REVIEW OF ROUTE-LEVEL RIDERSHIP PREDICTION TECHNIQUES (ABRIDGMENT)

An important trend in the local transit planning field is away from large-scale, capital intensive planning toward low-cost operational planning. Because most major transit and highway facilities are in place, greater consideration is being given to making minor changes to improve the efficiency and increase the capacity of existing transit services. Transit managers need to predict the effect of proposed service changes on ridership for a variety of reasons: (a) to allocate vehicle and manpower resources, (b) to prepare budget requests for proposed service plans, and (c) as inputs into the detailed route planning and scheduling that must accompany new service plans. To perform these tasks adequately, route-level patronage models must be sensitive to service characteristics as well as to the more traditional socioeconomic characteristics of the area the route passes through. The service quality measures most often affected by the route-level service modifications made by most transit properties are headway adjustment, route extension and contraction, limited and express service, shortlining, branching, through routing, creating transfer opportunities, fare adjustments, and new hours of service. A review of techniques that are currently used in the industry for predicting route-level ridership is presented. This review is based on discussions with the planning staffs of 40 transit agencies. Seven criteria were selected for evaluating the various techniques: accuracy, sensitivity, range of application, analyst dependence, cost of application, technical sophistication, and transferability.