This preview shows page 1. Sign up to view the full content.
Unformatted text preview: h other or that crossed over or behind boxes. This made the charts more comprehensible to humans. The history of how scheduling came to be done on computers is an interesting one that we do not have time to develop. However, a few paragraphs that summarize this period of development are in order to help the reader understand the meaning of the frequently used acronyms CPM and PERT. In the mid-1950s, E. I. Du Pont de Nemours & Company formed a working group with Remington Rand Corporation to study the planning and management of construction work for chemical plants. This team developed a mathematical theory of network planning and control called the Critical Path Method (CPM). The working group understood that for large projects this method required considerable computational capabilities. Dr. John Mauchly, of UNIVAC Corporation, one of the early researchers with electronic computers, joined the working group for the purpose of adapting the method to the digital computer. Concurrently, the Bureau of Naval Weapons set up a program evaluation office to provide senior management with continuous appraisals of the POLARIS submarine program. This office sought proposals from industry for the development of a program evaluation system, and in 1957 contracts were awarded to Booz, Allen, & Hamilton and Lockheed. This effort resulted in another mathematically oriented technique that has come to be known as the Program Evaluation and Review Technique (PERT). Although both of these efforts were undertaken independently, they both were based on computing the critical path, a concept that is discussed in the next section. The concepts of CPM and PERT form the basis of what we refer to as simple scheduling. Also during this period, papers were beginning to appear in the literature that described a technique that has come to be called resource leveling. Since that time, better algorithms have been discovered for resource leveling, but in general the problem is known to be intractable. However, it is often possible to Page 181 restrict the problem under a suitable set of assumptions to allow its solution or at least to find a schedule that meets the requirements even though it may not be optimal. A discussion of what resource leveling means and how it is accomplished is presented later in this chapter. Also during this time, computer programs were beginning to be put forward that could do resource leveling under certain assumptions. But these programs were not in general use. By the mid-1970s, though, an extensive literature on the subject had developed, and a number of commercial software products were available that could do resource scheduling (scheduling based on the concept of resource leveling). However, for large projects it could be computationally expensive to try to resource level a schedule in those days. So, for many years, resource leveling was possible but not widely used because of the time and expense involved in producing a schedule that was resource leveled....
View Full Document
- Spring '10