AN ANALYTIC MODEL FOR PRODUCTION SCHEDULING IN A JOB SHOP
Abstract
The thesis presents a general methodology for determining the optimal production schedule for a job shop. A model of the production process in a job shop is first considered. The job shop is characterized as a single firm, producing a variety of non-standard products to customer specifications. The model provides a homogeneous monetary measure for the inherently non-homogeneous product of the job shop. The value of the output of the shop is measured in dollars of revenue per unit time since the individual jobs to be produced specify the resources required for their production as well as their value upon completion. The profits of the shop depend upon the extent to which lateness penalties and the fixed costs of the shop carried by each job may be minimized. The production schedule specifies both of these. The number of possible schedules for a job shop (suitable specified ) is finite, and depends upon the number of permutations of the jobs to be scheduled. The number of alternate feasible solutions to a scheduling problem is often so large as to make complete enumeration uneconomic. Work on scheduling problems of this nature has heretofore concentrated on finding algorithms which would find optimal schedules in a finite number of steps. The method adopted here is, instead, based on the Theory of Discrete Optimizing, a random-search procedure which yields solutions with controlled accuracy. The methodology is computer oriented and in corporates the computational resource into the resource set of the job shop as a catalytic agent. The methodology is applied to a variety of problems using randomly generated jobs to simulate demands upon the jobs shop as well as the characteristics of the production process .
Degree
Ph.D.
Subject Area
Commerce-Business
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.