Artificial intelligence and DOE: an application to school bus routing problems
- PDF / 1,008,147 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 40 Downloads / 154 Views
(0123456789().,-volV)(0123456789().,-volV)
Artificial intelligence and DOE: an application to school bus routing problems Jonnatan Fernando Avile´s-Gonza´lez1 • Jaime Mora-Vargas2 • Neale R. Smith2 • Miguel Gaston Cedillo-Campos3
Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract This paper presents the implementation of simulated annealing (SA) method, an artificial intelligence technique, to solve the optimization problem known as the school bus routing problem (SBRP). A specific challenge in all artificial intelligence optimization techniques is the selection of appropriate value parameters. One contribution of this paper is the implementation of a design of experiments technique to provide statistical support for parameter selection. The SBRP is formulated as a 0–1 integer linear programming model, where the objective function is to minimize the total cost. Because this problem is combinatorial in nature, it is not possible to find exact solutions in an adequate time, calling for the use of an artificial intelligence optimization technique. The proposed technique is SA due to its modeling flexibility and processing speed. To demonstrate the performance of the proposed algorithm, several experiments with real instances were carried out, showing that the metaheuristic algorithm performs better in quality and time than the classic routing method. Keywords Design of experiments Simulated annealing School bus routing Optimization
1 Introduction Routing is an important task to provide quality of service to the users of transportation systems [1]. The present research has as a motivation, the possibility to contribute to the understanding of techniques and methods focused on solutions to improve transportation routing. Several authors have presented different methods and algorithms to address this situation. Even so, these methods are susceptible to be improved in terms of selection of parameters or functionality. To ensure various quality of service parameters for heterogeneous traffic types, it is important to use multiple
& Miguel Gaston Cedillo-Campos [email protected] 1
Escuela de Ingenierı´a en Produccio´n, Universidad del Azuay UDA, Av. 24 de Mayo 7-77 y Herna´n Malo, 0101981 Cuenca, Azuay, Ecuador
2
Escuela de Ingenierı´a y Ciencias, Tecnolo´gico de Monterrey, Eugenio Garza Sada 2501, 64849 Monterrey, Nuevo Leo´n, Mexico
3
Laboratorio Nacional en Sistemas de Transporte y Logı´stica, Instituto Mexicano del Transporte, KM 12?000, Carretera Estatal No. 431, 76703 San Fandila, Quere´taro, Mexico
good quality paths with an integrated approach of routing metrics in order to improve overall performances [2]. Routing is a technique for tackling the traffic demand uncertainty problem. A routing scheme derived from this principle intends to achieve a predicable performance for a set of traffic matrices [3]. Several authors have presented different approaches and algorithms to solve routing and transportation problems. For an example of a network design model,
Data Loading...
