基因演算法應用於兩階段平行機台 流程型排程績效之研究 –以預錄式光碟產業為例–

A Study of Mixed Flow-shop with Two Stages and Parallel Machines’ Scheduling Performance Using Genetic Algorithms - An Example of Prerecorded - CD Industry -

莊育修(Yu-Siou Chuang)

指導教授 ： 江瑞清

中原大學/電機資訊學院/工業與系統工程研究所/碩士(2011年)

https://doi.org/10.6840/cycu201100238

摘要

本研究應用基因演算法於，具有兩階段生產型態，且第二階段又分成兩種群組的平行機台混合型流程式生產問題，以最小總完工時間為目標。
驗證模式為第一階段M群組四台機器，第二階段分別為A群組三部機台及B群組二部機台，而產品生產模式為兩種，第一種M到A，第二種M到B。
在預錄光碟產業個案公司的實例驗證下，證明基因演算法可取得比改良式 The Nawaz Heuristic法更低的總完工時間。總完工時間由平均由85330秒降低到75653秒，約降低11.34%，變異數也由18532降低為15665，約降低15.47%，以個案公司月產能1500萬片計算。每月可增加170萬片的產能，並增加1360萬元的收益。

關鍵字

平行機台 ； 總完工時間 ； 兩階段流程行工廠 ； 基因演算法

並列摘要

This research is for two-stage flow shop model. The second stage can be divided to two different parallel machines style and finally to be one mixed process production problem using genetic algorithms. The goal is minimum makespan.
The verify model is as below：
1st stage, there are M group with 4 machines.
2nd stage, there are A group with 3 machines and B group with 2 machines.
There are kinds of process models of production process, one is M to A and the other one is M to B.
Taking Prerecorded-CD industry case company as example and verifying, it can prove genetic algorithms can squeeze much lower total completion time than improved Nawaz heuristic algorithm. The mean of total completion time dropped from 85330 seconds to 75653 seconds, around 11.34%. The variation decreased from 18532 to 15665, around 15.47%. Take 15 million pcs monthly production as example; it can increase 15% capacity per month about 1.7 million pcs production and profits will increase 13.6 million one month.

並列關鍵字

Parallel machines ； Makespan ； Two-stage flow shop ； Genetic algorithms

參考文獻

湯璟聖, 2003, “動態彈性平行機群排程的探討,” 中原大學工業工程研究所，碩士論文。

周文聖, 2009, “兩階段混合式平行機台流程型排程績效之研究 —以預錄光碟產業為例—,”中原大學工業與系統工程研究所，碩士論文。

Campbell, H.G., Dudek, R.A. & Smith M.L., 1970, “A Heuristic Algorithm for the N Job M Machine Sequencing Problem,” Management Science, 16, pp. 630-637.

Chen, C.L., 1995, “An Application of Genetic Algorithms for Flow Shop Problem,” European Journal of Operational Research, 80, pp. 389-396.

Eberhart, R.C. & Kennedy, J., 1995, “A New Optimizer using Particles Swarm Theory,” IEEE International Conference on Neural Networks, vol. 4, pp.1942-48.