Optimal locations for bikeshare stations: A new GIS based spatial approach

Abstract

This study identifies three new bike station locations for Baltimore City's bikeshare program using a location allocation spatial analysis tool. Over 1.6 million Global Positioning System (GPS) coordinates of bikeshare trips over a period of four months were used to identify the routes most frequently used by bikers. It was hypothesized that potential new bike station locations have a relationship with proximity to transit, attractions and restaurants/pubs. A negative correlation was found between route usage intensity and proximity to attractions and restaurants/pubs, which means that the closer the proximity to attractions or restaurants/pubs, the higher the road usage intensity is. A new methodology was developed by modifying Huff's gravity model, replacing facility location size with a suitability score. The bike station suitability score is calculated with regard to proximity to transit, attractions, restaurants/pubs and existing bike stations. A location allocation model is developed to maximize market share, i.e., the potential bike station locations will be accessible to maximum population while also being within a 400-m (0.25-mile) radius of attractions, restaurants/pubs and at least 300 m from existing bike stations. The location of the proposed bike stations through location allocation is within a block of the planned bike stations by the City of Baltimore, which corroborates the findings of this study. This methodology will be useful to identify new facility locations in any metropolitan city where maximum exposure is a necessity.