Optimization and Machine Learning

Optimization for Machine Learning and Machine Learning for Optimization

Rachid Chelouah,Patrick Siarry

• Éditeur: Wiley-ISTE
• Année: 2022
• ISBN:
  9781119902874
• Édition: 1
• Sujet: Informatik
• Sous-sujet: Informatik Allgemein

Machine learning and optimization techniques are revolutionizing our world. Other types of information technology have not progressed as rapidly in recent years, in terms of real impact. The aim of this book is to present some of the innovative techniques in the field of optimization and machine learning, and to demonstrate how to apply them in the fields of engineering. Optimization and Machine Learning presents modern advances in the selection, configuration and engineering of algorithms that rely on machine learning and optimization. The first part of the book is dedicated to applications where optimization plays a major role, and the second part describes and implements several applications that are mainly based on machine learning techniques. The methods addressed in these chapters are compared against their competitors, and their effectiveness in their chosen field of application is illustrated.

PART 1
Optimization

1
Vehicle Routing Problems with Loading Constraints: An Overview of Variants and Solution Methods

Ines SBAI¹ and Saoussen KRICHEN¹

¹Université de Tunis, Institut Supérieur de Gestion de Tunis, LARODEC Laboratory, Tunisia

This chapter combines two of the most studied combinatorial optimization problems, namely, the capacitated vehicle routing problem (CVRP) and the two/three-dimensional bin packing problem (2/3D-BPP). It focuses heavily on real-life transportation problems such as the transportation of furniture or industrial machinery. An extensive overview of the CVRP with two/three-dimensional loading constraints is presented by surveying over 76 existing contributions. We provide an updated review of the variants of the L-CVRP studied in the literature and analyze some of the most popular optimization methods presented in the existing literature. Alongside this, we discuss their variants and constraints, their applications for solving real-world problems, as well as their impact on the current literature.

1.1. Introduction

Although the vehicle routing problem (VRP) is the most studied combinatorial optimization problem, the challenge still remains to achieve the most optimal and effective results (Sbai et al. 2020a). The VRP aims to minimize total traveling cost in cases where a fleet of identical vehicles is used to visit a set of customers. The VRP is used in many real-world applications, for example: pharmaceutical distribution, food distribution, the urban bus problem and garbage collection. The basic version of the VRP is known as the capacitated VRP (CVRP); each vehicle has a fixed capacity which must be respected and must not be exceeded when loading items. It is aimed at minimizing the total cost of serving all the customers. The CVRP can be extended to the VRP with time windows (VRPTW) by adding time windows to define the overall traveling time for a vehicle. It can also be extended to the VRP with pickups and deliveries (VRPPD) where orders may be picked up and delivered. Another variant of the basic CVRP is the VRP with backhauls (VRPB). Here, pickups and deliveries may be combined in a single route; all delivery requests therefore need to be performed before the empty vehicle can collect goods from customer locations. Two surveys, conducte...

Table des matières

• Cover
• Table of Contents
• Title Page
• Copyright
• Introduction
• PART 1 Optimization
• PART 2 Machine Learning
• List of Authors
• Index
• End User License Agreement