The Researches on Subway Demand Forecast at Station Level: Smart card data, Space Syntax and Points of Interests.

International Conference on Manufacturing Technology, Materials and Chemical Engineering

机械制造;材料科学;化学工业

Wu, Cheng Cheng^1 ; Chen, Da Wei^1

School of Transportation, Southeast University, Nanjing, Jiangsu

210096, China^1

关键词: Built environment;    Demand forecast;    Energy density;    Forecast accuracy;    Multiple linear regressions;    Nanjing Subways;    Points of interest;    Spatial integrations;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/392/6/062113/pdf
DOI  :  10.1088/1757-899X/392/6/062113

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

This paper examines the possible 'attraction' factors at station level in subway ridership, such as built environment factors and Spatial connectivity. Direct demand models were built to estimate subway ridership based on multiple linear regression. Also, a comparison between predicted values and actual values in subway ridership were applied in order to analyze forecast accuracy. It reveals that energy density on rail scale of service and spatial integration could be an important asset for urban rail demand forecast models at station level. Based on such correlative factors, parameters of direct models are chosen and calibrated by means of multiple linear regressions using actual data from Nanjing subway. The results show that the direct models can be more accurate and efficient in estimate subway ridership at station level in a short circle.