Estimating Public Bicycle Trip Characteristics with Consideration of Built Environment Data

A reliable estimation of public bicycle trip characteristics, especially distribution and duration, can help decision-makers plan for the relevant transport infrastructures assist operators in addressing issues related to imbalance. Past research studies have attempted understand relationship between generation, attraction factors such as built environment, weather, population density, etc. However, these typically did not include distribution, detailed information on environment. This paper aims estimate daily i.e., attraction, duration using points interest smart card data from Nanjing, China. Negative binomial regression models were developed examine effect environment usage. Totally fifteen types (POIs) are investigated residence, employment, entertainment, metro station found be statistically significant. The results showed that 300 m buffer POIs restaurant, bus stop, station, amenity, school significantly positive effects generation while, counterintuitively, shopping, parks, attractions, sports, hospital negative effects. Specifically, an increase 1% distance leads a 2.36% decrease origin-destination (OD) trips or 0.54% duration. We also number other nearby stations reduce 0.19% OD trips. this offer useful insights better estimating usage providing services improve ridership.