One of the world’s foremost experts on capital project planning and scheduling talks strategy, tactics and the algorithms that will change the way we work
Dr. Gui Ponce de Leon is one of the world’s foremost planning and scheduling experts. His professional experience includes roles as investor, developer, construction manager, program manager, and EPC contractor planner and scheduler. He has continually pioneered innovations in project management throughout his 45-year career and is on a quest to transform scheduling from a task performed by specialists using a “black box” to stakeholder-centric processes that promote collaboration, enhance stakeholder interaction, and inherently result in reliable schedules.
VELOCITY MAGAZINE: Tell us a little bit about yourself, beyond your bio. Where did your career in capital project management begin, and how did your first job impact your thinking about capital project planning and execution?
GUI PONCE de LEON: The first company I worked with was small, and the only way they could compete with the titans was by selling themselves on the cheap. We were building power plants, and the country was in such dire need of power plants that they were offering a $5 million bonus for completing the project on time. That’s how we made money — by finishing on time and getting that bonus. So the schedule became a strategic, competitive advantage.
VM: What a great story. Today most capital projects are late — how did you get them done on time?
GPL: There are two types of project schedules: strategic and tactical. We’ve been mired in tactical schedules forever now, but when I got into scheduling in the 60s, the primary method of scheduling was strategic.
A strategic schedule is done by the responsible stakeholders — we got the key players around the boardroom table and they figured out how to build the project. They produced a high-level, summarized sequence of events, and the detailed tactical schedule came later and built on that strategic foundation. It’s a more nuanced form of project scheduling.
VM: What happened? Why don’t we use strategic schedules anymore?
GPL: When computers came along, we all suddenly became obsessed with data and the strategic schedule fell victim to project management software. Now only the scheduling technocrats can make a schedule. The responsible stakeholders simply can’t keep up with the volume of data in the scheduling software, and so they go MIA. They’re no longer active participants in the scheduling process. The problem is that a scheduler isn’t responsible for delivering the project, so they don’t have the same perspective.
Don’t get me wrong, I have great respect for professional project schedulers. The best ones can still put together a realistic schedule — I call these gladiator schedulers. After 15 or 20 years of wrestling with the 1,000-pound mastodon that is a modern capital project schedule, they become really good at it. That said, whenever you disconnect the scheduling process from the responsible parties, you set yourself up for execution problems.
When computers came along, we all suddenly became obsessed with data and the strategic schedule fell victim to project management software.
VM: So, we’ve moved from strategic to tactical scheduling and divorced the responsible stakeholders from the scheduling process. How do we fix this?
GPL: The heart of the problem is that modern scheduling software uses archaic algorithms and the vast majority of schedulers complete the schedule using deterministic data instead of probabilistic data. A small group of people has been trying to correct this since the 1990s, working to improve predictability by introducing more sophisticated statistical analysis, but they’re continually hamstrung by the archaic Critical Path method, which turns 65 next year.
I highly recommend the book Superforecasting: The Art and Science of Prediction, by Philip E. Tetlock and Dan Gardner. Reading this book was a seminal moment for me, it transformed my thinking about project planning and execution.
VM: You’re working on new software and scheduling algorithms — can you tell me a bit about your current projects without giving away your trade secrets?
GPL: We are building easy-to-use software that prioritizes strategic scheduling and allows responsible stakeholders to once again become active participants in the scheduling process. It’s next-generation software, intuitive, data-driven, instantly updated, and so on. It will be a five to seven year process to get everything into software, we’ve got seven patents in-hand and we’ll file 15-20 more over the next five years or so. That’s all I can say until our launch!
VM: What impact will the software have on predictability?
GPL: Using the new software, with the new algorithm, we’ll be able to see the true, most likely date of project completion. It will start out with some variability, and eventually we’ll be able to estimate with high probability.
VM: What’s clear is that we know how to deliver projects on-time and on-schedule — you were doing it back in the 60s. What’s not clear is why we continue to use outdated and inefficient project scheduling and execution strategies. Do you have any insights?
GPL: I think it’s human nature. On Day 0 of a new project, we’re 100% focused on the future. The first month goes by, and the future still hugely outweighs the past. The second month goes by, the third, and so on. At some undetermined point, the past starts to contain a goldmine of information that can improve future outcomes, but we’re not in the habit of looking back. I think it’s human nature to believe that you can figure out the future without understanding past performance. Who wants to look back? It feels like a waste of time.
VM: I think that’s a great spot to wrap up. Thank you for your time, this has been such an insightful interview.
GPL: Thank you!
