Analytics and machine learning in vehicle routing research

  title={Analytics and machine learning in vehicle routing research},
  author={Ruibin Bai and Xinan Chen and Zhi-Long Chen and Tianxiang Cui and Shuhui Gong and Wentao He and Xiaoping Jiang and Huan Jin and Jiahuan Jin and Graham Kendall and Jiawei Li and Zhengyong Lu and Jianfeng Ren and P. Weng and Ning Xue and Huayan Zhang},
  journal={International Journal of Production Research},
  pages={4 - 30}
The Vehicle Routing Problem (VRP) is one of the most intensively studied combinatorial optimisation problems for which numerous models and algorithms have been proposed. To tackle the complexities, uncertainties and dynamics involved in real-world VRP applications, Machine Learning (ML) methods have been used in combination with analytical approaches to enhance problem formulations and algorithmic performance across different problem solving scenarios. However, the relevant papers are scattered… 

An Overview and Experimental Study of Learning-Based Optimization Algorithms for the Vehicle Routing Problem

This paper reviews recent advances in learning-based optimization (LBO) techniques and divides relevant approaches into end-to-end approaches and step-by-step approaches, and concludes the applicable types of problems for different LBO algorithms.

Analytics and machine learning in scheduling and routing research

This special issue largely originated from various discussions during several cross-domain, multi-disciplinary conferences and workshops, especially the 9th Multidisciplinary International Scheduling

Combining Constructive and Perturbative Deep Learning Algorithms for the Capacitated Vehicle Routing Problem

The Combined Deep Constructor and Perturbator is developed, which combines two powerful constructive and perturbative Deep Learning-based heuristics, using attention mechanisms at their core, and demonstrates a cost improvement in common datasets when compared against other multiple Deep Learning methods.

Analysis of box and ellipsoidal robust optimization, and attention model based reinforcement learning for a robust vehicle routing problem

In this work, we consider a class of vehicle routing problem that uses simultaneous pickup and delivery and is constrained by a hard service time window with an objective to minimize costs. In a

Attention, Filling in The Gaps for Generalization in Routing Problems

This paper aims at encouraging the consolidation of the Attention model through understanding and improving current existing models, namely the attention model by Kool et al, and identifies two discrepancy categories for VRP generalization.

A Routing Method for Ridesharing Service by Applying CPLEX

This paper presents a VRPTW solving model that applies Mixed-Integer Programming (MIP) to optimize transport costs and number of vehicles and results show that the cost andnumber of vehicles are reasonably optimized for this model.

Multiple-Drones-Multiple-Trucks Routing Problem for Disruption Assessment

This study proposes a multiple-drones-multiple-trucks (MDMT) routing problem to assess infrastructure in the areas of disruption epicenters using multiple drones that operate in synchronicity with

Neural Combinatorial Optimization for Coverage Planning in UGV Reconnaissance

The urban coverage reconnaissance task is formulated as a vehicle routing problem, and a new traversal model is proposed, which takes account the connectivity characteristics of the road and proposes a neural combinatorial optimization technique to solve the problem.

Digital Twin Applications in Urban Logistics: An Overview

Urban traffic attributed to commercial and industrial transportation is observed to largely affect living standards in cities due to external effects pertaining to pollution and congestion. In order to

In-Vehicle Data for Predicting Road Conditions and Driving Style Using Machine Learning

A low-cost machine learning system that uses in-vehicle data is proposed to solve three categorization problems; road surface conditions, road traffic conditions and driving style.



RL SolVeR Pro: Reinforcement Learning for Solving Vehicle Routing Problem

Simulation results suggest that the proposed Reinforcement Learning Solver for Vehicle Routing Problem (RL SolVeR Pro) is able to obtain better or same level of results, compared to the two best-known heuristics: Clarke-Wright Savings and Sweep Heuristic.

Single vehicle routing with stochastic demands : approximate dynamic programming

A stochastic dynamic programming model is formulated and Approximate Dynamic Programming (ADP) algorithms are implemented to overcome the curses of dimensionality and reduce the computational time signicantly.

Reinforcement Learning for Solving the Vehicle Routing Problem

This work presents an end-to-end framework for solving the Vehicle Routing Problem (VRP) using reinforcement learning, and demonstrates how this approach can handle problems with split delivery and explore the effect of such deliveries on the solution quality.

Deep Reinforcement Learning for Solving the Heterogeneous Capacitated Vehicle Routing Problem

This article proposes a DRL method based on the attention mechanism with a vehicle selection decoder accounting for the heterogeneous fleet constraint and a node selectionDecoder accountingfor the route construction, which learns to construct a solution by automatically selecting both a vehicle and a nodes for this vehicle at each step.

Deep Reinforcement Learning Approach to Solve Dynamic Vehicle Routing Problem with Stochastic Customers

DRLSA is evaluated against the commonly used Approximate Value Iteration (AVI) and Multiple Scenario Approach (MSA) and shows that DRLSA can achieve on average, 10% improvement over myopic, outperforming AVI and MSA even with small training episodes on problems with degree of dynamism above 0.5.

Scenario-Based Planning for Partially Dynamic Vehicle Routing with Stochastic Customers

This paper considers a dynamic VRPTW with stochastic customers, where the goal is to maximize the number of serviced customers and presents a multiple scenario approach (MSA) that continuously generates routing plans for scenarios including known and future requests.

A simulation-based solution approach for the robust capacitated vehicle routing problem with uncertain demands

The aspiration to create models, which are more appropriate for real applications in which stochasticity is a major issue, is present in different optimization problems, such as the Vehicle Routing Problem (VRP), which was handled with the assumption that all inputs of the problem were deterministic and known in advance.

ADMM-based problem decomposition scheme for vehicle routing problem with time windows

An adaptive data-driven approach to solve real-world vehicle routing problems in logistics

An adaptive data-driven innovative modular approach for solving the real-world vehicle routing problems (VRPs) in the field of logistics with an innovative multistep algorithm that combines several data transformation approaches, heuristics, and Tabu search.