Normalization of Deviances: What is it?

A year ago I received an e-mail from our DE, asking us to consider a question that poses at the end of this story. Instead of trying to resume the story for you, I rather decided to share it because it is an important topic facing our daily activities as engineers and construction managers. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fifty years ago, NASA astronauts Gus Grissom, Ed White and Roger Chaffee died in a fire on the launching pad during a “live test” of their Apollo 1 capsule. Somewhere within their spacecraft, a spark occurred, perhaps from frayed wiring. This spark combined with the flammable materials inside the pressurized capsule containing 100% pure oxygen to create an inferno that swiftly engulfed the entire crew compartment. By the time rescuers were able to open the complicated access hatch the whole crew had suffocated. How could this happen at an organization with some of the best and brightest engineers and technicians in the world?

In the previously manned space capsules, Mercury and Gemini, engineers and technicians were very cautious regarding the potential for a fire inside the capsule. Flammable materials were religiously tested and prohibited from the capsule if they posed a fire risk. But slowly, as more and more successful missions were flown and “live tests” were performed, this prohibition was informally relaxed. Over time, highly flammable materials were allowed into the capsule as astronauts and technicians added Velcro and netting to keep small items in place. Before long they had an Apollo capsule that was “nearly wall-to-wall Velcro” as described by one engineer. And why was this allowed - this introduction of flammable materials into a 100% pure oxygen environment?; Because nothing bad had ever happened before. The previous Mercury and Gemini capsules had similar pressurized oxygen environments, and they never had a fire. So why would they expect Apollo to be any different? It’s called “Normalization of Deviance.”

 Normalization of Deviance is defined as: “The gradual process through which unacceptable practices and standards become acceptable. As the deviant practice is repeated without catastrophic consequences, it becomes the social norm of the organization.”

 This is what happened at NASA back in 1967. The longer they went without a fire in the capsule the more willing they became to waive their rules against flammable materials. Unfortunately a “normalization of deviance” didn’t happen just once at NASA. It happened again in 1986 with the space shuttle Challenger and again in 2003 with Columbia. Seventeen people in all lost their lives in these three separate, very different but then again very similar incidents.

 So what does all of this have to do with us? Everything. I submit we are an organization much like NASA. We have some of the best and brightest engineers and technicians in the world working here every day. We also have a long history of success, where bad things happen very rarely. But also like NASA or past success shouldn’t lull us into believing we’re always doing things right. We must be mindful of the potential normalizations of deviances around us. I submit there are probably things we have always done a particular way that we shouldn’t be doing. But, since nothing bad has ever happened, we’ve informally decided that it must be ok. “We’ve always done it that way!” isn’t just a phrase we use to be critical of bureaucratic processes, it’s a potential killer.

 So, on this 50th anniversary of the loss of three very skilled, very courageous, very committed, astronauts who were also husbands, fathers, and friends to those around them, I ask you to think about our safety practices. What are we doing that we have done so many times that it has become an acceptable practice even though it’s not safe (or efficient or ethical or….)? Where are our “normalizations of deviances”?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I thought for a second about these questions and found a couple of examples within my experience on the highway construction and inspection industry. In this occasion, I will discuss one of them. Coming to my mind is how comfortable we as construction managers accept construction schedules that are not meeting with minimum requirements or programs that fail to provide essential information needed to monitor and control the schedule efficiently. It seems that most schedules are done to comply with the requirement of submitting a plan, rather than realizing that the schedules are tools that help us take the project to its completion. This normalization of deviance poses a significant risk. The risk of finishing the project later than our stakeholders expect. Some of the many reasons I heard when I have tried to address this issue are; So it is late! What is the big deal? Or so what? All the projects complete construction later than expected and if they do, then Contractors has to pay the penalty or disincentive for not meeting the schedule. The reality is that completion dates in highway construction mater to the traveling public and the officials that funded the project. Commitments have to be met to continue the funding of more projects within a Program. Therefore very often disincentives or penalties do not recover the impact of removing projects within a Program due to failing to meet with a frame time or merely failing to notify in a timely fashion that projects will finish later than expected. Of course, this is not a life or death situations, but I believe them to be significant in our industry because of the trust that the public entrusted to us.

In my life, I have not only reviewed schedules for conformance to specifications and constructibility, but I have also developed schedules for monitoring the construction. Therefore, I understand the complexity, the time consuming, and the cost involved to create a complete schedule. With time I learned, that the amount and the time to build a schedule are relatively insignificant to the benefit of forecasting accurate completions dates. I think that an economical solution to this problem is to learn how to use SPI and Variances prediction as an alternative method or complementary method to monitor if projects will finish or not in time. I am currently writing protocols describing how in the Mobility35 Program we use that knowledge in our industry. I Hope to share soon. In the meantime what others “normalizations of deviances” can you identify?