Don't get caught in the incident pit

The incident pit is a concept I first read about in one of my mums diving magazines. At that time I realised how useful it is in describing incidents involving divers and how a lot of small things going wrong can result in a major incident.

In essence, the incident pit is something that you gradually fall into. One small thing might not initially go to plan or a relatively minor piece of equipment might break down or get left at home; when this happens you have put your first foot into the incident pit. Initially it is not a problem, but then something else goes wrong - that is another step into the pit. The challenge with the incident pit is that with each step, not only do you get closer and closer to the bottom of the pit and a serious incident, but with each step the sides of the pit get steeper and steeper. It gets more and more difficult to get out the further into the pit you go.

I think there is something to learn from this concept for engineers. I often see engineers taking steps into the incident pit with relatively minor problems, potentially causing much larger problems in the future. To define the incident pit I think that there needs to be two key elements. The first is that the incident pit is caused by a series of problems, any of which would not be a problem on their own, but each one exacerbates the next. The second is that you have to be in a situation where you are incrementally committing yourself. For divers this is relatively easy to understand, the deeper you are diving the more committed you are and the more difficult it will be to solve any problems. I think this is often the case for engineers; as we design and build something, we make more and more commitments and we have more and more invested in something, so just stepping away becomes difficult, if not impossible.

The trick is to know when to stop taking steps into the pit. Another key feature of the pit for divers is that you can always choose to stop taking steps into the pit, but this has a cost, because their dive may have to be cut short. It is the same in engineering, if you choose to stop taking steps into the pit there will probably be a cost.

I want to look at how the incident pits forms in engineering, so I'm going to consider a simple building being designed by a structural engineer. The design starts off well with no major problems and he makes his first submission to the client. They love the design, especially the huge front window that you have given them. At this point you haven't taken a step into the incident pit as nothing has gone wrong, but you have started to get committed.

The next day you get a call from the manufactures of the glass for the centrepiece window. They are no longer able to provide glass quite up to the specification you wanted. This is the first step into the pit. It might not be the designers fault but he does have the choice of whether to continue or not. At this point in time it is not a major problem, even if the glass is not quite up to the expected specification, it should still be strong enough so the engineer chooses to continue.

So the engineer goes and makes a few more submissions to district planners, the contractors who are pricing the project, and utility companies to check the sewers running under the house will be okay. As each one of the submissions goes in, the engineer gets more and more committed to the proposal; backing out gets more and more difficult.

And whilst the project is getting more and more committed, problems keep affecting the large glass window. The original calculations needed to be modified slightly imposing greater forces on the window, new codes of practise are issued which require a better performance of the window, the structural frame has to be modified slightly increasing the span of the window. The engineer is getting caught in the incident pit as he gets more and more committed. The small problems gradually mount up and if the engineer does not pull out soon enough he will be trapped - the design will not be adequate, but too many commitments have been made. All of a sudden the small problems escalate into one big problem.

There is no easy way to deal with the incident pit. You just have to recognise when things are not going as planned and know when it is time to pull out so you don't get trapped. By remembering how these small problems escalate and how easily you can get trapped, it is easier to deal with the problems.