The ML4VRP Competition has attracted strong participation from the research community, fostering advances in machine learning–assisted evolutionary computation for vehicle routing problems. Below we highlight selected past rounds and their key outcomes.
The 2026 edition of the competition has now concluded. This year’s competition featured two tracks, CVRP and CVRPTW, and brought together a diverse range of machine learning-assisted optimisation approaches.
Congratulations to Fei Liu and the HGS-EoH team, who achieved first place in both the CVRP and CVRPTW tracks.
We also congratulate GNN-PopHGS, led by Maria Rooeintan, and LFJ, led by Maksym Terekhov, for achieving second and third place, respectively, in the CVRP track. In the CVRPTW track, LFJ, led by Maksym Terekhov, achieved second place, while MengerFlock, led by Nuwan Ganganath, achieved third place.
The complete rankings of all participating teams are presented in the ML4VRP@GECCO 2026 poster.
We sincerely thank all participating teams for their valuable contributions to the competition, and we appreciate the support of the GECCO Competition Chairs and the wider GECCO community.
Focus: CVRP and CVRPTW
Resources: https://github.com/ML4VRP/ML4VRP2024
A.Tiwary, A.Mathur, M.Kalra, V.P.Prakash
Md Irtiza, R.Kalaria, A.S.M.Kayes
Focus: VRPTW
Resources: https://github.com/ML4VRP/ML4VRP2023
Team Lead: Shipei Zhou
Team Lead: Robbert Reijnen