Stuart Ockman is president of Ockman & Borden Associates, project management consultants based in the Philadelphia area. He started his career at Bechtel working on power projects but now specializes in project planning and control, claims management and claims avoidance. He talked recently with ENR Deputy Editor Richard Korman about the state of construction scheduling, what makes a good scheduler and how schedules are manipulated.
ENR We first met over a decade ago when you and some other old-school experts in critical-path method scheduling raised concerns about how the most popular scheduling software in the industry allowed users to "game" the results to create schedules that suited the users' purposes. Is this still a problem?
Ockman Yes. What's killing the scheduling part of the industry is that scheduling software products are available to anyone at low cost. The users rather quickly become good at using the product but don't know what they've got and don't understand that it's just a tool and the importance of honesty. Construction's a tough industry, with all sorts of agendas, and when it comes down to claims, law firms shouldn't be allowed to hire consultants that will give them any answer they are looking for. Schedulers must be intellectually honest.
How did you learn scheduling?
Some of us were lucky enough to come across a scheduling class, which I took while studying at the University of Michigan. At the time computers were not readily accessible, so I actually learned the theory behind critical-path method scheduling.
What signs are there now that standards remain lax when it comes to scheduling honesty?
I have a problem with the AACE International standard for forensic schedule analysis, which has been in its current form since 2011. The recommended practice gives you a group of different methods and subsets that people have used to do after-the-fact analysis. What it doesn't tell you is that many of the methods described for identifying and calculating delays will give you different answers. And if that's the case, then AACE has done a terrific disservice to the industry by not identifying which method ls giving the right answer.
(Note: John Livengood, president of AACE International, replies that AACE's Recommended Practice on Forensic Schedule Analysis, RP29-R-03, is the only peer-reviewed guide to the practice of forensic delay analysis in the U.S. and one of only three in the world. It recognizes that there are different methodologies that are appropriate in different circumstances and the methods include Windows, As-Planned vs. As-Built, Time-Impact Analysis and Collapsed-As-Built.)
Why is that the case?
Because it leaves the door open to methodology shopping in after-the-fact schedule analysis. Followers of the recommended practice may think that by giving you choices about how to calculate a schedule that it makes doing everything entirely right. The calculations that are wrong are ones where there is an adjusted as-planned analysis, or a collapsed as-built schedule analysis using methods that are not repeatable. If you use one of those you are guaranteed to get a wrong answer, a wrong schedule, unless you are lucky. The right method starts with a reasonable as-planned schedule to use as a benchmark for comparison to what actually happened on the project. The key to making these things work is a chronological comparison of the reasonable plan with what has actually occurred on the project and then making adjustments to the plan to reflect when each critical delay was resolved.
What are some other signs of schedule manipulation?
Too many user-assigned constraints, constraints are a bad thing, they override the normal schedule calculations, you are better off without them.
Constraints are date restrictions imposed on an activity's start or finish date in scheduling software. According to the website planacademy.com, applying a constraint will swap the computed date to a date imposed by the user and they are often used to impose deadlines or to delay activities in a schedule. Will eliminating them solve all the problems?
You don't have to do it only by constraints. You can manipulate a schedule by changing logic ties, leads and lags, shortening activity durations. There are an infinite number of ways to manipulate a schedule, and all are counter-productive and cost trillions of dollars around the world. It's something society ends up paying for.
What's your best idea about how to stop schedule manipulation?
Education. There's a new Project Management College of Scheduling that's independent of the Project Management Institute, and my goal is to have all the top schedulers in the world involved in it.
So what's the best way to find out about it?
Go to the website, www.PMCOS.org, to see what programs we have planned and how you can participate.