A case-injected genetic algorithm for single machine scheduling problems with release time

A hybrid approach of genetic algorithms and local optimizers in cell loading

The increasing market fluctuations and customized products demand have dramatically changed the focus of industry towards organizational sustainability and supply chain agility. Such critical chang ...

Bio-Inspired Self-Organisation in Evolvable Production Systems

ALHAWARI, OMAR IBRAHIM, M.S. November 2008, Industrial and Systems Engineering Operator Assignment Decisions in a Highly Dynamic Cellular Environment (130 pp.) Director of Thesis: Gursel A. Suer Operators are assigned to operations in labor-intensive manufacturing cells using two assignment approaches: Max-Min and Max. The major concern is to see how these two approaches impact operators’ skill levels, makespan and total processing time among cells. The impact is discussed under chaotic environment where sudden changes in product mix with different operation times are applied and under non-chaotic environment where same product mix is run period after period. In this thesis, operator’s skill levels are affected by learning and forgetting rates. Max-Min improved operators’ skill levels more significantly than Max in this multi-period study; particularly in chaotic environment. This eventually improved makespan and total processing times as well. Approved: ___________________________________________________________ Gursel A. Suer Professor of Industrial and Systems Engineering

/pdf/operator-assignment-decisions-in-a-highly-dynamic-cellular-uq8w8u2xmh.pdf

Operator Assignment Decisions in a Highly Dynamic Cellular Environment

Seeding is a technique used to leverage population diversity in genetic algorithms. This paper presents a quick survey of different seeding approaches, and evaluates one of the promising ones called the Seeding Genetic Algorithm. The Seeding GA does not include mutation, and it has been shown to work well on some GA-hard problems with binary representation, such as the Hierarchical If-and-Only-If or Deceptive Trap. This paper investigates the effectiveness of the Seeding GA on two problems with more complex non-binary representations: capacitated lot-sizing and single-machine scheduling. The results show, with statistical significance, that the new GA is consistently outperformed by the conventional GA, and that not including mutation is the main reason why. A detailed analysis of the results is presented and suggestions are made to enhance and improve the method.

Experimental study of seeding in genetic algorithms with non-binary genetic representation

In this paper, first manufacturing scheduling is briefly discussed and later the problem studied is introduced. The optimal solution to minimizing the average flow time in single machine scheduling is obtained by the Shortest Processing Time rule if ready times are zero for all jobs. In the case of non-zero ready times, preemption plays a key role in the solution. Preemption allowed version is solved optimally by using the Shortest Remaining Processing Time procedure. However, the version of preemption not allowed is known as NP-hard and delay and nondelay strategies might be used in a hybrid fashion. This paper focuses on minimizing the average flow time in the presence of non-zero times and when preemption is not allowed. The proposed method is evolutionary programming (EP). The results indicate that EP produces near optimal and consistent results in a short period of time.

Joel Santos

Papers

Evolutionary programming for minimizing the average flow time in the presence of non-zero ready times