KNOWLEDGE BASED CONTROL SYSTEM FOR AUTOMATED PRODUCTION AND ASSEMBLY (EXPERT SYSTEM, SCHEDULING)
Abstract
Modern manufacturing facilities vary more and more from traditional manufacturing facilities. The main concept behind modern manufacturing is flexibility. The need for a short response time to change in demand, increasing variability in products all together with the need for high production rate created the need for Flexible Manufacturing Systems (FMS), as a concept. The only means by which flexibility can be introduced in a manufacturing system is by imposing a computerized control or using Computer Integrated Manufacturing systems (CIM). The computer ties together the requirements from the system, the capability of the system and the control of the system to achieve the designed output. This dissertation describes an experimental investigation into the routing of jobs in an automated production facility using a knowledge based system. The control of an automated manufacturing (CMS) and assembly system is a complicated task. The routing problem of jobs to similar machines is a dynamic problem in a job-shop environment with no closed form analytical solution. In order to control the system a knowledged based system (KBRS) is developed. This procedure is composed of static and dynamic databases, a behavioral knowledge part, a procedural knowledge part and a simulation driver. The system is written in PROLOG and is implemented on a simulated environment. The KBRS system is able to recognize different assembly structures, failures in machines, and adapt its solution to the current characteristics of the environment including a change in the look ahead planning horizon.
Degree
Ph.D.
Subject Area
Industrial engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.