Throughout the years of experiences in Quality Control Tools training and application, we have seen many industry practitioners are able to apply the quality control tools for product/process improvement. However, industry practitioners nowadays have concerns on effectiveness of quality control tools application. The tools can easily be applied but how useful the analysis results for making right decision. We have quoted some common mistakes typically made in quality control tools application.
1) Cause and Effect Diagram - Lack of verification on the root cause
The Cause and Effect diagram which is well known as Fishbone diagram is commonly used to identify the root cause of the problem. However, root cause identified without verification will lead for wrong conclusion. The root cause shall be verifiable and actionable to solve the problem. Example, root cause identified is “Operator Error” but after verification is found that the “Operator skipped a process step”. Then, subsequent solution or action to take is “how to detect or prevent the process step which was skipped by the operator”.
Each root cause identified must be verified to confirm that it is a true root cause. The verification shall be done by the team members who are the subject matter expert in technical knowledge and engineering experience to solve the interested problem. Usually the process owner or the competent personnel fulfill the requirement.
2) Pareto Chart – Misconception of using defect occurrence data only for analysis
As you may know, Pareto Chart uses 80/20 rules of thumb to prioritise the area of focus for improvement. Most of time, industry practitioners use defect occurrence data to create the Pareto Chart. However, it is not the only data can be used but these are the data usually collected by Quality Control and Production Process departments to create the Pareto Chart. As compared to defect occurrence costing data, creating the Pareto Chart using costing data may show you different priority of area for improvement. The management has more concerns on cost reduction over defect reduction improvement.
Create 2 Pareto Charts to compare before and after improvement action done to evaluate effectiveness of action taken. Use costing data to create Pareto Chart will make more logical sense if you want to see bottom line significant improvement as each amount of cost gained will be quantified.
3) Scatter Diagram – Use of causal relationship between 2 variables data
We use Scatter Diagram to show linear relationship between 2 variables. When expense is Y variable and income is variable X, the data points are in ascending trend from bottom left to top right of the Scatter Diagram. This indicates that increasing Y is observed when X is increasing or decreasing Y is observed when X is decreasing. It does not imply that increasing Y is caused by increasing X. Unlike Cause and Effect Diagram, it does imply causal relationship.
Use continuous data for X and Y variables to display their linear relationships. The X and Y data are in pairs. More pairs of data are desired to plot the coordinates in the Scatter Diagram to observe and identify the linear relationship for the same samples. This will help to reduce sample to sample variation. Often, we calculate correlation to justify the acceptability criteria of correlation results. Subsequently, we will relate a regression model when there is a strong linear relationship between variables for prediction use.
4) Check Sheet – Perception of Check Sheet is identical to Check List
What do you think? In fact, manufacturing industry use many types of check sheet to collect data for information and facts. Example, we use QC Inspection Check Sheet to record or count the occurrences of product defect data daily during inspection. However, we use Tooling Verification Check List to inspect the adequacy and completeness of tooling/die change-over verification step-by-step in sequence.
Use Check Sheet to record the right data for analysis use to reveal the facts for decision making. The important need is to design the check sheet which is user-friendly for the users to record the data. The Check Sheet shall be easy to understand and only meaningful data are recorded as data collection can be time consuming and requires a lot of efforts like resources and commitments. The accuracy and integrity of data have always been the concerns to trust the validity of true data.
5) Control Chart – Incorrect application of control chart
The control chart is usually applied to detect special cause of variation from process. However, many blame that control charts are not practical and effective to control my process. The people are saying this whether they really use the correct control chart to control process variation. Learning the knowledge can be simple but applying it can be a great challenge. How to know you correctly use the control chart? There are guidelines on the selection of control charts to use to control your process. The subgroup size (samples size per subgroup) and subgroup frequency (number of subgroup) are vital to determine how effective your control chart to detect the process variation. The technical knowledge and process owner experiences are important factors to justify the out of control situation. The real time feedback to rectify the out of control situation has been a catalyst to effective implementation of control chart as a prevention control approach.
Use control chart to identify and detect out of control situation during the production process. Control chart tells us what has gone wrong with the process and when did it happen but it can’t tell why it happened. The need to use root cause analysis methodology is to verify the root cause and take necessary actions to rectify the special cause of variation.
Interested in discussing your requirements? Let’s talk.
Training / Online training / Capacity building