What good are Projects Schedules if they are not realistic at their core? I mean, we ask our Project Schedules to do so much for us. We use them to direct our work, and to validate our progress payments. We use them to resolve disputes over time usage, including claims of delay, acceleration, and loss of productivity. As an extension of our contracts, we use them to evaluate changes in Work Scope as well as changes in conditions of, or approaches to, Project Execution.
In August I wrote a blog entitled “The Role of the Schedule: Coordination or Prediction?” in which I identified a remarkable number of uses for the common Construction Project Schedule, including “as a planning tool, a coordination tool, a communication tool, work organization tool, resource management tool, performance measurement tool, forecasting tool, reporting tool, contract administration tool, cost control tool, marketing tool, financial planning tool, record-keeping tool, and dispute resolution tool.”
Striking the Optimum Balance
It stands to reason that a Project Schedule that is not realistic at its core jeopardizes every one of the above applications. So the central question is: how do we achieve realistic schedules? I can tell you, from decades of experience, that it is much easier said than done. It would be difficult enough for an individual Scheduler to find the perfect balance between the numerous variables that influence Schedule quality.
But, adding insult to injury, we compound that challenge we operate that Project Schedule in a politically-charged atmosphere of conflict and distrust. The Project Schedule is the rope in a tug-of-war. On one end, Owners insist on reviewing and approving the Project Schedule. On the other end, Contractors and their Subcontractors are reticent to reveal their hands, and so they intentionally obscure the details within the Project Schedule.
But let us leave for another day and another blog how to work around and through these political challenges. There is more than enough for us to discuss concerning what the Project Team can do to ensure that the Project Schedule can achieve and sustain intrinsic quality and integrity.
Realism, Above All Else
Before we get into specific suggestions on how to ensure Project Schedule integrity, let me explain why the blog’s title focuses specifically on realism above all other desirable attributes for our Project Schedules. After all, don’t we also want our Schedules to be accurate, timely, coherent, sufficiently detailed, relevant, achievable, and so forth? The answer is that in order to accomplish any of these other objectives the Schedule must be inherently realistic.
Let us never forget that a Project Schedule is nothing more or less than a model of an agreed-upon Project Execution Strategy. Developed through extensive deliberation that culminates in mutual agreement by the Project Team as to its feasibility and practicality, the Project Schedule charts a course for the most efficient performance of Project Work Scope.
Among all of the uses of a Project Schedule, surely the two most important ones are as a forecasting tool and as a coordination tool. In the October blog these two objectives were contrasted to show that when we optimize the Project Schedule for one objective we also reduce its ability to achieve the other objective. You are of course invited (encouraged) to read the referenced blog, but let me here extract a central point it struggles to make. As a prediction tool, we want our Schedule to be as stable as possible. Yet, as a coordination tool, we want our Schedule to be as precise as possible. These two goals are not always compatible with one another.
Owners as well as Executive Management of the Contractor rely on reports emanating from the Schedule to accurately record Project performance, as well as to consistently and credibly predict the likelihood of achieving downstream temporal objectives. Interpretation of such reports invariably boils down to close scrutiny of what are known as KPI’s, or Key Performance Indicators. For a lack of any better method, these executives resort to comparing statistics from prior and current reporting periods, looking for telltale signs of schedule erosion.
The most popular technique along these lines are a variety of trending analyses, with the common interpretation being that constantly shifting KPI’s reflect trouble ahead. In this climate it is understandable why Contractors would want to inject time cushioning to absorb, as much as possible, the natural ebbs and flows of Project performance, thus stabilizing downstream milestone projections.
But the Contractor’s use of the schedule as a coordination tool demands as much precision as possible. Every Activity Duration must reflect the minimum reasonable length of time required to perform the activity’s Work Scope. After all, it is Activity Durations that are the basis for the Critical Path Schedule’s four primary calculated dates: Earliest Start, Earliest Finish, Latest Start, and Latest Finish. Should Activity Durations ever become bloated with extraneous padding, those Earliest Dates adjust outward and set the stage for schedule slippage.
Without getting into a lot of detail but still attempting to explain the previous sentence, imagine two activities arranged in series, each having a one-week duration. Activity B follows Activity A, such that Activity A consumes Week 1 and Activity B consumes Week 2. Now imagine, in the interest of padding the schedule to ensure stability of downstream milestone predictions, we increase the duration of Activity A to two weeks. Now, Activity B would not commence until the start of Week 3.
Granted, the above example is extreme; even the most aggressive padding of Activity Durations would rarely if ever double the duration of an activity. But even a single day here in a single day there applied to dozens and dozens of activities throughout a Schedule can add up to weeks, even months, of schedule padding. And the effect described in the previous paragraph would be precisely the same. That is why, among schedulers, we refer to schedule elongation as a self-fulfilling prophecy.
Which brings us back to the question of (a) whether the schedule should be padded at all and, if so, (b) the best way to do this. As you might imagine there is hardly unanimous agreement within the Scheduling Community as to whether there should be any schedule padding. I am of the camp that believes that a modest and discrete amount of schedule padding is not only helpful in stabilizing the schedule, but is actually a central in achieving the kind of schedule realism that all schedule uses depend upon.
In the end, though, it comes down to a matter of degree. No padding at all creates an overly optimistic schedule, one that is likely unattainable. Too much padding in a schedule typically renders the schedule irrelevant and useless in a very short period of time, as actual progress betters the artificially protracted pace of performance modeled in the padded schedule. And once confidence in the Schedule is lost, it quickly finds its way to a dusty corner in the job site trailer, and just as quickly becomes irrelevant.
So, how much padding is acceptable? And, what are some suggestions as to where and how to place that padding?
Padding, in Activity Durations
Most any book on Project scheduling will advise that Activity Durations should be based on consideration of three key variables: the activity’s Work Scope, the configuration of the performing work crew or team, and the assumed productivity rates for that work crew or team. As a simple example, if a work crew (comprised of a journeyman technician and two laborers) can install 10 widgets per hour, and if the Work Scope of the activity entails the installation of 160 widgets, assuming an eight hour work day this work crew would require two (2) workdays. Hence, we would assign an Activity Duration of two days.
Now further suppose that this activity is wedged between two other activities such that its predecessor completes on Tuesday. If the linkage between activities is a Finish-to-Start, then the Widget Activity would be expected to start on Wednesday and complete by end of work, Thursday. Accordingly, the successor activity would be expected to commence on Friday.
But there is a flaw in the above reasoning, ever so subtle yet extremely significant. Overlooked by virtually all of those textbooks is the simple reality that Activity Durations have profiles of many different shapes, from bell curves to front-loaded to back-loaded to the double-humped. The one profile that is least commonly encountered, and yet ironically the one assumed by most of those textbooks, is a straight-line. That is, the general assumption is that an activity is performed in a steady level of intensity, commencing at the very first moment of the activity and lasting until the very last moment of the activity. In turn, the next (follow-on) activity is assumed to start a millisecond after its predecessor has finished. This is not how things work in the real world.
For any given trade and for any given installation process there is typically a ramp-up period at the start of the activity, as well as a ramp-down period at the end of the activity. In fact, Momentum Science (part of Cognitive Project Management) identifies three distinct stages in every activity.
- Production Stage: This is the primary portion of the activity when the Work Scope is being performed. It is called a “production” stage because effort during this stage produces the Work Scope of the activity.
- Pre-Production Ancillary Stage: This is a typically short “ramp-up” period of time needed to achieve positional and conditional readiness prior to performance during the Production Stage.
- Post-Production Ancillary Stage: This is a short “ramp-down” period of time required to demobilize the activity’s workforce as well as to clear and reset the workspace as a courtesy to the next trade in line to perform work in the same area.
ICS-Research has developed a fairly elaborate understanding of what typically takes place in these two Ancillary Stages. In the Pre-Production Ancillary Stage, first to happen are positional adjustments, where the workforce is moved from some other location into the current work location. In this sense, the workforce is put in a position to do the work. Once relocated to the current workspace, the workers ready the area itself: this is what is meant by conditional adjustments. Tools, materials, equipment, storage facilities, and even orientation to surroundings are part of the conditional adjustments made just prior to performance of the primary Production Stage.
At the other end of the activity, once the Production Stage has finished, the inverse of the Pre-Production Ancillary Stage takes place. The workforce clears the area of debris, removes any remaining materials and equipment, and generally cleans the area in advance of the next trade’s occupation of the same space. Finally, the workforce itself leaves. What we see in the Post-Production Stage, then, are conditional adjustments followed by positional adjustments.
It should be obvious from the previous two paragraphs that the amount of time required to perform the actual Production Stage of the activity is almost always less than the Activity Duration as assigned. For instance, in any ten-day activity, actual production may occur in the middle 8-9 days, with a Pre-Production Stage of 4-8 hours and a Post-Production Stage of 4-8 hours, as well.
Now, let us return to our earlier example of the two-day Widget Activity. The question we need to ask, and have answered, is whether the two-day duration anticipated the Ancillary Stages of the activity, or whether they spoke exclusively to the Production stage. Since virtually all textbooks only discuss productivity rates, crew configurations, and scope of Activity work — it would seem unlikely that the Ancillary Stages were taken into consideration.
And so in light of the above revelations, as well as a few other reasons that are best left for another discussion, that I am comfortable with recommending a practice whereby Activity Durations are increased by a very marginal percent, such as 5%. The 5% is not a hard-and-fast rule, but instead represents an ideological departure from the Conventional Wisdom. In other words, a little bit padding of Activity Durations, in my opinion, adds a certain degree of realism into the schedule that otherwise would be missing.
Padding, in Activity Paths
A very hot Project Management Topic these days is Risk Management. Recommended “best practices” with respect to Risk Management invariably involve the Project Schedule. The range of Risk Management processes and solutions span from a simple listing of major risks that may afflict the project — to a full-blown Monte Carlo regression analysis. But the important point to note about Risk Management is that it is designed to identify, address, quantitatively and qualitatively estimate, and ultimately develop response plans for … major risks (whether opportunities or threats). The operative word is “major.” Because Risk Management processes are fairly sophisticated and costly, they are rarely intended to flush out or prepare for minor incidents that may afflict a project.
Cognitive Project Management, however, does address the steady smattering of minor “incidents” that one can reasonably anticipate to afflict a Project. In Cognitive jargon, they’re called “dilemmas.” The thing about dilemmas is that we can be almost 100% certain that every project will encounter its fair share of unexpected “surprises,” and yet in advance we cannot estimate the number or severity of such dilemmas. How then can we incorporate into our Project Schedule some time contingency as a buffer against any cumulative impacts that might result from unmitigated dilemmas?
Here again there is no universal answer or agreement. My suggestion is to sprinkle strategically placed Dilemma Cushions throughout the schedule. Of course, the Owner would have to agree to such a tactic, because these Dilemma Cushions might otherwise be construed as a float sequestering technique, a tactic usually prohibited in most major construction contracts. But, with the Owner’s blessing, more realism can be built into a schedule by establishing a duration for each Dilemma Cushion that equals 5% of the length of the longest pass feeding into that Dilemma Cushion. For example, if there are three pence terminating at a Dilemma Cushion, and along this path is 10 weeks in length (50 workdays), then the duration of the Dilemma Cushion would be three days (technically, 2.5 days).
Overlapping Activities: Whether to Overlap
Another way to artificially elongated schedule, without resorting to Dilemma Cushions, is to string to activities in series that might otherwise be performable in some overlapped fashion. For instance, with two people in the kitchen, one can wash dishes while the other drives them. Assume Washing Dishes has a 20-minute estimated duration, and Drying Dishes has a 12 minute estimated duration. For all practical purposes, the one drying the dishes need not wait for all dishes to be washed before beginning the drying process. Suppose we have the Dish Dryer wait until dishwashing as reached 50% completion, or 10 minutes into its activity. If we commence Drying Dishes at this point, it will complete two minutes after Washing Dishes finishes. Combined, the two activities would take 22 minutes.
Of course, if we string the two activities in series, where Drying Dishes does not commence until Washing Dishes is completely finished, than the combined duration is 32 minutes. And so, stringing activities in series, that could otherwise be overlapped to some degree, is another way to elongated schedule. My major concern with this tactic is that it does not include a mechanism for documenting the reason why activities were linked in series, rather than overlapped. At least with the Dilemma Cushions there is the opportunity to put a brief explanation in the Dilemma Cushion’s Activity Description.
Overlapping Activities: By How Much to Overlap
The same textbooks referenced above fail to provide any guidance on just how much of a stagger one should create between two overlapped activities. Perhaps the thinking is that the reasons for overlapping activities in the first place are many, and that the extent of such overlaps is entirely dependent on the unique circumstances associated with the two activities.
All of this may be true but I still think some meaningful guidance can be given. A few weeks ago I wrote a blog entitled, “Schedule Logic Dependency Objectives.” The point of the article was to explore why we link activities together in the first place. Understanding why we might link activities together at all is also useful in better managing when, and by how much, we overlap activities. In that blog I discuss Cognitive Project Management’s Four Restriction Linkage Objectives, which are: Mandatory, Natural, Practical, and Logistical Restriction Objectives.
To get a full understanding of these Restriction Objectives, encourage you to read the blog, found at the Continuing Scheducation website. But for the sake of this article let me summarize them here:
- Mandatory Restriction Objectives reflect a mandatory sequencing requirement. For example, a contract may require the completion of Phase II before Phase III.
- Natural Restriction Objectives reflect a particular sequencing of activities as required by the very nature of the work being performed. For example, it is impossible to paint a wall before the wall is first built.
- Practical Restriction Objectives reflect what is deemed by the Project Team to be the most practical approach to the work. For example, the decision to complete floor finishes from the top down in a multistory structure.
- Logistical Restriction Objectives reflect logistical considerations of Project Execution, including procurement, supply, and maintenance of equipment; the acquisition, deployment, and movement of personnel; the availability of and access to workspaces, and so forth.
When we mentally superimpose the above four Restriction Objectives with our new understanding of the Production and Ancillary performance stages that comprise every Activity Duration, we begin to find an answer to our question about how much overlap between activities is appropriate.
Specifically, only the Natural Restriction Objective directly involves the actual performance of the work. In the definition’s example, where we cannot paint the wall until the wall is first constructed, we would want the Start-to-Start dependency to be large enough that the Production Stage of the Paint Wall activity occurs sometime after the Production Stage of the Construct Wall activity has not only begun but has also progressed for a sufficient period of time. [Note: Sufficiency in this context is determined by the particulars of what is natural between the performance of the two activities.]
As for the other three Restriction Objectives, it would seem that the Pre-Production Ancillary Stage might just be independently adequate in meeting their intentions. The most obvious of the three is the Logistical Restriction Objective, which explicitly describes the very conditional and positional considerations of the two Ancillary Stages. It speaks to both the mobilization and demobilization efforts associated with each activity.
As for the Mandatory and Practical Restriction Objectives, both express the will of the parties (Owner or Contractor, respectively) to impose an arbitrary order to the performance of work. Such an ordering of activities would, of course, apply every bit as much to the Ancillary Stages as to the Production Stage.
So, the take-away under this subheading is one that reeks of common sense: we should give full thought to why two activities are being linked to one another, and then let that understanding suggest a Restriction Delay (lead/lag) period that is compatible, supportive, and consistent.
I will close this subheading by noting that, in cases of schedule slippage, among the many possible reasons why is this one: the Restriction Delay was not large enough to stagger the Production Stages of the two involve activities. Imagine a circumstance where the Predecessor Activity requires a two-day Pre-Production Ancillary Stage while the Successor Activity only requires a half-day Pre-Production Ancillary Stage. Now suppose that the two activities are linked by a Start-to-Start with a lead/lag (Restriction Delay) value of two days. In such an example the Successor Activity’s Production Stage would commence at precisely the same time as they Predecessor Activity’s Production Stage.
The devil is, indeed, in the details!
Above, we were talking about Dilemma Cushions, explicitly identifiable activities, which hold modest contingency durations that are meant to safeguard against the random and unpredictable occurrence of “surprises.” Because they are unpredictable, this means that we also cannot associate them with any specific single activity or activity path.
Certain elements of a project are known or believed to contain intrinsic risk that is directly associated with the nature of the work being performed. A good example of this is earthwork excavation, where the amount of subgrade resistance is not fully known (even with the best prerequisite soils testing) until the work is actually underway. Maybe the soil will be sandy; perhaps it will contain a vein of granite.
Another good example often occurs in renovation projects where the extent of corrective work cannot be ascertained until performance of exploratory investigations. For instance, in old buildings with plaster ceilings, until the ceiling is cut open and the conditions in the overhead are visually inspected, it is not possible to know how much repair work will be required.
In situations like these, we can infuse greater realism into our schedule by adding a modest Time Contingency Buffer that (a) is strategically located in the schedule’s logic, (b) has a self-explanatory description, and (c) has a reasonable and prudent Activity Duration that is consistent with the level of estimated risk while not challenging the prohibition against sequestering of Total Float.
Schedule Recalibration Techniques
This last suggestion has to do with how we maintain and utilize our schedules, whereas the previous suggestions dealt with schedule content and integrity. Schedule Maintenance involves performing routine schedule updates in which the achieved progress of Work Scope is posted to the schedule, and any important changes in performance conditions, performance approach, or added/subtracted Work Scope is also noted in the schedule.
It is during these routine schedule updates that I recommend management of any Time Contingency Buffers existing in the schedule. As a practical matter I prefer to transfer any unused Time Contingency Buffer to downstream Milestone Buffers, such that urgency is maintained for all uncompleted work. I apologize for introducing this concept and then not giving it much discussion, but this blog is long enough already. For those interested in learning more about any of the concepts introduced in this blog, please feel free to write to me.
Why You Will Most Likely Be Unable to Apply Some (or Many) of the Above Ideas
As I bring this lengthy blog to an end, I find it necessary to point out that the current contractual climate might not conducive to many of the ideas presented here. Several times I noted that they were pushing right up against the Total Float Sequestering prohibition. But even the idea of freely adjusting buffers of any kind is certain to raise eyebrows in many Owner/Management meetings.
The obvious answer is that, in order for innovative ideas ever to see light of day, those at the highest levels of Project Management must be willing to revisit their preconceived notions of what Project Control is all about. That is why I created this blog site in the first place and entitled it, Thinking Outside the Box. I wanted to open a channel of communication with decision-makers who can change the current climate if they so choose. Here are some ways that they can shake things up (for the betterment of their projects):
- Expect Variances: Currently, there is an obsession with Variance Analyses. Schedulers are required to monitor and report on any variances between actual and planned project performance. But, as I noted earlier in this blog, the vast majority of such comparisons are traceable back to Earlier Dates, which any reasonable person would expect to be missed, given their unrealistically ambitious basis. Message to Owners: quit panicking every time Earliest Dates aren’t met or Total Float values shift by a day or two. Such fluctuations are normal.
- No Fluctuations: I would be much more concerned with schedules that show no fluctuations from one update cycle to the next. Schedules that consistently hold their end dates, or somehow always have the same Total Float value, are extremely suspicious.
- Embrace Changes in Project Execution Strategy: Rather then doggedly resisting any attempts by the Contractor to change his approach to the performance of the Work Scope, why not instead ask for explanations for the proposed changes? Life is unpredictable; it unfolds in unexpected and surprising ways that require each of us to be vigilant in spotting opportunities to better the project’s outcome. When you drive your car, are you not constantly (even mindlessly) inching the steering wheel left and right so as to keep it down the road? Aren’t you constantly on the lookout for opportunities to change lanes, slow down, or speed up, in order to maintain the most efficient and effective pace and direction along your route? The prudent Project Manager is constantly spotting such opportunities, and he/she requires the freedom and autonomy to seize those special moments.
- Poll; Don’t Grill: Use the frequency of the weekly and monthly meetings between the key Project Participants to learn as much as possible about what is happening, what is not happening, and what can be done to exploit project conditions for the betterment of the Project. Create an atmosphere of inquiry and collaborative brainstorming, rather than a dreaded session of Hot Seats and Wood Sheds.
Realistic Schedules Require Prudent Expectations
The title of this blog says it all. We must pursue the creation and maintenance of realistic schedules; that is for sure! But even if we do, there is still the matter of reasonable expectations. As a seasoned veteran of Project Time Management in the Construction Industry, I’m convinced that one of the best things we can do to improve the temporal outcomes of our projects is for those at the highest levels of management, both Owner and Contractor, to hold reasonable expectations.
But more than just being reasonable, these expectations must also be prudent ones. Upper Management must never forget the power of their word: your wish is my command. When Owners express a passing desire for this or that, their subordinates interpret those comments as operational directives. So it isn’t simply a matter of harboring more realistic expectations, it is also a case of more carefully self-scrutinizing and self-censoring what one says. Make sure that you have a fairly good understanding of what you’re asking for, what will be required of others to deliver on your request, and whether your expectations are prudent ones in the context of what is best for the project.