2019 started strongly here at Atlas Geotechnical, but almost immediately we found ourselves overwhelmed re-working problems that we thought we had solved. And that re-work distracted us from other commitments, to the point where we nearly landed on one of our project’s critical paths. And of course when we’re working faster than we should small details don’t get checked, like the PE expiration date on a permit drawing, causing more re-work. As soon as we frantically cut off one head, another grows in its place and the project continues to disrupt our workflow. Rinse and repeat. The past 4 weeks have been tough.
It’s increasingly obvious that the problem is not a phase. We’ll never just “get through this;” our workflow is not going to smooth itself out. Something that we are doing, or not doing, in how we approach these fast-paced, complicated problems is preventing solved issues from staying solved. We need to conduct ourselves differently if we want to achieve better outcomes.
Atlas takes on complicated projects. It’s unavoidable that we start our work while data are being collected, and sometimes it’s unavoidable that our partially-complete engineering needs to be set aside in favor of new strategies. Sometimes. Sometimes it’s unavoidable, not always. I’ve noticed that almost all of our frustrating projects never had a credible plan to begin with. We’re re-designing because the initial concept was not fully planned out. “Ready, fire, aim” is not the way to solve complicated problems.
Better, more thorough up-front planning is how we’re going to improve subsequent engineering so that solved issues stay solved, so that the cut-off heads stop growing back and fighting us while we’re trying to do other work.
The oil-and-gas industry, who build some of the most complicated infrastructure in the world, uses an explicit engineering process called Front End Engineering Design (FEED). We’ve participated in a few FEED studies and have seen how investing in up-front planning yields overall cost and schedule savings. Researchers at Delft Technical University wrote up a really excellent overview:
So, we’re going to make an effort to adapt FEED practices to our more interesting projects.
- We’ll assign ourselves more responsibility in the up-front work.
- We’re going to exert more leadership over conceptual designs and means-and-methods choices.
- We’ll host charettes, brainstorming sessions that include the full spectrum of stakeholders and subject matter experts.
- We’re going to identify the likely problems before our customers order materials and mobilize equipment.
And if we’re successful, our first-try engineering solutions are going to stick and our fallback positions are going to deploy smoothly. We’re going to make a proper plan for killing the hydra all at once so we can stop frantically hacking at solutions during construction.
Look for an update in July for how things are turning out.