Short‐Term Demand for Specialized Transportation: Time‐Series Model

Abstract

This paper focuses on regression and intervention modeling of Lexington/Fayette County specialized transportation, WHEELS, monthly riderships. The 1979–1985 time‐series ridership data are comprised of total monthly trips and their breakdown by passenger type, the passenger's ability to walk, and trip purpose. Intervention models from time‐series analysis properly replicated these monthly riderships. The superiority of intervention modeling when compared with regression analysis was found to be in capturing ridership seasonality, proper incorporation of the time‐lag structure, and the intervening events of fare and service changes. Post‐modeling analysis determined months of ridership peaks. The para‐transit planning and management utility of these modeling techniques in time‐series analysis is clearly evident. Although the selected models are dependent on the 1975–1985 data, the same methodology can be applied to any time‐series ridership data for short‐term demand modeling.