So...so what?? Debate over CCPM gets a verbal shrug from TOC guru Goldratt
Debate Over CCPM Gets a Verbal Shrug from TOC Guru Goldratt
by Jeannette Cabanis-Brewin
PROJECT MANAGERS LOVE to talk, especially about project management. And it's been a while since they had a topic of conversation as intriguing as the management theories of Eliyahu Goldratt, author of The Goal and Critical Chain (see sidebar). And boy, have they been talking! In the pages of PM Network, at PMI ’99 in Philadelphia in October, in articles, letters to the editor, e-mails, paper presentations, and discussion groups, the energy and decibels generated a shock wave that could be felt all the way to Goldratt's native Israel, where the author/guru, now retired from the Avraham Y. Goldratt Institute, makes his home. We reached him by phone to ask him questions drawn in part from issues raised by participants in discussions at PMI ’99.
A physicist's quirky clarity and the passion of a crusader: Eliyahu Goldratt muses on the wildfire spread of his ideas.
Q: Your ideas are hotly debated in the project management world these days.
Look what's happened! People are talking about [critical chain project management] at the biggest annual conference on the subject! This amazes me! But it makes perfect sense, if you ask yourself, where have been the breakthroughs in project management? Gantt charts, PERT charts and critical path method. The last one came out in about 1950 and since then there have been immense amounts of articles published on project management, with essentially no new ideas.
But what really is amazing is the number of companies with multiproject environments that are implementing CCPM. I never thought to see this because, although it's relatively easy to try it out on a single project, because of culture change involved, it's immensely difficult to do it in a multiproject environment. But people are trying it anyway! This brings back my belief in human nature!
Jeannette Cabanis-Brewin (firstname.lastname@example.org) writes on project management issues for the business press on behalf of the Center for Business Practices. A second article based on this interview will appear in the January issue of the CBP's Best Practices Report. Address comments on this article to email@example.com.
“Over 1,000 Case Studies”
Habitat For Humanity. A local chapter of this international charitable organization, which mobilizes volunteers and donated materials to construct low-cost houses for the needy, built a four-bedroom house in New Zealand in three hours, 44 minutes and 59 seconds, using Critical Chain to shave nearly an hour off the previous record, set in 1998 in Nashville, Tenn. Warren Jack, chairman for the planning team, said the 130-member team used MS Project and ProChain software to finish the project “ahead of schedule”—the plan called for the project to take four hours— despite the fact that “Murphy did strike,” causing two tasks to take significantly longer than planned.
Harris Semiconductor. An article published in the January 1999 issue of Midrange ERP magazine details the success of an expansion project at Harris Corporation's Mountaintop, Pa., semiconductor products plant. The plant had already used the Theory of Constraints (TOC) to transform an aging plant into an impressive profit center, growing from zero to 20 percent of the corporation's profits in just four years. The plant was rewarded with a $250 million investment in new facilities. Design and erection of the building, installation of equipment, hiring and training employees, and ramp-up to a 90 percent production rate would typically take about 54 months. “Project Raptor” delivered in just 13 months from kick-off to product sales—accomplishing this despite severe weather delays that cost the project about 40 days. The project was completed three days ahead of schedule and only 4 percent over budget—a cost that was vastly offset by the ability to produce and sell product 40 months earlier than the “industry standard.”
Lord Corporation. This manufacturer of vibration and noise control systems for the industrial and aerospace markets, suffered from poor performance on IS projects. In fact, in the five years prior to implementing TOC/CCPM, every IS project was late, even though the department rejected half of the projects presented to it. Yet from February to November 1998, IS pulled off eight projects using Critical Chain: two came in early, five were on schedule, and one late. Capacity increased by 60 percent and cycle time improved dramatically, while operating costs remained the same. In spite of this, job satisfaction improved, with employees reporting lower levels of stress—something Goldratt maintains is a hallmark of companies that implement his ideas.
Q: How many companies around the world are using critical chain?
No idea … but here are some indications: From talking to AGI Network [the consulting arm of the Avraham Goldratt Institute], about half their work today is in project management. Three years ago it was zero. [Software vendor] ProChain says they have hundreds of active clients. I'm coming across companies that have implemented it just from the books, without using any consultants.
I don't have a feeling about how big a movement it is, but it's worldwide—I just had three beautiful stories from India last week.
Q: Cultural change was one concern raised in a recent letter to the editor of PM Network. Dr. Jeffrey Pinto of Penn State University said he's seen other useful techniques fail because the cultural change needed is hard to sustain.
He is an optimist. Never mind sustaining it, it's hard even to start! The magnitude of culture change is immense. In Critical Chain I concentrated on single projects because my belief was that a lot of time would pass before it was possible to implement in multiproject environments.
With a new theory or tool, usually people want to check it out on one project; but in a true multiproject environment, you can't do a pilot. You see, in a pilot, you take only one project and do it only with dedicated resources. You will get a success if you do this, but everyone will say: “Big deal, dedicate resources to my project and I will also finish early!” So you have to do it on all projects in order to get a meaningful test.
Since software that supports CCPM is scarce, it's being done with conventional software, so a lot needs to be done manually. At some companies I have seen them with everything on the wall and moving things around on the schedules by hand—it's like World War II project management! I am surprised that there are not many more software packages. But the programming is so different; it's more programming how to manipulate the data, not just store and arrange it.
But when people are determined, it doesn't seem to be a barrier. When you explain things very logically and people can relate it to their experience, it's not so difficult. You just need consensus in order to start.
Q: In the case studies that show phenomenal schedule performance using the critical chain method, how do you ascertain that the results aren't at least in part due to the Hawthorne Effect (skewed results caused by personnel knowing their performance is being used in a study)?
On over 1,000 case studies?!
Listen, the one environment where [the Hawthorne Effect] is not really possible is the project organization. Here, the project manager is totally focused on his project; his neck is on the line—his neck can't be even more on the line than it already was. Especially on important projects … and you don't test CCPM on something unimportant. How can you increase the intensity? Many of the stories published are about projects that are so important and so under the gun, there's no way to increase the intensity, the scrutiny people are under.
But even if it does affect the outcome, so … so what???
Q: The pressure to “multitask” has never been greater, as organizations struggle with minimal staffs after downsizing and are hard pressed to hire enough trained personnel in a tight labor market. Aren't your ideas counter to trends in business today?
Oh, it's more than blasphemy! These days, you have a company trying to run multiple projects simultaneously by integrating departments and assigning the same resources to multiple projects, everyone is running around like mad trying to find out three different things about three different projects, everyone is always interrupting everyone else … all of this disappears. So much easier! In an ordinary multiproject environment, you hear people say, “Listen, I have to work so I'm going home, I can't work here.” With critical chain, you don't hear such nonsense. People are claiming they are working less, not more … environment more peaceful and pleasant. You get more done by using logic than by using the whip.
Q: Isn't this just “using psychology” on team members—giving them an earlier deadline to urge them to start sooner?
Any solution contains a major element of psychology. We are dealing with organizations composed of people, after all. To try to have a solution without a major component of psychology is therefore stupid. We cannot separate the mechanism by which we manage the project from the fact that this mechanism is imposed on people and influences their behavior.
Q: At the Project Leadership Conference, you drew a parallel between physics and the art/science of management, saying that “conflicts do not exist in reality—only in our assumptions about reality.” What, in your view, are some of the faulty assumptions inherent in project management?
(1) We assume that the amount of safety [in a realistic estimate] is not enough to buffer against actual glitches. (2) We assume the additive rule is applicable for projects: If one task takes 10 days then a project with two tasks will take 20 days. … That's wrong whenever you have the same resource doing more than one task. This is not new to mathematicians. They know systems in which the additive rule is applicable are very scarce. (3) We assume the only way to finish on time is for each task to finish on time. The project manager shouldn't care when the tasks finish … only when the projects finish. Finish tasks ahead of time, great. Finish late, fine, I'll give you help.
Q: There seem to be two opinions about your ideas among project management types: (1) You are a guru. (2) This is nothing new. One PM authority has said, “If you are applying the concepts in the PMBOK® Guide properly … CCPM hasn't much to offer.” Would you care to comment?
You haven't heard it from practitioners … only from consultants. Anyway, so what? Who cares if it's new or not as long as it works. I hear the same two things from some people: This is nothing new, we said it all along, and It's such a revolution, we won't do it. How can it be both? (Laughs.) Maybe the intuition was there, but there's a huge gap between intuition and verbalization. Until you verbalize it you can't do it.
Old, repackaged, who cares. It's still correct. Go and do it.
Q: Does CCPM make earned value measurement obsolete? If not, how can the two tools be used in conjunction?
Can they be used together? Definitely yes … as to how in the last chapter of Critical Chain, I raised the question “How do you evaluate the financial aspects of the project?” Isn't the way we are doing it today a conceptual mistake? We are trying to measure time plus money in one unit. In environments other than government we can use another factor—interest rate—to represent the value of time. In government this has no meaning, since it's all budgetdriven.
I put the question on the table, and now people are debating it. Perfect!
Anyway, this whole financial issue revolves around how to choose between projects … it really has nothing to do with how to plan and manage projects.
ProChain Solutions Inc.
Scheduling Technology Group Limited
(U.K.) (Offers production scheduling software based on TOC principles)
The Goal [North River Press, 1992], available from PMI Publishing Critical Chain [North River Press, 1997], available from PMI Publishing Leading the Way to Competitive Excellence: The Harris Mountaintop Case Study [Avraham Goldratt Institute, 1999]
Q: How do you suggest a company start—particularly if senior management isn't convinced about CCPM?
People like to blame lack of change on management. Sometimes it's just an excuse. Once they look into the cultural changes needed and know how difficult it is to change anything, they get cold feet.
Take a few copies of the book and give them to top management. Don't assume they won't listen. You go to any large company and ask top management how they'd like it if a large project finishes three months ahead—they will pay anything for that!
Ask for the kind of two-day CCPM seminar that's offered by several companies now. People do need time to sit together with an expert and discuss how to go about it in their particular environment.
Top management has to decide which projects will be frozen, because in order to decay multitasking, you must focus on one single project. They have to accept that even the project delayed until last will finish earlier by this method. It's almost a leap of faith.
Then put the buffer at the end instead of covering your ass, which everyone is so ingrained to do.
Q: What about implementing CCPM as a “project rescue” tactic—is this a valid way to introduce it to an organization?
That's exactly what's happening. Recently I was at a conference in California where a representative from Boeing described a very important project on which it became obvious that the test stage would have major problems. … They decided, since it's dead anyhow, let's try critical chain … and the project finished a week ahead of time, 4 percent under budget, with more content! Here's an example of exactly what you were talking about … a project that's both very important and already beyond hope. Management says: “What do we have to lose? We're dead anyhow. Let's try it.”
Q: You must have a vision of how business and society would change if your ideas were implemented across all organizations. Can you share with us what that “critical chain world” would look like?
A better place to live in, that's all. If the starting point is that all conflicts are only faulty assumptions and we have a practical way to evaporate these conflicts … it's a better world… show me a person who is not suffering from conflicts!
Reader Service Number 012
December 1999 PM Network