THE EFFECTS OF INCLUDING SOCIAL FACTORS IN RIDE-MATCHING ALGORITHMS ON THE PERFORMANCE AND THE QUALITY OF MATCHES

Advancement in communication technologies has fostered alternative transport modes, such as ride-sharing. Ride-sharing aims to increase vehicle occupancy rates by matching riders with the drivers, who have empty seats on their vehicles and have similar routes and time schedules. Regarding to the success of a ride-sharing system, many researchers have been interested in efficient ride-matching algorithms. Ride-matching optimization problem is considered as NP-Hard Problem. In most of the ride-matching algorithms in the literature, to be able find matches at short notice some parameters were omitted. Hence, social characteristics and choices of participants, such as gender, age, employment status and willingness to socialize, were not included in many ride-matching algorithms. In this paper, the effects of including such factors in a ride-matching algorithm on the performance and the quality of the matches are investigated. Several ride-matching algorithms in the literature are simulated with randomly generated data. The simulation results show that when social factors are included the computation times and the quality of the matches increase significantly.