A new binary genetic programming approach to design public transportation systems according to transit-oriented development criteria

This study introduces a new evolutionary approach called binary genetic programming (BGP) to design and assess public transportation systems from sustainable development perspective. The BGP combines system identification techniques with k-fold cross-validation obtain an accurate model between the land use parameters urban point of view. To model, two including tram line Antalya (Turkey) bus rapid transit Bhopal (India) were considered. was applied classify into transit-oriented (TOD) non-TOD. solutions generated by compared those classic decision tree (DT) as well state-of-the-art random forest (RF) models evolved benchmarks in this study. results showed that is highly efficient may provide less than 5% classification error. It superior DT RF solutions, which typically require higher datasets avoid overfitting. Furthermore, explicit formulation combination multicriteria evaluation method increases human insight on factors affecting transportations