The addition of transit routes transforms urban development by disrupting the existing equilibrium that land parcels have achieved over time and promotes revitalization. It is based on the relationships between land parcel variables and transit route characteristics, including feeder routes and road infrastructure. Traditional parametric methods for explaining this relationship have problems with multicollinearity and generalizability while non-parametric methods are not used with the multiple variables of both transit route and land parcel changes over time. This study applies the C5.0 decision tree algorithm, a non-parametric model that creates a decision tree with leaf nodes that predict the relationship. Using the BRT Lahore case study, the time series data of parcel variables in the 2 km circle of five transit stations before BRT 2010 and after BRT 2018, as well as transit route characteristics including feeder routes and road infrastructure, were collected and analyzed. The model identified eight important predictors and explained the relationship in the form of a flowchart. Property condition emerged as the strongest predictor, followed by property value, parking, population density, land use, building height, access routes, and distance from transit stations, in that order. The results show that well-developed transport infrastructure, parking spaces, and feeder routes enable sustainable urban transformation.