Optimizing the Carpool Service Problem with Genetic Algorithm in Service-Based Computing

Abstract

Carpooling increases the occupancy rate of cars by decreasing the number of empty seats, thereby creating an effective solution to traffic congestion. This paper proposes an intelligent carpool system, BlueNet, which comprises two important modules. These modules are called the Mobile Client module and the Cloud Global Carpool Services module. By using smart handheld devices, users can submit carpool requests and obtain matches within the Mobile Client module via the Cloud Global Carpool Services module. The Cloud Global Carpool Services module generates acceptable matches via the Genetic-based Carpool Route and Matching algorithm. The proposed algorithm furthers the solution to the carpool service problem by dramatically reducing the time required to match a large number of users. In regard to the quality of the matches and processing time, the experimental results show that the proposed Genetic-based Carpool Route and Matching algorithm is able to find carpool route and matching results that are among the most optimal, and operates with significantly less computational complexity to require less services computing time.