Sunday, August 31, 2014

Four approaches to problem solving

Car breakdown? Internet not working? Boss or spouse upset? Garbage everywhere? Long commute times? Corruption? Poverty? Everyday, we experience problems at different levels and of varied complexity. Some, like car problem, appear solvable, some like, commute time appear more difficult to solve and some others like corruption / poverty appear hopeless. Nevertheless, problem solving has fascinated mankind for centuries. From Francis Bacon who championed inductive methods to Buddha who championed meditative methods various people have proposed different approaches to solve problems. Here is an attempt to classify these approaches into 4 categories – system centric, problem centric, solution centric and solver centric approach. I will argue at the end that as human dimension associated with the situation gets more complex, we need to use more of solution and solver centric approaches.

To simplify things, I have considered the process of going from problem definition to hypothesis generation (or solution creation). And hence, I have kept the steps of problem definition and the hypothesis validation (experimentation) out of scope. Most of the times, problem solving is an iterative process where problem (re-) definition, solution creation and experimentation get repeated until one is satisfied. Thomas Edison epitomized this method by running tens of thousands of experiments to get his light bulb right.

Four approaches to problem solving:

System centric: Tata Nano is the result of questioning the position, material and various properties of car’s components. For example, engine’s position has moved back from front. Plastic was given a serious consideration as a body material before abandoning it because it didn’t fit in the budget. In the early stages of conceptualization the role of doors was questioned.  In the systemic centric approach, a system is looked upon as a combination of various sub-systems. And then the arrangement of sub-systems (and its sub-systems) gets questioned and changed to meet certain constraints like cost, performance etc. TRIZ methodology and its variants like Systematic Inventive Thinking (SIT) are examples of system centric approaches. For further study, SIT’s five thinking tools (subtraction, multiplication, division, task unification and attribute dependency) is a good place to start.

Problem centric: Malaria has been one of the deadliest diseases for mankind. It may have contributed to the decline of Roman Empire. However, once it was known that it is spread through mosquitoes, its arrest became easier. Bed nets and insect repellents are low cost and yet effective measures to reduce the risk of malaria. Root Cause Analysis helps us find the root cause of the problem and then address it. Techniques like fishbone or Ishikawa diagrams and asking five-why’s have been developed to help analyse root causes. This site shows how root cause analysis is applied to various famous failures in various industries like aviation, healthcare, business, environmental, legal etc. For example, this article presents the application of repeated “Why” questioning to arrive at root cause(s) of devastation due to Hurricane Katrina.

Solution centric: By the time Gyanesh Pandey realized that Jatropha based biodiesels based approach won’t be a viable approach, he had already spent significant portion of his pension fund and five years in iterative experimentation. He was struggling to hold on to his dream of electrifying rural India and had run out of ideas. In a chance meeting, he came across a gasifier based electrification technology being used in forty villages in Bihar. He spent a month-and-a-half at one of the plant studying it and improved it further. This led to Husk Power Systems, a venture Gyanesh co-founded that sells gasifier based power generators that supply electricity several off-grid villages in North East India. Note that Gyanesh’s breakthrough happened when he looked at a solution rather than the root cause of the problem. Hence this approach is called solution centric approach.

As the social system becomes large and complex (like rural India), it becomes more difficult to do root-cause analysis. Nassim Taleb calls such systems high causal density systems. The number of variables which are affecting the system is very high – perhaps thousands, perhaps millions. Hence, to solve a problem, it makes more sense to start with “what is already working in the same setting” (a solution). These are referred to as “bright spots”. The idea is to focus on scalable bright spots. The psychotherapy method called Solution Focused Brief Therapy (SFBT) focuses on bright spots as a starting point. Similarly, an approach called Positive Deviance advocates solving social problems by scaling bright spots. Chip & Dan Heath advocate “Follow bright spot” approach in their book “Switch:  How to change things when change is hard”.

Metaphor (or analogy) is another powerful solution centric method of solving problems. A metaphor tries to create an equivalence between two seemingly disconnected concepts by asking, “Why can’t this be more like that?” For example, McDonalds and an eye hospital (Aravind Eye Care), moon and apple (Newton & gravity), water in the tub and water in the stream (Hellen Keller’s first breakthrough concept) etc.

Solver centric: In 1993 when Dr. Kiran Bedi became IG(Prison) of Tihar Jail, she asked an interesting 
question – Why can’t a jail be more like an ashram – a spiritual retreat for prisoners? Soon, an experiment was conducted in which 96 inmates and 23 jail staff participated in a 10-day Vipassana meditation course conducted within the jail premises. It was a success and since then it has become a regular practice at Tihar. Moreover, it has been introduced in other prisons in India and abroad.

What has meditation got to do with problem solving? A reason why a jail term may not reform a convict because the thought patterns which lead to a wrong-doing are stuck deep inside the mind. Similarly, we are not able to solve a problem because cobwebs of judgments / desire / fear prevent us from seeing the world as it is. The same tape of complaints and labelling is getting played again and again inside our heads most of the time. The only way out is to see clearly the futility of this wasteful involuntary compulsive thinking. All meditation approaches are intended to get one’s attention away from this compulsive thinking. Hence, meditative approaches are solver centric approaches. A practice like dialogue championed by people like David Bohm, Peter Senge and Edgar Schein is also an example of solver centric approach. In a dialogue, the emphasis is placed in clarifying one’s own assumption rather than judging others’ comments.

Which is the best approach? Of course, each approach is relevant in some context and you may need to combine them as well. However, I feel that as the social complexity of a problem increases, the role of solution and solver centric approaches increases. My limited experience of working with organizations over the past eight years tells me that solution and solver centric approaches are under emphasized and under-utilized. Hope to see more people using them in future.