Why Construction Project Timelines Always Seem to Fail

1. Book-experience: in the field of construction scheduling that you can’t cut your teeth on books alone. As I discussed in my last post, a lack of field experience is a disadvantage when it comes down to the nitty-gritty: optimization, mitigation, and recovery schedules. Such, schedulers are over-reliant on input from the project team as they will be be hard-put when tasked with generating a baseline from the project drawings.

2. Ignorance: whether willful or not, many contractors struggle mightily with the business of generating even the most mundane of CPM construction schedules to their clients, merely because they lack proficiency in the science. Some will go through the motions – not surprised when their updates are rejected, with the hope and expectation that once the project is built, the never-approved schedule will be a moot point. This latter is what I would call willful ignorance.

3. Cheapness: contractors who like to keep general conditions on a short-leash may be loath to make the requisite investment in (of all things) a CPM schedule – if they think they can get away with it. Indeed, the schedule may be the first place they plan to cut corners – assuming they had a notion of what was required of it in the first place. If owners don’t hold contractors to even the most basic reporting requirements, then they can expect exactly no more than that. This acquiescence of sub-standard schedules in the industry always comes back to haunt.

4. Amateurism: let’s face it, construction project scheduling software can difficult to master, esoteric, and not for everybody. If I had to hazard a guess, I would say that perhaps 90% of the schedules I review are not ready for prime-time. Although there tends to be project integrity cut-off – where an amateur may no longer flourish as he once did in perhaps less serious times (the rollicking 1990’s), he is still commonplace nonetheless. Smaller contractors are often hard-put to qualify the work of such amateurs, in the way they might judge a carpenter’s framework. That’s because scheduling is an abstract concept to most. Despite there being many talented project managers in the pool, frankly even they are invariably described as scheduling amateurs. After all, scheduler, is not their job-description.

5. Team failure is a top contributing factor baseline schedules tank before their time. Scheduling a construction project should never be the responsibility of the scheduler alone, nor should he be the only one held blameful when the schedule deprecates at a fundamental stage. Building construction schedules requires participation from all contractual parties: it is a collective effort, not an isolating one. That being said, if the scheduler lacks the skills necessary, it won’t matter how good the rest of the team is – they won’t have a scheduler to plot it for them.

6. Omission of the timeline verification phase by vendors, suppliers, and sub-contractors: too many schedulers prepare unilateral timelines without running them by the respective parties. Sometimes I prepare timelines for vendors, suppliers, and sub-contractors, because a schedule is due before their contract is let, but I always have the good-sense to defer to the respective parties before I make adjustments, finalize, and consider the baseline complete. For critical contractual parties, GCs and CMs generally stipulate time-frames, if not contractual schedule clauses that they must adhere to.

7. Float-scatter: high and low float values appear in too many sequences where they should not. Naturally, this sequencing, if left unresolved, would lead to endless stopping and starting, access issues, and coordination issues in general. Few baselines are optimized (minimum free-float without straining resources) before they are published. One look at the float – free and total, column betrays this shortcoming right out of the box. This condition begs for attention, before the inevitable float feeding-frenzy ensues – where everyone competes for the slack-time. The closer you get to substantial completion, the more the owner will vie for the slack-time for his close-out phase; which is perfectly understandable.

8. Lack of Optimization: as I say above, this condition is always evident in the float columns. A seasoned scheduler knows enough to pre-optimize. My short definition of optimization as being a baseline schedule that is compressed as much as possible without requiring extra resources, or creating compression for other activities.

9. Premature Updates: the baseline is not approved in time before the first update: this makes the schedule impossible to track because you must integrate the actualized dates into the baseline – a sort of hybrid schedule. Big problem right? Read at the end of my list for my approach to the hybrid schedule.

        10. Dollars-to-donuts the schedule is an irreconcilable bust after Update 1 because:

1. (see above) there wasn’t an approved baseline to compare it to.
2. Activities will likely be irrevocably out-of-sequence (OOS). Some owners will reject OOS sequence out-of-hand (OOH).
3. I have found the best approach to rectifying (another’s) deprecated schedule is more often than not re-baselining, i.e., starting over.

Managing the Hybrid Schedule

It seems to be the rule rather than the exception, that a baseline is drafted and approved after work begins in earnest, especially with a schedule that has design-deliverables, and other preconstruction activities to plot. How do you reconcile the baseline with the actualized work completed without creating a monster?

The trick is to force the (actualized) dates in the schedule without them being, well – actualized, i.e., showing progress. Here’s how I do it. Let’s say a project must begin site work and foundations before the schedule is approved. I do want the dates to appear in the backward-pass without mucking up the sequences to follow. I can only do that by showing that the activity neither started nor finished.

For this, I merely tweak the lag between activities until the dates agree with those actualized. It’s that easy. Your client should not grouse too much, because those activities will have the highest degree of exactitude in the entire network: they have to, because they represent actualized dates.

Beware the pitfall of forcing too many out-of-sequence OOS hybrid activities. They will give you even more of a headache than they will in future updates, because you are using dynamic (non-started) activities to use lag to drive other activities. If those driving activities are later adjusted, you will have to adjust sequences and lag up the line accordingly.

Good housekeeping practice dictates that you clean-up the lags after you actualize your activities in the first update. Besides, lag always gets the owners (unwanted) attention. The less superfluous information you have in your data-base, the better the quality of your reporting down the line.

Thus, try to pick a relatively static activity – such as notice to proceed, that is less volatile – or subject to change, to use as an anchor lagging your hybrids. I generally don’t like to use lag in this way, as it is somewhat the tail wagging-the-dog, but when a hybrid schedule turns up, there’s often no better choice.