Page 99 - Untitled-1

78 CRITICAL PATH SCHEDULING

Estimating Task Durations There are several ways to assign a duration to each
task. The most common is to just come up with an estimated time and establish
that as the task duration. Depending on the type of work, these durations may be
in days, weeks, hours, and so on. In each case, these are considered elapsed times.
For example, a task that has been assigned a 10-day duration is expected to take 2
weeks (assuming a 5-day-a-week calendar). It does not necessarily mean that the
task is worked on for all of these 10 days, or that it is a 10 man-day effort. Later, in
our discussion of schedule risk, we introduce the concept of multiple duration es-
timates. But we’ll leave these alone for now.

A second popular method of determining task durations is called effort-driven.
With this method, we enter the total hours to be applied by each resource that is
working on the task, as well as specifying the rate of effort. For example, a wall is
built by two bricklayers, working full time, for a total of 80 hours. The task dura-
tion is calculated as 5 days (40 elapsed hours). If there are multiple resources and
rates of effort, the task duration is determined by the longest assignment.

There are numerous variations and fine points that can be introduced here,
but we proceed with our coverage of critical path scheduling using these two clas-
sic methods.

Task Precedence Relationships and Date Constraints Scheduling, of course, is
deciding when the work will be performed. In a few instances, this timing will be
free of constraints. But this is the exception to the rule. It is more likely that the
timing of the work will be influenced by one or more factors. These may include:

• A date committed by contract or other agreement.
• Dependencies on other tasks.
• Availability of required conditions (weather, space, permits, funding, etc.).
• Availability of materials.
• Availability of labor resources.

The critical path method allows the specification of any and all of these con-
straining and dependency conditions. The normal process calls for the identifica-
tion of task dependencies, followed by the imposition of dates that will force the
calculated timing to be overridden. Let’s examine these options in greater detail.

Defining Dependencies The default dependency is a Finish to Start (FS) rela-
tionship. That is, task B cannot start until task A is finished. However, conditions
may exist where an overlap is possible. For instance, task B can start 2 days after
task A starts. This is designated as a Start to Start (SS) relationship. There can also
be instances where 2 tasks can start independently of each other, but the comple-