Complexity, Murphy’s Laws and Artificial Intuition in Product Development

The origin of the well known Murphy’s Laws may be traced to Edwards Air Force Base in 1949. A few of the most popular of these laws are listed below:

• If anything can go wrong, it will
• If there is a possibility of several things going wrong, the one that will cause the most damage will be the one to go wrong
• If you perceive that there are four possible ways in which something can go wrong, and circumvent these, then a fifth way, unprepared for, will promptly develop
• Left to themselves, things tend to go from bad to worse
• Everything goes wrong all at once
• Nothing ever gets built on schedule or within budget
• Nothing is as easy as it looks
• Everything takes longer than you think
• It is simple to make something complex, and complex to make it simple

Murphy’s Laws may sound funny but most of us will agree that they correctly reflect the reality more than simple anecdotes. However, we can state that they essentially point in the following direction:

Things tend to become more complex and not simpler

In other words, Murphy’s Laws state that, when given a chance, complexity will go up rather than go down.  In effect, when we say that a “situation is bad” or has “gotten worse” we often imply that it has become more complex. Highly complex situations are difficult to assess and to manage and frequently spawn unexpected behavior and this is why humans prefer to avoid them. In other words, Murphy’s Laws are saying just that. Nobody likes to deal with high complexity.

Somehow, this increase of complexity seems inevitable. On the one side, Nature does it for us, and if this is not the case, we ourselves make sure that complexity increases. We have this incomprehensible urge to pump every possible bit of new, often immature technology, into new products. Just because something is possible doesn’t mean it needs to be done. Almost every modern product is software-heavy, and this increases immensely its complexity, dimishing our capacity to understand why things break, how to fix them, or how to make them better.

Excessive complexity is like an autoimmune disease, attacking a system from within.

Self-inflicted complexity seems to be the inevitable leit-motif , the hallmark of our times. But this is only the beginning. With the increase of popularity of Artificial Intelligence, the introduction of these new technologies into modern products elevates the issue of product complexity to a totally new realm. Such products are increasingly more difficult to test before market launch. It is becoming impossible to debug them completely and it is often up to the customers to do that instead. Basically, the risk is being cast onto the users and away from the manufacturers. When an Original Equipment Manufacturer assembles a high-tech product, containing a myriad of “intelligent” components manufactured by others, how can it guarantee overall quality? It cannot. Remember the Principle of Incompatilibity:

High precision is incompatible with high complexity

In essence, if you build something highly complex you will never be able to make precise statements about it. These include, for example, performance, quality, risk, recalls, warranty and liability costs, maintenance costs, product longevity, side effects, etc. This is of course true for engineered as well as pharmaceutical products. Pity, because sometimes human lives are at stake.

What can be done to counter this? There are basically two approaches:

• A: Adopt complexity as a design attribute (like stiffness, mass, fatigue life, etc.) and develop products trying to reduce their complexity while satisfying the usual design requirements and constraints. This requires being able to measure product complexity and this is what Ontonix does since 2005 via its QCM technology. This is what product development will look like in the future.
• B: Develop a product in the traditional way, neglecting complexity altogether but use QCM technology to identify as many Fragility Hotspots as possible and try to implement remedies. This is a poor man’s approach but it can produce excellent results.

The fundamental reason why it is not a good idea to develop immensely complex products lies in the Principle of Fragility:

In other words, a highly complex product, functioning in an uncertain environment, this includes uncertainty generated by a human operator, not just that of the environment, leads to a context of fragility. High complexity means problems. It is a fact of life.

QCM technology is often referred to as Artificial Intuition. This is because QCM can spot that someting is wrong – in other words, it feels that a Murphy’s Law is about to hit you – without resorting to expensive and long training. QCM doesn’t rely on any form of Machine Learning, precisely because it works like intuition.

Merriam Webster: Definition of intuition

1a : the power or faculty of attaining to direct knowledge or cognition without evident rational thought and inference
b : immediate apprehension or cognition
c : knowledge or conviction gained by intuition
2 : quick and ready insight

This is what QCM is. This is what QCM does.