Y Network Techniques Pert And Cpm

With the exception of Gantt charts, to be discussed below, the most common approach to project scheduling is the use of network techniques such as PERT an CPM. The Program Evaluation and Review Technique was developed by the U^ Navy in cooperation with Booz-Allen Hamilton and the Lockheed Corporation o' the Polaris missile/submarine project in 1958. The Critical Path Method was deve-oped by DuPont, Inc., during the same time period.


In application, PERT has primarily been used for R & D projects, the type of projects for which it was developed, though its use is more common on the "development" side of R & D than it is on the "research" side. CPM was designed for construction projects and has been generally embraced by the construction industry. (There are many exceptions to these generalities. The Eli Lilly Company, for example, uses CPM for its research projects.)

The two methods are quite similar and are often combined for educational presentation. Throughout most of this chapter we will not distinguish between them except where the differences are of direct interest to us. We will write "PERT/CPM" whenever the distinction is not important. Originally, however, PERT was strictly oriented to the time element of projects and used probabilistic activity time estimates to aid in determining the probability that a project could be completed by some given date. CPM, on the other hand, used deterministic activity time estimates and was designed to control both the time and cost aspects of a project, in particular, time/cost trade-offs. In CPM, activities can be "crashed" (expedited) at extra cost to speed up the completion time. Both techinques identified a project critical path whose activities could not be delayed, and also indicated slack activities that could be somewhat delayed without lengthening the project completion time.

We might note in passing that the critical activities in real-world projects typically constitute less than 10 percent of the total activities. In our examples and simplified problems in this chapter, the critical activities constitute a much greater proportion of the total because we use smaller networks to illustrate the techniques.

Before explaining the mechanics of these methods, we must note that their value in use is not totally accepted by everyone. Research on the use of PERT/CPM |11, 24, 25] conducted in the 1960s and early 1970s found that there was no significant difference in the technological performance on projects where PERT/CPM was used and where it was not. This research found, however, that there was a significantly lower probability of cost and schedule overruns when PERT/CPM was used. In our experience, the use of network scheduling techniques has increased markedly in recent years, particularly with the proliferation of project management software packages that are inexpensive and reasonably friendly to PMs who are familiar with the fundamental concepts of PERT/CPM, and who are also sensible enough to avoid trying to construct complex networks by hand.

Recent research (9] finds that a greater use of "project management techniques" (PERT/CPM among a number of others) occurs on R & D type projects, on projects with greater levels of complexity, and on projects with resource limitations, than on other types of projects or those with lower levels of complexity and fewer resource limitations. Unfortunately, this otherwise excellent research did not investigate whether or not the use of project management software influenced the number of project management techniques used. The use of project management software for scheduling projects will be discussed and illustrated in Chapter 10.

Let us now define some terms used in our discussion of networks.

Activity A specific task or set of tasks that are required by the project, use up resources, and take time to complete.

OType 1 ; Type 3 Type 2 ^ Figure 8-1: Three sequential activities,

"Start" "Continue" "Finish" AOA format.

Event The result of completing one or more activities. An identifiable end state octffj curring at a particular time.

Network The combination of all activities (usually drawn as arcs) and events (usuli ally drawn as nodes at the beginning and end of each arc) define the project and thei activity precedence relationships. Networks are usually drawn starting on the lefti and proceeding to the right. Arrowheads placed on the arcs are used to indicate thefi' direction of flow—that is, to show the proper precedences. Before an event can be -realized—that is, achieved—all activities that immediately precede it must be com-pleted. These are called its predecessors. Thus, an event represents an instant in time^' when each and every predecessor activity has been finished. Events themselve: have no time duration and use no resources. They are merely points on the networl conditions of the system that can be recognized. Path The series of connected activities (or intermediate events) between any two% events in a network.

Critical Activities, events, or paths which, if delayed, will delay the completion of the project. A project's critical path is understood to mean that sequence of critical^ activities (and critical events) that connect the project's start event to its finisht^ event. »*'

To transform a project plan into a network, one must know what activities com-; prise the project and, for each activity, what its predecessors (and/or successors^ are. An activity can be in any of these conditions: (1) it may have a successor(s) bu no predecessor(s), (2) it may have a predecessor(s) but no successor(s), and (3)j may have both predecessor(s) and successor(s). The first of these is an activity tha starts a network. The second ends a network. The third is in the middle. Figure i shows each of the three types of activities. Arrows are labeled with the appropriate type numbers. More than one arrow can start a network, end a network, or be in t.Q middle. Any number of arrows can end at a node or depart from a node, as in Figu^ 8-2.

The interconnections depend on the technological relationships describedj the action plan. For example, when one paints a room, filling small holes and crac in the wall and masking windows and woodwork are predecessors to painting thi walls. Similarly, removing curtains and blinds, as well as pictures and picture hooR^

from the wall are predecessors to spackling and masking. It is the nature of the work to be done that determines predecessor-successor relationships.

In the examples above, arrows represent activities while nodes stand for events. This is an AOA (activity-on-arrow) network. Another format for drawing networks is AON (activity-on-node). In this case, activities are represented by nodes and arrows to show the precedence relationships. In AON notation, when there are multiple activities with no predecessors, it is usual to show them all emanating from a single node called "start." Similarly, when multiple activities have no successors, it is usual to show them connected to a node called "end," as in Figure 8-3.

The choice between AOA and AON representation is largely a matter of personal preference. Our impression is that users of PERT favor AOA and users of CPM favor AON, but both approaches appear in the educational literature. Both are also used in commercially available computer packages, though AON is typically used in the most popular PC-based software. AOA networks are slightly harder to draw, but they identify events (milestones) clearly. AON networks do not require the use of dummy activities (defined below) and are easier to draw. Throughout most of this chapter we adopt the AOA format of PERT. In Section 8.4, we use the AON representation that is standard with that method. In this way, the reader can become familiar with both types of networks. This chapter is intended as an introduction to project scheduling at a level sufficient for the PM who wishes to use most commercial computerized project scheduling packages. For a deeper understanding of PERT/CPM, we refer the reader to |4, 12, 15, 17, 19, 37, 40, 55].

Recall the planning documents we developed in Chapter 5. In particular, the action plan contains the information we need. It is a list of all activities that must be undertaken in order to complete a specified task, the time each activity is expected to take, any nonroutine resources that will be used by the activity, and the predecessor activities for each activity. For example, we might have an action plan like that shown in Figure 8-4.

Let us start by assuming the node numbered 1 denotes the event called "START." Activities a and b have no predecessors, so we assume their source is at START (node 1) and their destination at nodes we will number 2 and 3, respectively (Figure 8-5). As explained above, the arrowheads show the direction of flow.

Activity c follows a, activity d follows b, and activity e also follows b. Let's add these to our network in Figure 8-6. Note that we number the event nodes sequentially from left to right as we construct the network. No great damage occurs if we do not use this convention, but it is convenient.

Project Management Made Easy

Project Management Made Easy

What you need to know about… Project Management Made Easy! Project management consists of more than just a large building project and can encompass small projects as well. No matter what the size of your project, you need to have some sort of project management. How you manage your project has everything to do with its outcome.

Get My Free Ebook

Post a comment