Carl Chatfield: Top 10 Problems, #6: Don't reign in effort-driven scheduling when it shouldn't apply

Carl here. ProjHugger is for Microsoft Office Project newbies, enthusiasts, and zealots. I publish new posts every Monday morning, but you can add comments any time. This week’s ProjHugger post continues the “Top 10 problems new (and not so new) Project users have, and what you can do to ease the pain” series, which started like this: #10, #9, #8, and #7.

Problem #6: Don't reign in effort-driven scheduling when it shouldn't apply

Some things make perfect sense in one context but perfect nonsense in a different context. Effort-driven scheduling is one such thing. It's a powerful feature in Project that you should know well.

The basic idea of effort-driven scheduling (or EDS to its friends) is simple enough: if one person working full-time on one task should take, say, 20 days to complete that task, could two equally able resources complete the same task in 10 days? Depending on the nature of the task the correct answer could range from "Sure!" to "No way!"

In Project 2007 and 2003 at least, EDS is on by default. Project has a task-level setting that you can use to apply EDS or not. That setting is on the Advanced tab of the Project menu/Task Information dialog box, and is the Effort-driven check box. EDS is user-adjustable only for Fixed Units or Fixed Duration tasks, but not for Fixed Work tasks.

There are two hurdles to managing EDS effectively: one conceptual and the other mechanical. Let's consider both in turn.

The recipe says to cook it at 200 degrees for 4 hours, but I went for 800 degrees and 1 hour. Hey what's that smoke?!?

As a project manager, you really have to understand the nature of the work required to complete a task to determine if EDS should be enabled or not. Consider the following:

Task: Neatly stack up a pile of bricks currently strewn on the ground. One stacker resource would require about four hours to complete this task. Could two stackers get-er-done in about two hours? Most likely yes. How about four stackers in one hour? Well, maybe but there are some issues to consider. Will four workers bump into each other in their mad rush to complete the task in one hour? Is the workspace large enough for four stackers? Clearly, there's some time advantage to having more people work in parallel on this task, but it scales only so far. Ten eager stacker resources, for example, almost certainly could not complete this task in 0.4 hours (although mathematically that is the correct duration with EDS enabled and 10 resources assigned).

Or how about this one?

Task: Perform a liver transplant on a single patient. One surgical team can accomplish this task in, say, eight tense hours. Could two surgical teams working in parallel accomplish this task in four hours? No way. The nature of the task doesn't accommodate additional resources.

Project is binary about EDS-it's either on or off per task. You as a project manager need to apply your judgment to answer the question: For a given task does EDS make sense, and if so to what scale? If the answer is not clear to you, by all means ask the resources who will do the work.

Trust me, Project always does the math correctly (but you may not like the answer)

Continue reading at projhugger.

Carl and Tim Johnson are currently updating Microsoft Office Project 2007 Step by Step for Office 2010.