« 12th DIMACS Implementation Challenge: Vehicle Routing Problems
April 05, 2022 - April 08, 2022
Location:
Online Event
Organizer(s):
Claudia Archetti, ESSEC Business School
Tamra Carpenter, DIMACS
Nicholas Kullman, Amazon
Catherine McGeoch, D-Wave Systems
Jorge Mendoza, HEC Montréal
Panos Pardalos, University of Florida
Mauricio Resende, Amazon
Eduardo Uchoa, Universidade Federal Fluminense
Thibaut Vidal, Polytechnique Montréal
Implementation Challenges trace back to the early years of DIMACS and to the vision of David S. Johnson to use experimentation as a companion to theory to gain more holistic perspective on the performance of algorithms. The over-arching purpose of a Challenge is to assess the practical performance of algorithms for a particular problem class, while fostering interactions that transfer ideas between research areas related to algorithms, data structures, implementation, and applications. This 12th DIMACS Implementation Challenge will be held in honor of David S. Johnson on the topic of Vehicle Routing Problems.
The Vehicle Routing Problem (VRP) and other related dispatch problems have been widely studied for over fifty years because they are of both practical relevance and theoretical interest. Designing efficient routes for vehicles performing distribution or service functions translates directly to cost savings, making vehicle routing a topic of great commercial interest. Moreover, the fact that it generalizes the Traveling Salesman Problem, but is substantially more difficult, has kept it in the sights of theoreticians for decades. The VRP exists in a myriad of variations that arise from practical considerations like vehicle capacities, delivery time windows, delays in road networks, and the ability to split deliveries.
Because of the expansive problem space, this Challenge considers multiple VRP variants, representing a mix of classic VRP variants and newer variants inspired by practical considerations. The Challenge addresses problem variants that include some "classics" and some of the most challenging of the VRP family. These problems focus on features that are critical to bridging the gap between application and practice, but they lead to different structural characteristics favoring different solution approaches.
The list of variants is:
This workshop will conclude the Challenge. It will take place online April 5-8, 2022 from roughly 9:00 AM (EDT/UTC-4) to 1:00 PM (EDT/UTC-4). Each day will be comprised of an early session (approximately 9:00-11:00 EDT) and a late session (approximately 11:00-1:00 EDT) and will include a keynote talk, as well as talks by selected teams that competed in the Challenge.
After you are registered, you can click here (or the picture) to enter the virtual venue.
For information please:
If you have questions for the organizers please send email to the organizing committee. (dimacs_vrp_challenge_committee@email.rutgers.edu)
View the VRP Challenge flyer.
Challenge organizers Mauricio Resende and Eduardo Uchoa, together with Luciana Buriol and Celina Figueiredo, recently published The Guide to NP-Completeness is 40 Years Old: An Homage to David S. Johnson, which talks about Johnson's contributions to NP-completeness and experimental analysis of algorithms, including the Implementation Challenges.
Tuesday, April 5, 2022
Welcome from Organizers
Keynote: A Tribute to David S. Johnson
Mauricio Resende, Amazon
Introducing the Ride-hailing Platform
Nicholas Kullman, Amazon
A Novel Repositioning Strategy for Ride-Hailing Problems
Christian Ackermann, University of Hildesheim
CorailRPC-DQN: a Deep Reinforcement Learning and Heuristic Approach for the Dynamic Ride-Hailing
Augustin Parjadis, Polytechnique Montréal
Break + Q & A/Discussion
Iterative Fast Optimization for the Capacitated Vehicle Routing Problem
Luca Accorsi, University of Bologna
Hybrid Genetic Search With Ruin-and-Recreate
Martin Simensen, SINTEF
Eduardo Queiroga, INRIA
An Improved Hybrid Genetic Search with Data Mining for the CVRP
Marcelo R. H. Maia, Universidade Federal Fluminense
Q & A/Discussion
Wednesday, April 6, 2022
FHDSolver: Fast and Effective VRPTW Solver
Shibiao Jiang, Huawei Cloud
Hybrid Genetic Search for VRP with Time Windows: a High-Performance Implementation
Wouter Kool, ORTEC
Adaptive Large Neighborhood Search for Vehicle Routing Problems
Stefan Røpke, Technical University of Denmark
35 years of Metaheuristics for the Vehicle Routing Problem: A Retrospective
Michel Gendreau, Polytechnique Montréal
Break + Q & A/Discussion
A Three-Stage Matheuristic for Multi-vehicle Inventory Routing Problem
Shihao Huang, Huazhong University of Science and Technology
A Branch-and-cut Method with Warm-start for the Inventory Routing Problem
Jørgen Skålnes, Norwegian University of Science and Technology
A New Branch-and-cut Algorithm for the Inventory Routing Problem
Thiago Guimaraes, Federal University of Paraná
A Network Simplex based Matheuristic for the Inventory Routing Problem
Bruno Castro, Pontifical Catholic University of Rio de Janeiro
Q & A/Discussion
Thursday, April 7, 2022
FHCSolver: Fast Hybrid CVRP Solver
Shibiao Jiang, Huawei Cloud
Jiongzhi Zheng, Huazhong University of Science and Technology
AILS-II: An Iterated Local Search-based Algorithm for the Capacitated Vehicle Routing Problem
Mariá C. V. Nascimento, Federal University of Sao Paolo
Hybrid Large Neighborhood Search for CVRP and VRP with Time Windows
Stefan Voigt, Catholic University of Eichstätt-Ingolstadt
Break + Q & A/Discussion
Keynote: Last-Mile Deliveries at Amazon
Renato Werneck, Amazon
Dual-level Local Search Memetic Algorithm for the Vehicle Routing Problem with Time Windows
Yunhao Li, Huazhong University of Science and Technology
A Vehicle Routing Heuristic Based on Accelerated LKH-3 Coupled with Set Partitioning
Francesco Cavaliere, University of Bologna
A Hybrid Algorithm for the Vehicle Routing Problem with Time Windows
Carlos Neves, Federal University of Paraíba
Q & A/Discussion
Friday, April 8, 2022
An Efficient Iterated Local Search Heuristic for the Split Delivery Vehicle Routing Problem
Weibo Lin, Huawei Cloud
Exact and Heuristic Methods for the Split Delivery Vehicle Routing Problem
Stefan Røpke, Technical University of Denmark
Keynote: Routing Problems: Yesterday, Today, Tomorrow
M. Grazia Speranza, University of Brescia
Break + Q & A/Discussion
General Edge Assembly Crossover Driven Memetic Search for Split Delivery Vehicle Routing
Pengfei He, University of Angers
Solving the VRP with Split Deliveries
Simen Tung Vadseth, Norwegian University of Science and Technology
Redoute - Pruning Instances of E-VRP-NL Problems
James Fitzpatrick, University College Dublin
Break + Q & A/Discussion
The Implementation Challenge workshop is open to all who register. Registration is free, but you must register to be admitted to the online venue. Presentations at the workshop are by invitation and will largely be given by teams that competed in the Challenge. (Please visit the Challenge website for details on the competition.)
After you are registered, you can click here to enter the virtual venue.
Please note that you may not be able to register once the event has begun.
Presented in association with the Special Focus on Bridging Continuous and Discrete Optimization and the Implementation Challenges.