Public bicycle-sharing (bike share) programs have become increasingly popular in recent years, particularly in Europe, with a number of cities recently implementing systems and high levels of usage. North American efforts have been more limited to date, with high-profile recent examples including a small program in Washington, D.C., and a substantial seasonal program in Montreal, Canada. Because there are no established large-scale programs in the United States, planners exploring potential system designs and feasibilities are faced with an unusual degree of uncertainty about who would ride, where they might ride, and how often they might ride. A large-scale bike share system is under consideration in Philadelphia, Pennsylvania. This paper discusses the methods and findings of a two-phase project that (a) used a raster-based geographic information system analysis to identify a primary geographic market area for a bike share program and (b) applied bike share trip diversion rates observed in peer European cities to estimate daily bike share trips in the primary market area. This analysis resulted in estimates for daily usage in Philadelphia that ranged from roughly 6,000 to 23,000 for two scales of market area and three demand scenarios (low, middle, and high). As bike share systems continue to proliferate in different settings, new data can refine the methods used here to provide increasing levels of certainty in the future.
