This preview shows page 1. Sign up to view the full content.
Unformatted text preview: art date and the last box has the same finish date as the project finish date. This is rarely the case since there may be more than one task on the critical path whose start date is the start date of the project or whose finish date is the finish date of the project. Therefore, it would be more suggestive if the critical path were called a critical subnet or a critical spanning set, since, usually, there will be a family of "paths" that are all critical rather than a unique one. The concept of the critical path has proved to be a very useful concept. Since it consists of a family of tasks within the project that have zero float, it must be dealt with very carefully because there is no room for delay in getting any of the critical tasks started or in completing them. If any one of them finishes even a day late, the total project schedule will slip by at least a day, unless a way is found to complete at least one of the other critical tasks a day early. Conversely, it is clear that if we concentrate our attention on reducing the durations of the critical tasks, then we might be able to reduce the time it takes to finish the project, or at least "buy some insurance" that the project will finish on time. When project staff talk about using CPM, they often are referring to periodically analyzing the critical path. On many large projects, the project schedulers periodically prepare an analysis of the critical path for presentation to the program Page 187 manager. As tasks are completed, the actual finish dates can be inserted into the scheduling system. The scheduling system can use these new dates to compute new start and finish dates for all the tasks that have not yet completed. This may cause the critical path to change for better or worse. If the duration of the total project as computed by the scheduling system continues to grow, it is a bad sign. Some growing and contracting is to be expected. A normal situation is for the total duration to grow a little, and then project management, through analysis and by capitalizing on opportunities that arise, then figures out how to resequence the work and reduce the critical path a little. As a result, the scheduling system can compute a total duration back in the range of where it was before. Then the total duration grows a little more, and again project management finds a way around the problem such as reducing duration of some of the critical tasks, and so on throughout the life of the project. But unrestrained growth of the total project duration is unacceptable. Experienced project managers know they need to be constantly aware of the critical path if they want their performance evaluation and productivity measures to remain acceptable. One of the questions that arises naturally is: How often should the project schedule be recomputed? Clearly, major changes in the scope of the work, major quantification variances, and major productivity trends are reasons for recomputing the schedule. However, experience has shown that periodic recomputation o...
View Full Document
- Spring '10