The world of project controls – project scheduling and cost estimation, is fraught with misconceptions that are often undetectable to the untrained eye. Even within the industry, an extraordinary level of inexact or fabricated data is presented in lieu of validated timelines and budgets. It is this obfuscation of reality that makes much of the industry unproductive and often untrustworthy.
Inexact, manipulated, or fabricated project scheduling and estimation, may be a factor of ineptitude, willful ignorance, or simply sleight of hand. In any form, it is unacceptable. In the private sector, the policing of such trespasses is largely left to oversight consultants who may have even less proficiency in the science – the blind leading the blind. Public – and especially governmental work, is subject to far more scrutiny, though attention spans can vary.
Certainly copious project scheduling specifications and requirements are far more stringent in public and government work. Such specifications often appear as boilerplate in private sector construction, but they are seldom enforced if understood. For example, the requirement of a cost-loaded schedule is as ubiquitous as it is routinely ignored.
In order to understand what’s wrong with a schedule or cost estimate, working knowledge of what is correct must first be mastered. Without such knowledge, it’s too easy to be misled by flawed or manipulated data sets that have no means or models for comparison.
Estimating
Rounded cost values are allowed for budgeting, but seldom represent calculated criteria, more specifically they point to an absence of true cost values. Lump sums for trades often obscure the fact that breakouts of larger components or activities of those trades are either hidden or not available.
Blind trust in cost guides – such as RS Means are regularly accepted, despite the fact that they are only cost averages for limited production scenarios. For example, 30,000 SF of flat partition doesn’t distinguish inside and outside corners, reveals, or coves in those areas. If gypsum board is to be finished at Level 3 or 5, there is no factor to adjust for the extra labor rate. Finally, change order work is carried simply as a random factor of 1.3 x base contract cost.
Of far more concern is the inexactitude of luxury or high-end estimating. RS Means and other cost guides have a glass ceiling that renders them irrelevant in the face of high-skilled labor and exotic finishes: they don’t really profess to have such knowledge. Estimators must have their own, which is either from first-hand experience, or from subcontractor and vendor cost estimates. If they lack the former insight, they are more likely to cut and paste third-party SOW and costs into their proposals without due consideration.
Project Scheduling
Signs that point to scheduling deficiencies are generally simple to detect. For example, the * denoting a constraint next to an activity start date is used more often to force project logic than it is to input a hard constraint. Leads and lags are also evident in gaps between activity bars.
In the large scheme of things, experienced project scheduling professionals know when a budget or schedule is too aggressive. Contractors who admit the challenge will fare better than those in denial, however, both can benefit from lessons learned: for example, if a project has been running at 65% rate of production, the thought that the contractor could suddenly reverse that trend mid-step – doubling rates to 135% production is counterintuitive, unless a substantial weight of resources or shift-work is planned.
The absence of an adopted project scheduling metric and benchmarking systems is the reason why schedules can flourish in an indeterminate universe of semi-truth and blan approximation. Such systems are widely used by other industries – aerospace, O&G, nuclear, aviation, and aeronautics, but the construction industry is blissfully ignorant of these tools. For example, I once presented a Deltek Acumen validated CPM schedule to an oversight consultant who had given a schedule low marks. I showed that my Acumen indices – FSI and FLI were in the 90th percentiles, and that these two perceptions could not be true at the same time. SInce he had never even heard of Deltek, this fact was a moot point.