AN EXPERIMENTAL INVESTIGATION OF PRIORITY ASSIGNMENT IN A JOB SHOP,
Abstract
The study principally concerns a discipline in which, at each point of decision, the job with the shortest processing time from among those available for assignment is chosen. Such a pro cedure is known to have optimal properties for certain simple, small-dimensional systems. It is shown in this study to be of dominating importance in networks as well. Regardless of the measure of performance, the simple version of this discipline is among the best procedures of the many tested. Moreover many variations of the basic discipline, and combinations with other disciplines are exhibited, demonstrating the possibility of designing an appropriate discipline for any stated measure of performance. The magnitude of the difference in performance between this shortest-operation discipline and the nominal standards of comparison--such as ''first-come-first-served'' or ''random'' selection--depends upon the particular measure of performance and such conditions as the level of utilization in the system, but factors of two or more are frequently displayed. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1964
- Accession Number
- AD0429970
Entities
People
- R. W. Conway
Organizations
- RAND Corporation