Quality management tools and techniques

Introduction to SPC

ISO Navigator Pro™ is a free tool that provides practical, expert guidance for businesses wishing to interpret and better implement the requirements of ISO 9001:2015, ISO 14001:2015 and ISO 45001:2018.

Our range of templates cover the requirements of ISO 9001:2015, ISO 14001:2015 and ISO 45001:2018, and offer an easy way to implement your next management system.

Statistical process control

Tools and techniques

Quality management tools and techniques are the means to implement and analyse your quality management system. Statistical techniques assist in identifying, measuring, analyzing, interpreting and modelling variability to maintain or improve effectiveness and efficiency.

Techniques use mathematical concepts and statistical methods to interrogate and make sense of the data collected from quality management system outputs to facilitate better use of available data to assist in decision making.

Statistical techniques

The analysis of the data can be undertaken using statistical techniques to help break down the information and present it in a format that allows a determination to be made of whether actions are needed to improve performance. Key quality management techniques include:

  • Gage Repeatability and Reproducibility (R&R) is a statistical tool that measures the amount of variation in the measurement system arising from the measurement device and the people taking the measurement.
  • Measurement uncertainty is the concept of evaluating uncertainty is fairly straightforward. It requires a detailed knowledge of the nature of the items being measured and of the measurement method, rather than an in-depth understanding of statistics.
  • Control charts are a method of measuring and controlling quality by monitoring the production process. Quality data is collected in the form of product or process measurements or readings from various machines or instrumentation. The data is collected and used to evaluate, monitor and control a process.
  • Histogram chart how often something happens and therefore highlights its likelihood. Histograms are used to show distributions of variables, while bar charts are used to compare variables.
  • Scatter diagrams can be used to verify a cause-and-effect relationship and find the correlation between two variables – dependent and independent.
  • Pareto analysis helps to evaluate the importance of items where resources may be best focused.
  • Product characteristics include drawings or specifications that define dimensions, tolerances, assembly interfaces, surface contours or shapes, materials, requirements for materials that would be an exception to an identified material specification, processes, surface preparations, etc.
  • Significant Product Characteristic (SPC) are those product characteristics that if not controlled within specified limits will negatively affect form, fit, function, safety, performance, agency approvals, or any governmental regulations.
  • Critical Product Characteristic (CPC) are those product characteristics that if not controlled within the specified limits, may have an unacceptable affect to form, fit, function, safety, performance, agency approvals, or any governmental regulations.
  • Tree diagrams are also called a systematic diagram, tree analysis, analytical tree, or hierarchy diagram, shows all the possible outcomes of an event. It can be used to calculate the probability of possible outcomes.
  • Data collection sheets/check sheets are a valuable way of collecting data, which then transforms into information to inform the quality process. Data in collected in real-time and at the location where the data is generated. Data capture instructions might include; procedures, quality plans, data collection sheets, data acquisition software, and sampling techniques.
  • Sampling sizes may be informed by process risk assessments or a statistical rationale based upon defined Acceptance Quality Limits (AQL), Unacceptable Quality Level (UQL). A sampling plan defines the product type, features, characteristics, batch/lot size, sample size, inspection levels, type of inspection, and the need for qualifying measurement systems analysis (Gage R&R),

The inspection data recorded should identify each characteristic and feature required by the design data or customer requirements, including the allowable tolerance limits and the actual dimension measured as objective evidence that each characteristic and feature has been inspected and is acceptable.

Ensure that statistically relevant representative samples are taken and that all information on the sample and the sampling procedure is recorded and documented. Key points to remember for sampling:

  1. Sampling should follow documented, quantitative acceptance criteria and/or sampling procedure;
  2. Sampling plans should be based on statistical theory or recognized standards;
  3. The sampling plan should be based on statistical methods;
  4. The sampling procedure should describe the selection and withdrawal of samples;
  5. The sampling location and procedure, the person who took the sample, and any other relevant information about the sampling process should be recorded.

Whatever statistical methodology is used, ensure your organization has determined the use of appropriate statistical tools and techniques and that statistical concepts are understood and used by the employees involved in the collection, analysis and management of statistical data.

Only competent personnel should use statistical tools and techniques. This includes part-time as well as full-time employees, as well as all management levels. Competence and skills may come from education, experience, or training; as evidenced by training records, competence matrix or job descriptions.

Tools and techniques to download for free...

Free SPC templates and guidance
Determination of Calibration Intervals.pdf Guidelines for the determination of calibration intervals of measuring instruments
Use of Data From Control Charts.pdf Use of Data from Control Charts for the Estimation of Measurement Uncertainty
Validation of Excel Spreadsheets.pdf Validation of Computerised Systems - Annex 1 – Validation of Excel Spreadsheets
Measurement Systems Analysis Tools.xlsx Including Repeatability, Reproducibility and Youden Plots, Method Comparison, Kappa, McNemar's Test, and Sensitivity and Specificity
Capability Study Template.xlsx Conduct a process capability study to determine the extent to which a process can meet expectations
Statistical Process Control Chart.xlsx Includes X-bar chart, range chart, histogram and probability plot
 

SPC control chart introduction for non-mathematicians

Standard deviation

All processes vary and the the aim of SPC is to minimize this variation. Variation is expressed as 'sample standard deviations' (SD) and SD is a mathematical term, based on probability theory. Click the links to download copies of a Capability Study Template and Statistical Process Control Chart.

Consider the process of driving a car along a dry, straight, level road; look at your hands on the steering wheel - they move as you make slight adjustments. Why are the adjustments necessary? Because the road is not totally flat, there's some play in car's suspension, etc.

If you draw a graph of the movements you will have a 'normal distribution curve'. In a normal distribution, most of the time your hands are in the middle. (In the secret language of SPC the 'middle' is called the 'average' or 'mean' or 'x-bar')

But sometimes your hands are a little to the left, at other times they're a little to the right, and on rare occasions they're on the extreme left or extreme right.

Normal Distribution Curve

Statistical process control application

Using the sample standard deviation (SD), you can then calculate and chart the usual extremes of a process under a given set of conditions. This variation is 'normal' and we cannot reduce the variation without making fundamental changes to the process (e.g. going on an advanced drivers course, buying sports suspension, etc.)

Because of the properties of the normal distribution it is usual in SPC or quality control applications to multiply the standard deviation by three and then add that to and take it away from the average which gives a confidence level which encompasses 99.73% of the observations or 'population'.

For example: 

average = 20

1 SD = 1.3

3 SD = 3.9

average + 3 SD = 23.9

average - 3 SD = 16.1

therefore 99.73% of all that we see will be between 16.1 and 23.9

Back in the car... 

A child runs out in front of you and you swerve to avoid him. The chart now shows a blip which is outside the normal curve. This indicates that something unusual has happened.

Non-normal Distribution Curve

In SPC applications it is often useful to recognize what is normal and what is non-normal. This case is clearly non-normal. In some situations also helps to know when things happen. Therefore, an SPC control charts also has a time axis.

Statistical process control chart

We can then apply these concepts in an SPC chart for monitoring customer complaints.

Control Chart

Having collected the basic information we can then add the average and +/- 3 SD markers.

Control Chart

What we are saying is:

  1. Given the normal variation of the process we will always have a level of complaints. There is nothing we can do without making fundamental changes to the process.
  2. As long as the individual plots are between the red lines then the process is exhibiting normal variation.
  3. BUT clearly something unusual (non-normal) happened in March - we had far more complaints than usual. We need to investigate and find out why.
  4. THIS IS THE CLEVER BIT - February's low result is equally unusual. We need to find why it happened and build that in to the process. Do it often enough and the average line will go down, and eventually become zero.

To help the investigation, use appropriate problem solving tools.

The problem with traditional SPC

Most SPC control chart techniques are based around the +/- 3 SD method, which covers 99.73% of a population. 99.73% sounds good until you realize that this means that 0.27% are not accounted for, there is 1/4 percent chance of something slipping through the net. 

If you are trying to control critical processes or very high volume (e.g. aerospace, building nuclear reactors, making silicon chips, etc.) then you will get it wrong once in every 370 attempts. 

Such critical industries tend to use Six Sigma (+/- 6 SD) methods as it gives a 99.9997% (3 in a million failures) confidence level. Three in a million sounds much better than one in three hundred and seventy.

To learn more, we recommend reading the articles from SPC Press & Statistical Process Controls, INC

Want to know more?

SSL certification

A certificate guarantees the information your internet browser is receiving now originates from the expected domain - https://www.iso9001help.co.uk. It guarantees that when you make a purchase, sensitive data is encrypted and sent to the right place, and not to a malicious third-party.

Free PDCA guidance

ISO Navigator™ is our FREE online training tool that shows you how to apply the principles of PDCA to your operations. We also offer many helpful templates that get you on the road to documenting your management system, please visit the download page.