Ridesharing architecture for patterned commute traffic

Abstract

Major cities around the world face traffic congestion due to population growth and increase in the number of vehicles on the roads. It is also estimated that up to 85% of daily commutes in personal vehicles are undertaken by a single individual, although most non-commercial vehicles are capable of seating up to five or more passengers. This repetitive behavior of drivers can be identified and modelled as a patterned commute. This paper proposes an architecture to help identify ridesharing options for individuals travelling along the same patterned commute. The proposed architecture is tested on a database of university students. This database contains student's demographic information, including their residential address, commuter status and their course schedule. From that information, their commute patters are modelled and created. Ridesharing options are evaluated based on this information to test the effectiveness of our proposed ridesharing architecture. Results show a potential reduction of up to 30% of the total number of trips can be achieved through appropriate use of ridesharing amongst these students.