between ∼630 origin-destination (OD) pairs during 2004. The problem of identifying the optimal number of subset of routes is modeled as a nonlinear integer-programming problem. The proposed model is then solved using NEOS server, a common optimization solver available over the Internet. A simple heuristic for the proposed model is also presented.