5.7 Basic Lean Tools: Cause and Effect Diagrams

Another critical tool in the improvement process is the cause and effect diagram (or the Ishikawa diagram, after its developer Kaoru Ishikawa). Due to its appearance, it is also often called the “fishbone” diagram.

In a previous section we identified issues and categorized them into groups within the affinity diagram. The cause and effect diagram helps us further examine the cause of problems and group them into categories to identify overlapping areas.

This diagram is a visual tool that presents information in an easily understandable format. Just as the Pareto chart helps us identify and prioritize issues, the cause and effect diagram allows us to visualize potential root causes or problems in our system. The only downside to this tool is that with highly complex and interrelated issues, it can become cluttered, which reduces its effectiveness as a “simplifying” visual tool. However, it is convenient and helpful for most problems, allowing users to see the various causes simultaneously.

Step 1 — Identify the Problem

To create and use a cause and effect diagram, you should follow a four-step process.

The first, and most simple, step is identifying the problem you are trying to solve. Continuing our example with the distribution center manager, we’ll take the largest error found in the Pareto chart example. We want to determine why so many pallets are shrink-wrapped incorrectly. Once you’ve determined the problem you wish to identify, you will draw the “spine” of the fishbone diagram with the problem at the end (note: there is not a “correct” left or right direction to draw the chart, but it is preferred to have the problem placed on the right side).

Step 2 — Add Main Causes

Next, we will add the “ribs” to the spine. Again, there is not a specific, correct way to do this, but there are some standard practices. The goal is to identify three to five leading causes of the problem. Each of these items will be a rib attached to the spine.

There are several “best” practices for naming the leading causes.

• Use your group’s titles from your affinity diagram. These will then be your leading causes in the diagram.

• Use the “3Ms and a P” approach (demonstrated in the figure below). This method uses four common causes within which most problem areas fall: Methods, Machinery, Materials, and People (3Ms and a P).

• Use the “4Ps” approach: Policies, Procedures, People, and Plant.

These are all reliable methods that you can use. Many organizations also include “Environment” as an additional cause to incorporate other issues such as sustainability, reverse logistics, etc.

All approaches work well and should provide you with a useful cause and effect diagram. The most significant difference between the naming of the leading causes of the spines is where your exact reasons will be added in the next step.

Step 3 — Add Detailed Causes

Next in creating the cause and effect diagram is to add more detailed causes to each of the leading causes.

The goal is NOT to assign blame but rather to identify problems and causes within the process that your team can correct.

Here are a few examples of more detailed causes you might add.

• A lack of new employee training may be part of the problem. Therefore, you would add this detailed cause (also referred to as a level one cause) under either method, people, or both.

• You fail to train new employees on the machine. This a method problem in your training process.

• The shrink wrap is too thin and keeps tearing. That could be added under materials.

You may be tempted to consider it a failure of the people operating the shrink wrap machine. Remember, this diagram is not meant to assign blame, but to identify problems that can be corrected. You may identify factual problems (such as Bob being late to work), but you do not want to treat Bob as the problem, rather the fact that he is arriving late. This is a thin line, but it is an important one to walk.

The figure above demonstrates adding the first few detailed causes. As you continue, you will fill out the rest of the diagram as needed to capture all the applicable causes. For example, the following is more filled out to demonstrate a halfway completed diagram.

Your final diagram would look similar to Figure 5.9, Cause and Effect Diagram Example, at the beginning of this section.

Step 4 — Analyze the Results

The final step in the cause and effect diagram process is to analyze the results. The previous figure would be much more complete and provide an overview of all the leading and exact causes. In this step, you want first to see if there are similar or related causes. This procedure helps to prioritize work efforts for improvement. For example, in our modified cause and effect diagram, we see two trends.

First, there is a people issue: the inadequate training upon hiring, which leads to errors in the process. Second, there is a machinery issue: the complexity of the machine setup and failure to ensure the shrink wrap device is within specifications before use.

Sometimes the connection between items is obvious (i.e., untrained workers). Other times, the impact of one error is a second-order effect in another (i.e., poor training impacts and is impacted by the machine’s complexity and setup requirements). The purpose of the analysis is to determine what corrective steps you should prioritize. For example, if poor worker training is a frequent cause, it is likely an excellent area to target initial improvement steps. This may also have unintended positive impacts in other areas.

For example, improved training should help to ensure the machine is correctly set up and maintained. This action may reduce the number of times the pallet cuts the shrink wrap because the shrink wrap no longer “spills over” onto the pallet and therefore doesn’t get torn by the edges.

In conclusion, the cause and effect diagram is a great tool to help identify causes that impact the effectiveness of a process. It is also helpful to prioritize corrective actions. Finally, it helps to visualize the relationships between various causes across the process.