Back to Basics: Chains and Challenges

William R. Duncan   Some hot topics: critical chain project management and outsourcing recommendations. I’ve chosen to address some potentially controversial topics this month: critical chain project management and dumb decisions by smart people. Q.   I keep hearing that "critical chain project management" is the wave of the future. Your thoughts? Are there some good ideas being put forth by the advocates of CCPM? Yes. Are they new and innovative ideas? Not as far as I can tell. CCPM has its antecedents in something called the Theory of Constraints (TOC). TOC borrows heavily from systems dynamics (developed by Jay Forrester at MIT in the 1950s) and from statistical process control (which dates back to World War II). The two key tenets — think in terms of systems with complex interactions rather than in terms of unidirectional flows, and find and fix the big problems first — are absolutely the right approach, but they are also old news to experienced project managers. In applying TOC to project management, CCPM follows the same trend—some good ideas presented as new insights. Here are a few of the good ideas: make sure your schedule reflects resource availability and not just activity dependencies; don’t let people work on activities that aren't ready to be worked on; and try to keep resources working on one activity at a time to avoid the overhead of context switching. But they aren’t new—all of them are well-documented in the project management literature. Presenting old ideas in a new light is a time-honored approach that can be found in everything from ancient religious texts to Reengineering the Corporation. Some proponents of CCPM, however, have advocated for the "goodness" of their approach by depicting current project management practices as "bad." I think that has the potential for causing some organizations to throw out the baby with the bath water. For example, one advocate for CCPM argued that scheduling activities on their early start date would divert the attention of the "project manager and team from what must be done." Just not true. Unless the network logic is improperly defined, scheduling activities to start on their early start date is absolutely the right thing to do — it is actually the result of using the critical chain approach! Under the same set of assumptions (full-time resources, resource-constrained scheduling, 50% probability of timely completion, don’t spend your float if you don’t have to), a critical chain schedule will be identical to a CPM "early start" schedule. CCPM also asks for full-time commitment of resources. This approach will, in fact, improve an individual’s productivity, so do it whenever you can. Remember, however, that getting part-time resources assigned to your project full-time means that something else (another project, your organization's day-to-day activities) will suffer. Make sure that these resource decisions are made appropriately. There is no free lunch. I’ve also read material from several CCPM advocates that seems to confuse the process of mathematical analysis with the creation of a schedule. The former facilitates the latter, but they are not the same. Mathematical analysis (whether critical path or critical chain) is a method for calculating early and late start dates so that the scheduler can decide when to schedule each activity. So, if you’re having trouble with your projects, CCPM may merit a look-see, but remember that there is little magic in tools and techniques. Traditional project management, properly understood and applied, will be just as effective. Q.   My employer is thinking of outsourcing our product development department (which includes both project managers and technical staff) based on an analysis from our accounting department that it will save money. I contend that it will cost money, but I don’t seem to be able to get anyone to listen to me. In essence, you accuse smart people (the out-sourcing proponents, your accounting department) of making a dumb decision. My experience is that these situations usually turn out to be the result of the two-sides working with different data. I believe it was Alan Kay of Apple Computer who said "point-of-view is worth 50 IQ points." Try putting yourself in their position. Try figuring out why they are making these decisions. Here are some possibilities: Outsourcing will convert a variable cost into a fixed one. If the economy turns down or you have a product failure, the company may be able to survive better long-term even if it means higher current costs. You aren't communicating accurately. The accountants may think you've questioned their process when in fact you've questioned their assumptions. You may be perceived as self-interested — trying to protect your own hide rather than having the organization's best interests at heart. If that is the perception, whether it is true or not may be beside the point. They may be focused on project costs while you're concerned with life-cycle costs. A common example: less testing keeps development costs down (low project cost) but increases the cost of later fixes, customer ill-will, etc. (life-cycle costs). You might try surfacing the issue by addressing it head on: "I'm really puzzled. I know you guys are smart, and this seems like a really dumb decision. Could you help me understand it?" © 1998 Project Management Partners

Return to Articles

About Us | Consulting | Training | Resources | Home