This special issue aims to attract the state of the arts of techniques used to solve optimization problem appearing in different areas of logistics and transportation systems. https://www.sciencedirect.com/journal/transportation-research-part-e-logistics-and-transportation-review/about/call-for-papers

Hipster list

Keywords:

  • Logistics
  • transportation systems
  • machine learning
  • combinatorial optimization
  • optimization
  • solution methods

Guest editors:

  1. Dr. habil. Shahin Gelareh - Universite d’Artois, Bethune, France ( shahin.gelareh@univ-artois.fr )
  2. Dr. habil Nelson Maculan - Federal University of Rio de Janiero (maculan@cos.ufrj.br)
  3. Dr. Rahimeh Neamatian Monemi - Université de Lille, (rahimeh.monemi@icam.fr)
  4. Dr. Pedro Henrique González - Federal University of Rio de Janeiro ( pegonzalez@cos.ufrj.br)
  5. Dr Xiaopeng Li - University of Wisconsin-Madison, Madison, WI, United States ( xli2485@wisc.edu )
  6. Dr. Fatmah Almazkoor - University of Kuwait ( fatmah.almazkoor@ku.edu.kw)
  7. Dr Ran Yan - School of Civil and Environmental Engineering, Nanyang Technological University ( ran.yan@ntu.edu.sg )

Special issue information:

A significant body of literature focuses on methods of optimisation in mathematical programming, particularly for addressing logistics and transportation problems. In addition to the theoretical developments that have led to highly efficient techniques, researchers have always tried to exploit the inherent structure of data and problem instances in handcrafting techniques to tweak and improve the performance of their ad-hoc methods. Furthermore, recent advances in machine learning and deep learning had also led to some significant improvement in the performance of solution techniques that learn to solve optimisation problems to optimality or to some provably near-optimal solutions. With that, classification and regression methods come in support of the classical techniques in search algorithms, branching and cutting decisions, estimating the primal dual properties and even further in more advanced and complex methods. Deep learning approaches use various neural-based methods such as graph convolutional networks, attention mechanisms, and reinforcement learning to learn policies for finding optimal solutions. Many promising results have been reported on many problems, especially for combinatorial optimization problems, that are building blocks for more complex optimization problems in logistics and transportation systems Although there is still much work to be done to improve efficiency of such techniques, there are already highly sophisticated techniques available.

Topics of interest:

This Special Issue invites authors to submit articles focusing on optimization methods that rely on learning techniques to address problems in logistics and transportation. Theoretical papers are acceptable, provided that they have case studies/numerical examples in the logistics/transportation field; models and algorithms that utilize learning to better understand the problem structure, physics, and behavior fall in the scope of the special session. We are particularly interested in contributions that are comprehensive enough to also cover or address problems in logistics and supply chains, that consider sustainability, IoT, electric vehicles, energy efficiency, and other relevant areas. We welcome both original research and review articles. Possible contributions may include, but are not limited to, the following topics:

  • Enhancing classical methods via ML
  • Markov Decision Process
  • Neural methods
  • Learning for primal-dual techniques
  • Reinforcement learning based methods
  • Novel classes of methods