What Went Wrong with Maintenance?

What went wrong with Maintenance theories and models?

I think -- almost everything in Maintenance Engineering is simply wrong. Most of the tools and methods are inappropriate. This is because the theory behind those methods and tools is not applicable to maintenance engineering. Unfortunately, this led to 'maintenance' earning a bad name -- 'necessary evil.' I didn't have the courage to say this any earlier. But now I do.

What really went wrong?

There are two approaches we can take to understand something -- a) Static b) Dynamic

Maintenance Engineering is a dynamic phenomenon.

Any dynamic can further be broken down to -- a) Simple (linear) b) Complex (non-linear and chaotic)

Maintenance Engineering falls in the complex domain.

This is where the problem begins.

Complex systems or the subject of Complexity can be further sub-divided into two definite branches, which are as follows:

A) Detail complexity (snapshot approach)

B) Dynamic complexity (the process of changes).

Maintenance, as far as I understand, is about the interpretation and explanation of dynamic complexity to clearly bring out the levers for actionable changes.

Most of the present thinking in Maintenance Engineering represented by its associated theory, methods and tools are definitely in the region of 'detail complexity' and do not address 'dynamic complexity.' Hence we go round in circles. We solve the issue of dynamic complexity we go on creating more and more solutions, which when put together makes the whole situation more complex and therefore unmanageable. A simple examination of an evolving maintenance plan, often known as the PM plan will reveal the truth of this inherent flaw in addressing failures through 'detail complexity.'

'Detail Complexity' definitely helps to create a snapshot showing how a system works at a moment in time. That in turn helps to rearrange the elements of that system into a more ideal picture. But in doing so we keep creating more and more complex solutions to an already complex situation.

Hence this approach of "detail complexity", is no different from conventional static systems analysis tools. It offers no understanding of how the problem we have today have developed over time, especially if the causes are non-obvious. Nor will it help in our understanding of the likely consequences of our future actions.

The big danger with the present approach is -- we might take actions that make things better today but worse tomorrow.

But our goal in Maintenance is to depict dynamic complexity to avoid or eliminate failures and their consequences. This can be done by understanding the process of changes that are happening over time and how to understand and find leverage points to reach a dynamic equilibrium position to make things better than what it is today both in the short and long term.

There is another aspect to this. To paraphrase Albert Einstein our present problems can't be solved at the level of thinking at which they were created.

This is significant. When we design machines it is always done with a two-parameter approach. For example, if we are designing a shaft we might consider load or torque to determine the basic diameter of the shaft. Then if we want to make it better we might also consider the torsion and introduce torsional rigidity in the shaft. Clearly, it is a two-parameter approach that we take to introduce the features of each element of a machine and then manufacture and assemble a host of such elements to give form to a functioning machine.

While Machine Design is always taking a two-parameter approach, Maintenance on the other hand is at least a three-parameter problem. For example, the same shaft we designed well enough considering the various combination of two parameters might fail because of a combined interaction with 'feedback loops' of torque, torsion, bearing wear and coupling unbalance (four parameters acting together). Unfortunately, the interaction along with their feedback loops of three or more parameters is never considered while designing and manufacturing a machine. Clearly, problems of Maintenance Engineering can't be addressed by the level of thinking at which the machine was created. It needs a different level of thinking.

So, the present approach and understanding of Maintenance must be rejected and a new approach based on 'dynamic complexity' must take its place if Maintenance Engineering is to be treated and respected as a 'Profit Centre.'