| Summary: | 博士 === 國立臺北科技大學 === 電子工程系博士班 === 103 === Traffic congestion has been a serious problem in many urban areas around the world. Carpooling is one of the most effective solutions to traffic congestion. It consists of increasing the occupancy rate of cars by reducing the empty seats in these vehicles effectively. In this paper, an advanced carpool system is described in detail and called the Intelligent Carpool System (ICS), which provides drivers and passengers the use of the carpool services via a smart handheld device anywhere and at any time. For help in coordinating the ride matches via the carpool service agency, we apply the family of Genetic Algorithms (GAs) to propose the variant GA approach by three schemes: (1) Genetic Constitution Scheme, (2) Fuzzy Controlled Genetic-based Scheme and (3) Low-Complexity and Low-Memory Computation Scheme for the combinational optimization problem called the Carpool Service Problem (CSP). In the experimental sections, the test was first conducted to evaluate the proposed approach, which is based on the GA, in regard to optimum solution involving the primary and secondary optimal objectives. Then, the efficacy of integrating the GA with fuzzy logic controller was verified. Ultimately, the experimental analyses demonstrate that our low-complexity and low-memory approach by using compact GA achieves the highest degree of performance in regard to solution quality, processing time, and memory requirements of all evaluated methods in a comprehensive scale.
|
|---|
