It is a technique used when a product is not dismantled and therefore it is not possible to use the research component technique.
It is noted that if the defect is ” good/Not good “, this technique can be used to identify a measurable characteristic and thus reformulate the problem.
1 – Select Samples
The principle of the tool is based on the comparison by pair of good and bad elements. We randomly choose as good as bad, then we randomly Shape pairs of good and bad. For the selection:
- If the defect is measurable: samples with the most extreme values of the defect will be taken.
- If the defect is attribute: We will take samples with the most pronounced defects.
2-Observing the differences
On a first pair, one identifies the difference (or differences) that distinguish them.
3-Identify the characteristic
The previous step is reiterated to identify a trend in a characteristic. The challenge is to show that a characteristic is clearly different between the good group and the bad parts group. Shainin considers that in excess of 5 to 6 pairs with a different characteristic, the cause of the variation is clear.
We are trying to understand why some disc brakes are more braking than others. We decide to take 50 good and 50 bad. It is believed that the thickness of the disc is at issue. We decide to measure the thickness in eight different places. The graph below shows the results of the measurements on point 4. Right brakes, left brakes less efficient. It is noticed that the defective brakes have a greater thickness, which is confirmed on the points.
It is concluded that this characteristic is at the origin of the defect.
R. D. Shainin (1993) – Strategies for technical problem solving
K. R. Bhote, A. K. Bhote (2000) – World class quality
K. S. Vinay, P. Gowda, H. Ramakrishna (2004) – Industrial scrap reduction using Shainin technique