AIMay 15

Petri Net Induced Heuristic Search for Resource Constrained Scheduling

arXiv:2605.1598350.4
Predicted impact top 66% in AI · last 90 daysOriginality Incremental advance
AI Analysis

For operations research practitioners, it provides a new exact method for RCPSP that is complementary to MIP, performing better on resource-tight instances.

The paper formulates the Resource-Constrained Project Scheduling Problem as optimal search over a Petri net reachability graph and solves it with A* using a consistent heuristic. On PSPLIB benchmarks, it outperforms strong MIP solvers (SCIP, CBC) in both success rate and solve time.

We formulate the Resource-Constrained Project Scheduling Problem (RCPSP) as optimal search over the reachability graph of a Timed Transition Petri Net with Resources, using relative-delay tokens so that scheduling decisions correspond to transition firings in the induced state space. We solve the resulting problem with $A^*$ guided by a heuristic that combines Critical Path and resource-based lower bounds, and prove that it is consistent under our token-based time semantics. Experiments on the PSPLIB benchmarks show that the approach outperforms strong exact Mixed-Integer Linear Programming (MIP) baselines (SCIP, CBC) in both success rate and solve time. Per-instance analysis shows that heuristic search and MIP degrade along independent axes, resource tightness for $A^*$ and formulation size for MIP, with resource strength mediating which solver benefits from scale.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes