5 Why Method - Metodologia 5 Porques

How to proceed “Why-Why Analysis” To identify the root cause for quick improvement - Recurrence prevention of root cause October 1st, 2003

Yazaki Corporation Toyota Business Unit 1

1. Introduction • Under the Toyota Production System, “Why-Why Analysis” has an ironclad principle to repeat “Why” 5 times at Toyota, in order to identify the root cause of a trouble. • This principle shows it is very important to identify the root cause and complete the recurrence prevention as quickly as possible when a certain trouble occurs. • These days, “Why-Why Analysis” brings about some successful results in identifying the root cause of a quality problem and implementing recurrence prevention. • I sometimes hear such complaints, “I don’t understand how to do Why-Why Analysis”, “I can’t repeat “Why” 5 times”, or “I can’t identify the root cause because I’ve found many possible causes”. 2

• For people in need, I give you tips to proceed “WhyWhy Analysis” in this document. I hope this document helps you understand and utilize “Why-Why Analysis”.

3

2. Before beginning Why-Why Analysis (1) Clarify the content of a problem, and understand the fact exactly. Clear up the problem with 5W1H. When

Date, Time

Where

Place, Process

Who

Person, Responsible person

What

Component, Part

Why

Reason, Cause

How Condition

Method, Means, Frequency, 4

(2) Understand the structure (mechanism) and function (role) of the problem. Check drawings, QC process chart, Inspection specification, Operation standard etc. to completely understand the parts, characteristics, functions and structures.

(3) Check the problem part of defective products and actual things of relevant parts. When checking the defective products, it is necessary to keep such pictures and data as dimensional accuracy and measured characteristics etc. 5

(4) Check the actual processing/assembly sites of defective products or relevant parts. On the day that the defective product is manufactured, ask the supervisors and operators if they’ve found some differences before/after the manufacturing, and record the findings. If changes in 4M (Machine, Material, Method, Man) are found, record the details.

6

3. 5-time Why If you ask “Why” only once, you cannot take a permanent measure for recurrence prevention because one-time “Why” provides superficial causes, ending up first aid. There are numbers of examples that repeating “Why” 5 times has reached to the root cause, based on the fact of actual product and production site in an attempt to identify the root cause. Followings are essential for the analysis.

7

Processing/Assembly conditions at the production site must be the same as those when the defect product was manufactured. It’s not too much to say that you can’t find out the root cause if the processing/assembly conditions are changed, because the existing situation at the production site is also changed. Processing and assembly must be checked right after processed or assembled. This is a very effective condition to identify the root cause based on the fact. 8

• Keep in mind that the more you are away from the time when a defective occurred, the more you have difficulty in identifying the root cause.

9

4. Case study from Toyota Production System Here is a model case of “Why-Why Analysis” for the phenomenon: “Machine does not work”

With 5-time “why”, Toyota identified the root cause.

10

Case study: “Machine does not work” (1) “Why the machine stopped?” Because the fuse blew out due to the overload. (2) “Why the overload was applied?” Because the lubricant for the bearing was not enough. (3) “Why the lubricant was not enough?” Because the lubricant pump did not pump up enough. (4) “Why the pump did not pump up the lubricant enough?” Because the axis of pump had looseness resulted from abrasion. (5) “Why the axis wore out?” Because some swarf came in resulted from the fact that there was no filter. 11

Repeating “why” 5 times, Toyota reached to the countermeasure: “Install a filter”

As this example shows, You must confirm based on the principles.

Phenomenon (result)

1st cause Machine stopped.

2nd cause Fuse blew out due to the overload.

3rd cause Lubricant was not enough.

4th cause Pump didn’t pump up enough.

5th cause Axis wore out.

12

Swarf came in.

• It is very important to stick to Genbutsu (actual product), Genba (actual site), and Genjitsu (fact) when you are trying to identify the root cause. • If you come to deadlock, it is important to - observe thoroughly. - check if there are some problems with each item and activity. - cling to questions you may have. - repeat “Why” until you are fully satisfied. 13

Let’s think about the following cases. 1) 1-time “why” • The supervisor replaced the fuse. This is a good measure as first aid. However, the same problem would reoccur in a couple of months because the measure is not taken for the root cause. 2) 3-time “why” • The lubricant pump could be replaced. However, the same problem would reoccur in a few months because this measure is not taken for the root cause as well. Also, the measure costs too much. 14

3) 4-time “why” • The axis of pump could be replaced. However, the same problem would reoccur in a few months because this measure is not taken for the root cause as well. The cost would be cheaper than that in the case of 2) above, but it takes long time to determine the frequency of replacement. 4) 5-time “why” • Permanent measure can be taken, and the machine will never stop due to the same reason. Production problems come to the end when recurrence prevention for the root cause is completed. 15

• In this way, if you pursue “why” 5 times, you can find out the root cause and can take permanent measures.

16

Now, if I repeat “why” further, followings come out. (6) Why the filter was not installed? Because the machine designer didn’t realize the necessity of the filter. (7) Why the designer didn’t realize? - Because he/she had no thoughts about swarf. - Because there was no system to feed problems occurred at a plant back to the Equipment Design Department. (8) Why the designer didn’t realize the treatment of swarf? Because he/she didn’t care about issues other than the conditions for product processing. 17

• In this case,

“Why-Why Analysis” is not focused on the relationship between “result and cause” with actual product and production site, but on the designer or system. In other words, this analysis is far away from the original purpose that is to take the recurrence prevention of root cause. If it is necessary to improve the system, an appropriate department at the head quarter must be in charge and enhance the system.

18

5. Notices on “Why-Why Analysis” When analyzing, identify the cause from both aspects of “made the problem occur” and “made the problem flow out”. To figure out if you reach the root cause or not, confirm that the problem will never occur due to the same cause. It is quite unusual that one phenomenon (Why) is caused by two causes at the same time. Production problem should be solved by identifying root cause in production process errors but not as design errors. Do not take up an assumed cause which is not found from the fact. 19

Sentences to specify “phenomenon” or “Why” must be simple and short. After finishing the analysis, read the report from the last “Why” to the first “phenomenon” with using the conjunction of “thus”, in order to confirm that the analysis makes sense, is reasonable, and does not have any leaps in logic.

Why Swarf came in.

Thus,

Why Axis wore out.

Thus,

Why

Thus, Phenomenon

Fuse blew out. 20

When continuing “Why”, stop analyzing a possible cause if you could determine, based on the investigation of actual products and production site, that a phenomenon is not caused by the possible cause. Do not find causes in human mentality. Ex: The operator was out of mind. The operator was tired. Do not use the word of “bad” and unclear words in the sentences.

End

21