Assessing Physical Environment of TOD Communities around Metro Stations : Using Big Data and Machine Learning

Policy makers and city planning professionals who work on transit-oriented development are often interested in evaluating the quality of physical environment around metro stations. How to carry out this task comprehensively, effectively and repeatedly, with limited time and budget? Under the GEF Sustainable Cities Integrated Approach Pilot Project (P156507), the task team has explored the possibility of utilizing street view photos and machine learning models. The analysis measures physical environment from four aspects, id est, convenience, comfort, vibrancy and characteristics using 14 subsets of indicators. It covers 201 stations within the 5th Ring Road of Beijing and all indicators are measured for areas within 10-minute walking distance from the metro stations. The analytic results can be used to support data-driven and evidence-based city planning and zoning