Benchmarking of Meta-heuristics in RCPSP with Mandatory Pauses under Random-Key Mapping

Vol 57, 2025 - 340412
Poster
Favorite this paper
How to cite this paper?
Abstract

The Resource-Constrained Project Scheduling Problem (RCPSP) with mandatory breaks brings the modeling closer to everyday real-world situations, such as ergonomic pauses or rest shifts for workers in construction and manufacturing. This dynamic introduces variability in workforce capacity, increasing the complexity of finding a feasible schedule. This paper compares the performance of metaheuristics based on continuous representation: Particle Swarm Optimization (PSO), Cuckoo Search, and three GRASP variants. Solutions are encoded via Random-Keys and decoded in discrete time, with an objective function based on normalized resource idleness. The computational evaluation employs 750 instances with different topological complexities. The results, validated by the Wilcoxon test, indicate the superiority of PSO, due to its efficiency in exploring rugged search landscapes. Cuckoo Search shows intermediate performance, while GRASP exhibits severe stagnation, a behavior suggesting limitations associated with the misalignment between the continuous representation and its discrete local search structure. Additionally, precedence bottlenecks (hourglass effect) increase computational cost, highlighting the impact of these constraints on decoder performance.

Share your ideas or questions with the authors!

Did you know that the greatest stimulus in scientific and cultural development is curiosity? Leave your questions or suggestions to the author!

Sign in to interact

Have a question or suggestion? Share your feedback with the authors!

Institutions
  • 1 Universidade do Estado do Rio de Janeiro
Track
  • MH – Metaheurístics
Keywords
RCPSP
Random-Key
Metaheuristics Benchmarking