Identifying residential and workplace locations from transit smart card data

public transit is highly promoted worldwide to reduce traffic congestion. an evidence-based planning of stop locations and routes with regard to residential and workplace locations can reduce walking distances to transit and the number of transfers, which can improve service quality of public transport and thus increase ridership. this paper proposes a novel method of identifying residential and workplace locations from smart card data. the proposed method identifies relevant stops first and then refines their catchments to narrow down residential and workplace locations in three steps: defining constraints from the design of the public transport network, movement logic, and land use. in 2017, we tested the method using beijing smart card data. the results show close to 69% residential locations inference rates and more than 72% workplace locations inference rates. the mean value of inferred areas is approximately 20% of the areas derived by traditional methods. available data on alighting stops verify the inferred results at least for flat fare systems.