Identifying trip purpose from a dockless bike-sharing system in Manchester

This study explores the application of a Bayes ruled-based algorithm to uncover the purpose of trips from a dockless bike-sharing system in Manchester. The purpose is to produce a picture of the activity patterns at an aggregated level to understand the role of such service. We relied on GPS data generated from every bicycle and a geospatial dataset of buildings and parks in the Manchester central area. The method proved to uncover trip purposes that, otherwise, would only be known through a survey and capable of distinguishing between activities in mixed land-use areas. Our findings indicate that the user's relationship with the service is complex. Trips to residential areas are the most important ones, especially in the evening whilst trips to work are predominant in the morning. This produces an inward-outward movement of bicycles throughout the day. With over a quarter of the total trips, retail activities proved to be more popular than cycling to education. The study provides an insight into people's travel preferences when allowed to cycle freely through the city and searches to bridge the gap between new mobility systems and traditional transport strategies.