How Project Costing’s Influence Can Destroy Scheduling Integrity

With each passing decade, the Project Cost Control community has become more and more reliant on a fully cost-loaded, network-based Project Schedule, in order to perform its primary cost control functions. With this increasing dependence has come an associated wave of performance criteria aimed at regulating the configuration, content, and quality of data contained within the Project Schedule. The procedural rules and parameters have often had the undesired effect of eroding the Project Schedule as a temporal tool — the project’s only temporal tool. This is too high of a price to pay for Cost Control information that, as it turns out, is of questionable credibility.

The Gradually, Barely-Detectible Encroachment of Cost on Schedule

It took little time for the broader Project Management community to accept, without little to no questioning, the apparent necessity of the Project Schedule to accommodate whatever restrictions and demands might be mandated in the interest of Cost Control. Cleverly, the Project Cost Control community depicted its role and contribution as being essential to  the achievement of fiscal and fiduciary accountability. What responsible Owner, especially in the public sector, would not fully embrace a program that promised:

  • To provide constant correlation between payments made to contractors/vendors and equivalent value received by the Owner;
  • To keep contractors and vendors honest in their invoicing, progress reporting, and dispute posturing;
  • To provide factual, objective assessments of true project status;
  • (And, for contractors) to yield a quantitative picture of resource utilization efficiency?

Encroachment, Not Without a Price

And while the Project Management world is adamant that Cost Control should be performed within the schedule, I am at best a reluctant supporter of this arrangement — and, even then, only on a fairly restricted basis. Let me be clear: I am not opposed to cost processes taking advantage of the power of an automated Project Schedule. Having said that, I nonetheless believe that:

  • Cost Management processes, wherever performed, must not be allowed to negatively impact Project Time Management operations.
  • Cost Control, as a needed discipline, can function just as effectively, from outside of the Project Schedule.

My position on this topic is that the above-listed goals of traditional Cost Control can be accomplished just as effectively from outside the Project Schedule. In fact, I happen to think that the quality and credibility of conclusions reached by cost engineering processes would actually be enhanced, if Costing did not have such a dominant say on how schedules are to be developed, maintained, and used!

Traditional Cost-Oriented Uses of the Project Schedule

This blog has the potential to become controversial in its conclusions. So perhaps I can begin with some points that we might all agree on. For starters, let us consider the value of cost-loaded schedules. To fully appreciate why a cost-loaded schedule can be quite helpful to the Project Team, one must look at it from two different perspectives: that of the Owner and that of the Contractor.

  • Owners use the cost-loaded schedule to generate cash flow projections, to acquire a sense of progress being made on the project, and to validate vendors requests for payment.
  • Contractors chiefly use the cost-loaded schedule to support progress payment requests, to keep track of their own performance budgets, and (less often) to assess cost resource efficiency.

Does Cost Control Require a Cost-Loaded Schedule?

Among senior scheduling practitioners there is a growing opinion that Cost Control processes may be better served through dedicated job costing software, leaving scheduling software to focus on temporal issues without having to endure cost-imposed encumbrances. The chief differentiator is that scheduling software addresses the element of time, whereas cost accounting software is strictly tabular in its calculations. Said differently, scheduling software must take into account the order or sequence of work activities, whereas costing software can sort the column of data in any order and always derive the same totals on the infamous Bottom Line.

The Weakness of Earned Value’s Schedule Performance Assessments

This same distinction can be found at the center of another, related debate: the credibility of Earned Value as a temporal prediction tool. I am talking about Earned Value’s Schedule Performance Index (SPI) which, if we are honest with one another, ignores the interwoven logic of the Critical Path schedule. That is, Earned Value calculations (and conclusions) do not take into account the relative criticality of activities whose resource utilization levels are being measured.

Now, compound this omission with another fly in the ointment; that Earned Value adopts a simplistic (and unproven) assumption that time and cost correlate with one another. The fact of the matter is that the rate at which project dollars are spent has no direct correlation to the amount of temporal progress being made, and this is true for a number of logical, common-sense reasons (especially in Construction):

  1. There is no consistency across the activities of a Project Schedule when it comes to labor crew configurations. Some activities employ two-worker crews, some eight-worker crews, and some are single-worker activities.
  2. There is no consistency across trades as to what a worker gets paid per hour.
  3. There is no consistency across activities as to how many crew days are required to perform an activity.
  4. The cost of some activities is heavily skewed by unusually large equipment or material costs.

I could go on, but let’s just stay with the four items I mentioned. On what dollar basis should the Earned Value’s “earned value” be determined?  Should we use the total cost of activities, including equipment and material, in addition to labor? The fourth bullet points out the distortion created by equipment/material cost skewing. If we use labor dollars only, how do we account for differences in labor rates, composite crew rates?  And if we simply use labor hours (not even labor dollars) in order to avoid the disparate labor rate concern, how do we deal with the varying labor intensity of different activities?

Finally, let us return to my earlier point, that Earned Value is blind to the relative criticality of activities (as measured by Total Float). It is a simple exercise of junior high school complexity to show by way of a “project” with just two activity paths (each with the same required labor hours and yet with significantly different Total Float values) that if the high Total Float path is reported at 100% performed, and the low Total Float path has not yet begun, the project may be reported “on schedule” per Earned Value, yet dramatically behind schedule in terms of Total Float

Where the Encroachment Happens

Earlier I mentioned the encroachment of Costing requirements and parameters that are imposed on the design, development, and maintenance of the Project Schedule.  Specifically, I am referring to requirements that dictate the level of detail of the schedule, the forced use of a Work Breakdown Structure, and certain rules that seek to matrix the work breakdown with the organizational breakdown.

Without getting into the nitty-gritty, let me simply say that most of this is entirely unnecessary for Project Time Management purposes alone. It is only in order to accommodate Cost Control interests in the Project Schedule that we are forced to jump through such hoops. My complaint is not simply one of interference and inconvenience. Far more significant, my gripe is that these impositions of rules and parameters and requirements have the unwanted effect of weakening the Project Schedule’s ability to function as a time management tool.

Erecting a Digital Fence

After years of experience and experimentation I can reach no other conclusion than that each interest should stay on its own side of the fence. I happen to think that a cost-loaded schedule can be used to extract data into integrated cost management software — but not backward again in the other direction. [Incidentally, I would make the same argument with respect to tracking schedule performance against key intermediate deadlines. We know that each additional Finish No Later Than date constraint further complicates the tracing of critical paths. So why not export scheduling data into a spreadsheet and compare to desired deadlines outside of the CPM schedule?] The same approach can work for cost budgets. Why not simply export cost data from the schedule into a cost management program, or even into a simple spreadsheet program?

Incongruities with Schedule-Based Cost Management

I have already noted that the cost-loaded schedule, if allowed to over-extend its influence, can easily erode the temporal functionality and reliability of the Project Schedule. This is quite a price to pay for the value of the information gained by cost-loading the schedule in the first place.

But what if the costing information gleaned from the Project Schedule isn’t even all that reliable?  Well, there are plenty of senior scheduling professionals who will agree with me that Earned Value should not be used to opine on temporal aspects of project performance. First, as noted above, the very rules and regulations imposed by Cost Control protocols erode the credibility of the Project Schedule. Second, the use of cost data as a basis for opining on schedule performance is, as also noted above, unsubstantiated.

Has the Day Come to Reverse the Tide?

So there we have it. On the one hand, the overbearing influence of Cost Engineering on the inner workings of the Project Schedule all too often has been instrumental in weakening the Project Schedule as Project Time Management tool. On the other hand, the Cost Control processes that are the primary beneficiaries of the information taken from the injured Project Schedule are themselves known to have identified, and thus far unresolved, shortcomings.

Taken together, these two observations beg us to confront a most obvious question: why do we continue to incubate Cost Engineering from within the Project Schedule? Why not perform these functions outside of the Schedule?

If this topic piques your interest, you may want to read The Great Divorce: Cost-Loaded Schedule Updating,” a paper presented by John Orr at the 2011 AACE International Annual Conference.

Leave a Reply

Your email address will not be published. Required fields are marked *