Success of the ASAP team at the MISTA 2013 Challenge on "Multi-mode resource-constrained multi-project scheduling"

Scheduling software from an ASAP team consisting of Shahriar Asta, Daniel Karapetyan, Ahmed Kheiri, Ender Özcan and Andrew Parkes wins first place in the MISTA 2013 Challenge on “Multi-mode resource-constrained multi-project scheduling”.

The performance of the proposed approach is significantly better than the other competing approaches, producing the best solution for 17 out of the 20 test instances and performing the best around 90% of the trials across all instances (see final results).
Project scheduling is a common real world optimisation problem, and in the competition multiple projects had to be scheduled whilst taking into account the availability of local and global resources.

The challenge was part of the 6th MISTA international conference on “Multidisciplinary International Scheduling Conference: Theory & Applications”.

The ASAP team, 3 of whom were appointed under the LANCS (S&I) Initiative, won against an initial registration of 21 teams, with 11 teams from around the world going on to submit full entries. Their winning approach combined Monte-Carlo tree search and hyper-heuristic methods with the ability to exploit the computing power of multicore machines.