Volume 20, Issue 3 (9-2023)
jor 2023, 20(3): 87-107 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Sadra M, Zaferanieh M. An Online Supervised Machine Learning Algorithm to Solve a Bi-Level Network Design Problem. jor 2023; 20 (3) :87-107
URL: http://jamlu.liau.ac.ir/article-1-2146-en.html
URL: http://jamlu.liau.ac.ir/article-1-2146-en.html
Mathematics and Computer Sciences Department, Hakim Sabzevari University, Sabzevar, Iran, Hakim Sabzevari University, Iran , mehdi.zaferanieh@gmail.com
Abstract: (925 Views)
This paper investigates a bi-level network design problem and proposes a hybrid optimization-machine learning algorithm to solve it. Bi-level problems are typically NP-hard, and even in the simplest scenario where both upper and lower level problems are linear, they remain NP-hard. Machine learning methods have gained popularity in recent years due to their accuracy and reasonable running time. In this study, we apply a hybrid optimization-machine learning algorithm using an online learning method to solve the proposed bi-level network design problem. Our primary objective is to select a set of candidate edges that minimize traffic flow as much as possible. The upper-level objective function aims to minimize the average total travel time and fixed cost expenses required to establish the newly designed edges. Meanwhile, the lower-level objective function focuses on minimizing individual travel time from the viewpoint of users. We provide several numerical examples to demonstrate the effectiveness of our proposed model and its hybrid solution approach.
Keywords: Bi-Level Programming, Supervised Machine Learning, Traffic Assignment Problem, Network Design Problem
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
