Metaheuristics for scheduling in industrial and manufacturing applications

Más información sobre el libro

Over recent decades, scheduling has emerged as a pivotal optimization problem and remains an active research area. It manifests in various scientific, engineering, and industrial contexts, adapting to the specific restrictions and optimization criteria of each environment. In optimization and computer science, scheduling is defined as the allocation of tasks to resources over time to achieve optimality in one or more objective criteria efficiently. In production, it refers to the planning of operations to ensure jobs progress through machines optimally according to certain criteria. While a standardized form exists for stating scheduling problems—efficiently allocating n jobs on m machines, each processing one activity at a time to optimize job completion times—scheduling encompasses a diverse array of problems. Several parameters influence the problem definition, including job characteristics (such as preemptiveness, precedence constraints, and release dates), resource environments (single or parallel machines, unrelated or identical machines), optimization criteria (minimizing tardiness, late jobs, makespan, flowtime, or maximizing resource utilization), and scheduling environments (static vs. dynamic, where job numbers and characteristics may change over time).