Using DEA for Simulation Output Analysis of an AGV System

• Main Authors: Ting-Yu Li, 李亭昱
• Other Authors: Fu-Haw Liu, Ph.D.
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/08387055764791848541

碩士 === 國立交通大學 === 工業工程與管理系 === 91 === This thesis uses Chance Constrained Data Envelopment Analysis (CCDEA) for simulation output analysis of an automated guided vehicle (AGV) system. The AGV system has three design factors: mean jobs arrival rate, vehicle loading capacity, and number of vehicles employed. Setting their levels, we simulate 96 system designs. Each design is called a Decision-Making Unit (DMU). Each DMU simulated 20 times and data of four output indices are collected. Number of work-in-process and the average job travel time are the two to-be-minimized indices. Average hourly throughput and average vehicle utilization are the two to-be-maximized indices. CCDEA enable us to select the best DMUs without pre-specify the weights of the four output indices. The dynamic behavior of the simulation system is handled by the function of CCDEA.