連續週期下縱列混製式各產品之生產力分析(I)

Abstract

連續週期下縱列混製式各產品之生產力分析 在多樣產品之生產工廠中，管理者常須進行各生產線運作效率之評估，我們將各項接受評量的產品稱為受評單位(Unit of Assessment, UOA)。績效評量情境依據時間性質與空間性質分類。以時間分有：1. 單週期，2. 多週期。以空間分有：1. 單製程，2. 縱列式之雙製程，3. 縱列式之多製程。交叉混合時間與空間，共有六個子問題，其中單週期單製程的績效評量，以＂共同權重＂的分析結果為本人國科會計畫之成果發表了國際期刊三篇。另外五個標為子問題將於本計劃進行。 各種產品在各製程之多項投入與多項產出均為已知，且中間任一製程的各項產出即為其下一製程的各項投入。在此一多績效評量指標、連續多製程、多產品、多生產週期的系統中，分析各產品在各機台的投入資源與產出量的調整，找出使得工廠的獲利最大的生產策略，將各種產品之重要性，加以排序，作為面對許多不同產品之訂單時，接單以及排程的參考。 共同權重模式的績效基本概念為利用所有UOA整體最少的投入，獲得所有UOA整體最大的產出。本人國科會計畫之成果於國際期刊發表了多週期Malmquist績效評量的方法，本研究之第一子問題另以＂共同權重＂分析單製程多週期的績效評量。第二、三兩子問題以＂共同權重＂分析雙製程及多製程於單週期的績效評量。第五、六兩子問題以＂共同權重＂分析雙製程及多製程於多週期Malmquist的績效評量。

Productivity analysis for the tandem job shop with multiple periods In a production system with multiple processes for producing multiple items, the manager desires to analysis the performance of the items. There is a given set of input and output performance indices for each process. The data of all items for those indices in each process are collected. In the tandem processes, the output indices for the intermediate product of a item in a process are the input indices of the successor process. Further, the data for multiple production periods are also available. This research is to develop a set of procedure to analysis the performance of those products and the entire production system. The performance analysis would enable the manager to rank those products according to their aggregate performance scores then the receiving orders could be ordered also. Since the general problem that we are dealing with involved different factors such as multiple items, tandem process, and multiple production period. The mathematical model and its notations are too complicated to present. We introduce six sub-problems that are special cases of the general problem with certain relaxed factors so that the researchers and the readers would be able to observe the inherent possible research and practical issues for each sub-problem. The initial sub-problem that only multiple items in one production process for a single production period has been solved by Common Weight Analysis (CWA) which has been presented by the author. The second and the third sub-problems of this proposal are the cases that multiple item respectively through two and multiple processes for a single production period. New mathematical models that imbedded with CWA are proposed to compute the set of weight for the input and output indices of the individual production processes. The proper adjustment of the input and output for each product in each process would be suggested to improve its performance. The aggregate performance scores of the items are ordered for selecting the receiving production orders. The problem settings for the first, third, and the fifth sub-problems are the multiple items are produced respectively through one, two and multiple production processes in multiple production periods. The models for the single production period are implemented to obtain the set of weights for the input and output indices in each period. Then, the Malmquist production index that measures the productivity change of each product in every pair of consecutive periods is computed according to the new methods we are proposing. The Malmquist production index is further decomposed into two indices: Self Competitiveness Shift, SCS and Group Competitiveness Shift, GCS. The geometric means of the SCS and GCS for the multiple periods of each item are also computed.