電梯產業生產排程最佳化模型之研究

電梯產業是一種橫跨多種作業型態的產業。在整個建築業的供應鏈上，是扮演設備供應與工程安裝的角色。高單價、生產時間長、交貨日期不確定是電梯產業的特色，同時在現今建築推案講究獨特風格、高辨識度的需求下，已經讓電梯生產模式從庫存生產轉變成為訂單生產的模式，並且配合客戶需求進行高度客制化的設計與製造，增加電梯製造商在調度設計人力、物料需求、生產排程與工地安裝等有限的資源上的困難度。啟發式演算法近年來大量用在解決組合最佳化、生產排程、網路佈建與運輸路徑規劃等問題。其中基因演算法有著平行搜尋、演算架構與問題的相依性低的特性。故本研究希望建立一個基因演算法為基礎的電梯生產的排程模型，配合電梯生產時所評估的生產指標，產生能夠符合電梯製造流程，且能快速計算並具參考價值的生產排程方案，透過實際的訂單資料來驗證排程結果的可行性。

關鍵字

基因演算法；生產排程；電梯產業；系統模擬

並列摘要

Elevator industry has several industries characteristics, as the role of equipment suppliers and installation contractors in supply chain of the building industry. Besides highly dependant on the business cycle the building industry. High price, long production time and date of delivery uncertain are the characteristics of the elevator industry, simultaneously modern buildings require unique style and the high demand recognizable. Elevator manufacturing mode from MTS (make to stock) change to MTO (make to order). Provide design and manufacture of highly customized according to customer demand. Increase the elevator manufacturing in dispatch design workers, material requirements, production schedules and site installation on the limited resources of the difficulties.The range of research in this thesis is elevator production scheduling problem, and the method of Genetic Algorithms is adopted to construct solution models. The genetic algorithm has been applied to the combination optimization problem, network implement and logistic transport planning. However, the elevator production scheduling haven't been researched. This thesis tries to verify the better solution can be found using the genetic algorithm.