The Incomplete Solution

Somewhere between wrong and useful

Published: Wednesday, February 22, 2017 - 13:03

The world of systems is very wide and deep, and this column can’t be perfect and all-encompassing. My goal here is to emphasize that solutions based on incomplete models lead to incomplete solutions. I’m not calling them incorrect solutions, just incomplete solutions.

Every problem model is a mental construct. Unfortunately, this means that the problem “reality” and the problem “model” are not identical. The mental construct of the problem model depends very much on the person constructing the model. This is affected by the person’s mental models, heuristics, knowledge, wisdom, and biases. This leads to what I call “the incomplete solution.”

The system model must be as close to the actual system as possible. The problem model must be as close to the actual problem as possible. However, this cannot be done. Thus, the problem model is an incomplete construct. Furthermore, the solution must match the problem construct.; thus, the solution derived from the incomplete problem model is also incomplete.

The concept that a model of the system is required before regulating it comes from Roger Conant and Ross Ashby, who said: “Every good regulator must be a model of that system.”

They identified that “any regulator that is maximally both successful and simple must be isomorphic with the system being regulated. Making a model is thus necessary.” Daniel L. Scholten has stated this in terms of problem and solution as, “Every good solution must be a model of the problem it solves.” And, “Every good key must be a model of the lock it opens.”

However, humans are terrible at creating accurate models of systems due to limitations of their mental capabilities. This idea was put forward by Herb Simon, the great American thinker who won a Nobel Prize for Economics in 1978, with the concept of “bounded rationality.” Wikipedia currently defines bounded rationality as “the idea that when individuals make decisions, their rationality is limited by the tractability of the decision problem, the cognitive limitations of their minds, and the time available to make the decision.” The complete knowledge of all the details, and the consequences of the actions, cannot be known. This indicates that a mental construct of a system is incomplete.

This concept is further echoed by the American statistician George Box, who stated in the proceedings of a 1978 statistics workshop, “All models are wrong, but some are useful.” And, “Remember that all models are wrong; the practical question is how wrong do they have to be to not be useful?”

The notion of cause-and-effect is paramount in the problem-solving process. However, this idea cannot be as simple as that. One can use the idea of cause and effect to determine the complexity of the system. In an ordered system, the cause-and-effect is direct, and thus a problem statement is very straightforward. For example, turning the switch does not turn the light on, because the bulb is burned out. Replacing the bulb thus solves the problem.

In a complicated system, there are more layers, and the cause-and-effect relationship is not straightforward. However, with the help of experts and solid problem-solving processes, a good solution can be found. There will be several solutions that can work. The ordered and complicated systems use the approach of hard systems. They are deterministic in nature. An example of a complicated system might be the entire electrical wiring in a house. The cause-and-effect relationship may not be direct for the inexperienced, but it can be established. In the manufacturing world, the processes can be ordered or complicated, and there is a desire for high predictability from their operations.

In a complex system, there are several interwoven parts that make the cause-and-effect relationships murky. There are definitely no linear, cause-and-effect relationships. Here the hard-systems approach cannot be used. Moreover, the problems in a complex system might be messes. One problem is most likely linked to other problems. Russell Ackoff, the great American systems thinker, called this a mess. Ackoff said, “Managers are not confronted with problems that are independent of each other, but with dynamic situations that consist of complex systems of changing problems that interact with each other. I call such situations messes. Problems are abstractions extracted from messes by analysis; they are to messes as atoms are to tables and charts.... Managers do not solve problems; they manage messes.”

Thus, focusing on one problem may not show the whole picture. There can be hidden portions not visible to the team. For instance, in soft systems methodology, Peter Checkland advises not forming the problem statement until the rich picture is understood. In his book Systems Thinking (John Wiley & Sons. 2003), Mike Jackson writes that “analysis in soft systems approaches should consist of building up the richest possible picture of the problem situation rather than trying to capture it in system models.”

In ordered and complicated systems, the incomplete solutions may be adequate. In complex systems, this can have unintended consequences. Hard systems are based on a paradigm for optimization, whereas soft systems embrace a paradigm of learning. A good reference quote for this concept is from Dwight D. Eisenhower, who said, “In preparing for battle, I have always found that plans are useless, but planning is indispensable.”

Final words

Incomplete solutions may be adequate in systems where the cause-and-effect relationships are linear and direct. However, in systems where the cause-and-effect relationships are murky and nonlinear, incomplete solutions can have unintended consequences. Moreover, this detrimental impact may not be understood even in hindsight.

Some of the ways we can improve our system models are to:
• Involve the people close to the system
• Go to the gemba
• Encourage opposing and diverse worldviews and perspectives
• Understand that the solutions are incomplete, and thus never “done”
• Build in feedback systems
• Encourage diversity
• Understand long-term thinking
• Understand that the complexity of the solution must match the complexity of the problem. Using a simple checklist or more training as the solution for a complex problem will not work.
• Not go for shortcuts and fast solutions (silver bullets). In some regards, this also explains why silver bullets do not exist. Simply copying and pasting methods (e.g., lean, Six Sigma) without understanding your systems and their problems do not work. It can actually cause more harm in the long run.
• Understand the cause-and-effect relationships
• Stay curious and always keep on learning

The corollary to the incomplete solution is that there is almost always a better solution than the one on hand. Thus, there is always room for improvement.

I will finish off with one of my favorite Zen koans that looks at the dynamic nature of perspectives:

Two monks were watching a flag flapping in the wind. One said to the other, “The flag is moving.”
The other replied, “The wind is moving.”
Huineng overheard this. He said, “Not the flag, not the wind; mind is moving.”

Koans are beautiful because they raise questions in your mind when you hear them. There are no correct or wrong answers to the questions. They are meant to make you think. In this koan, the question might be—what did Huineng mean by the mind is moving? Perhaps Huineng is saying that the two monks’ minds are like the wind and the flag—not settled. The monks are fighting over who is right or wrong. The monks, who should be able to control their minds and focus on a still mind, are letting their minds flutter in the wind like the flag. The reality is that there is a flag, there is wind, and the flag is moving.

Always keep on learning....

About The Author

Harish Jose

Harish Jose has more than seven years experience in the medical device field. He is a graduate of the University of Missouri-Rolla, where he obtained a master’s degree in manufacturing engineering and published two articles. Harish is an ASQ member with multiple ASQ certifications, including Quality Engineer, Six Sigma Black Belt, and Reliability Engineer. He is a subject-matter expert in lean, data science, database programming, and industrial experiments, and publishes frequently on his blog Harish’s Notebook.