Structured-thinking frameworks

A structured-thinking framework is a set of steps or stages that a data scientist can use to resolve a business problem in an evidence-based scientific manner. The three fundamental steps that form the basis of all the structured-thinking frameworks include identifying the problem, finding out the root cause, and creating solutions to address the root cause.

Structured-thinking frameworks are not rigid. They can be modified according to the context of your business problem statement.

Now you’ll learn about five structured-thinking frameworks in detail.

1. The six-step problem-solving model

The six-step problem-solving model is straightforward and can be used to solve any data science business problem. As the name suggests, this technique uses six steps to solve a problem:

1. Have a clear and concise problem description.
2. Examine the roots of the problem.
3. Brainstorm possible solutions to the problem.
4. Explore the possible solutions and select the most suitable solution.
5. Execute the solution effectively.
6. Evaluate the results.

This model follows the theory of continuous development and improvement. Therefore, it enables you to go back to choose an alternative solution in Step 4 if your results don’t solve the problem at the end of Step 6.

2. The drill-down technique

The drill-down technique is suitable for solving highly complex business problems. It works by breaking down the problem to identify the actual root cause. Once you know the root causes of all the smaller problems, you can develop solutions for the bigger problem. To use the drill-down technique, follow the steps given:

1. List all the important problems.
2. Identify the factors causing the problem.
3. Continue the factor identification until you reach the root of the problem.
4. Develop a plan that addresses the root cause of the problem.
5. Implement the plan and review the progress periodically.

Although the drill-down technique does not provide a quick solution to the problem, it is the most comprehensive method that breaks down all the potential causes of the problem to identify the root cause and reach the solution.

3. Eight disciplines of problem-solving (8D)

The 8D methodology is a team-oriented, structured problem-solving methodology used to identify, correct, and eliminate recurring problems. This technique provides a systematic and practical approach to solving a business problem and provides a documented database of improvements.

The eight steps in this methodology are as follows:

1. Build a skilled team.
2. Define the problem by identifying the problem’s ‘who’, ‘what’, ‘where’, ‘when’, ‘why’, and ‘how’.
3. Develop a practical plan to contain the problem.
4. Determine and identify root causes using diagrams and flowcharts.
5. Assess the plan and rework, if required.
6. Implement the revised action plan.
7. Evaluate your results.
8. Appreciate the team’s hard work.

4. The ‘5 whys’ technique

The ‘5 whys’ technique is a popular method of structured problem-solving. This technique follows a simple approach of asking ‘why?’ five times until you get to the root cause of the problem.

For instance, start with the central problem and ask why it happened; then, keep asking ‘why?’ until you reach the source of the problem. However, sometimes you may need to ask more or fewer than five ‘whys’ to get to the root cause. This intuition is something that data scientists acquire with a detailed understanding of the context of the problem and continuous practice.

Example of using the ‘5 whys’ technique

Problem statement: You are driving back home from work, and your car stops in the middle of the road.

The ‘5 whys’ technique application:

• Why did your car stop? – Because the engine broke down.
• Why did the engine break down? – Because I didn’t service my car this year.
• Why didn’t you get your car serviced? – Because I didn’t have time.
• Why didn’t you have any time? – Because I’m so busy balancing work and kids all the time.
• Why are you unable to balance work and kids at the same time? – Because I struggle with multitasking effectively.

In this example, the ‘5 whys’ lead to the root cause of the problem: the person needs a helping hand to manage their work and kids since they are unable to do two tasks simultaneously, resulting in a lack of attention to other tasks.

5. Mutually Exclusive and Collectively Exhaustive (MECE) framework

MECE is a problem-solving technique where you need to list all possible options before choosing a solution. It’s a three-step approach and the steps include:

1. Write down a clearly defined problem statement.
2. List solutions in a branching fashion, as Options 1, 2, and 3.
3. List the pros and cons of each option.

The image below illustrates the branching design of the MECE framework.

Click to enlarge

Next, you will choose a structured thinking framework to solve your data science business problem.