A Simulated Annealing Approach to Identical Parallel Machine Scheduling

TL;DR

This paper applies a simulated annealing heuristic to the identical parallel machine scheduling problem, effectively minimizing maximum machine runtime with near-optimal solutions faster than exact methods.

Contribution

It introduces a simulated annealing approach tailored for identical parallel machine scheduling, demonstrating its efficiency and effectiveness over traditional exact algorithms.

Findings

Simulated annealing yields near-optimal solutions.

The heuristic outperforms exact algorithms in computational time.

Solutions are comparable in quality to optimal ones.

Abstract

This paper studies the application of the simulated annealing metaheuristic on the identical parallel machine scheduling problem, a variant of the broader optimal job scheduling problem. In the identical parallel machine scheduling problem, $n$ jobs are to be assigned among $m$ machines. Furthermore, each job takes a certain amount of time that remains constant across machines. The goal of the paper is to schedule $n$ jobs on $m$ machines and minimize the maximum runtime of all machines. Both exact and heuristic methods have been applied to the problem, and the proposed algorithm falls in the heuristic category, making use of the simulated annealing metaheuristic. Compared to exact algorithms, simulated annealing was found to yield near-optimal solutions in comparable or less time for all problem cases.