工廠排班最佳化模式之研究-以一LED藍寶石基板廠為例

NCUIR > Engineering College > Executive Master of Civil Engineering > Electronic Thesis & Dissertation > Item 987654321/63406

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/63406

Title:	工廠排班最佳化模式之研究-以一LED藍寶石基板廠為例
Authors:	蔡孟緯;Tsai,Meng -Wei
Contributors:	土木工程學系在職專班
Keywords:	LED;0-1整數規劃;最佳化公平指派模式;Lingo;LED;0-1 integer programming;optimized mode of fair assignment;Lingo
Date:	2014-01-21
Issue Date:	2014-04-02 14:49:39 (UTC+8)
Publisher:	國立中央大學
Abstract:	台灣由於產業結構的轉型，鑒於環保意識抬頭及中下游客戶需求，發光二極體（Lighting Emitting Diode，LED）產業急速擴展，其產品並大量運用於交通運輸控制標示設施及綠建築照明上。目前LED藍寶石基板廠大多屬24 小時生產模式，人工經驗排班作業一直是公司常態性耗時問題，因此經常令決策者花費大量時間協調及安排，其排班結果亦非最佳化公平指派。應考慮現實需求及公平性。本研究係依LED藍寶石基板廠作業人員，週期內之日班及夜班勤務需求量之人員排班做為研究範圍，先將現實問題所有限制資源量化，轉換並建構數學模型，並以0-1整數規劃模式為架構，輸入撰寫Lingo套裝軟體進行模擬測試，輸出求解其最佳化公平指派模式，以驗證本研究最佳化模式產生的排程結果，確實優於人工經驗排班，且更具效率及公平性。此外依據各參數之調整進行敏感度分析，藉此求得所有作業人員之平均排班次數，幫助決策者適度規劃及人力安排，以避免勞工超時過勞及違反勞動檢查機關之規定。; Transformation of industry structure, coupled with rising consciousness on environmental protection and demands from downstream customers, is causing the industry for lighting emitting diode (LED) to grow rapidly in Taiwan. These LED products have been used in large quantity on traffic control lights and on illumination for green buildings.Most LED sapphire substrate plant operate under a 24-hour mode of production where shift scheduling based on experience has always been a frequent, time-consuming problem. This causes policy-makers to spend a large amount of time coordinating and arranging. The outcome of this is also not an optimal or fair assignment. What should be included in consideration are actual needs and fairness. The scope of this study is overtime shift scheduling, based on the periodical demands for day and night duties on operators of an LED sapphire substrate plant. All limited resources for realistic issues were quantified, converted, and constructed on a mathematical model. Then, a 0-1 integer programming mode was selected as a framework for inputting and running simulation tests on Lingo software package. The output of this test was sought to arrive at the most optimized mode of fair assignment. The result of scheduling produced under the optimized mode from this study is proven truly superior to scheduling based on experience and is proven more efficient and fair. Moreover, every parameter is adjusted for sensitivity analysis to arrive at the average number of shifts scheduled for all the operators. This helps policy-makers properly plan and arrange human resources to prevent fatigue due to overwork and to avoid violating regulations from labor-force supervisory agencies.
Appears in Collections:	[Executive Master of Civil Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	700	View/Open

社群 sharing

Loading...