8 Tools For Medical Device Quality Control

The quality of medical devices can be assured by using one of the many basic tools developed over the years for use in quality work. As examples of such methods can be mentioned Pareto diagram, check sheets, control charts and histograms. According to Ishikawa, 95% of all quality-related problems can be resolved by using any of these methods.

1. Pareto diagram

This tool is named after the Italian economist Vilfredo Pareto, and it represents a type of frequency chart where the data is arranged in a hierarchical order. It can help you identify the most significant problems that need to be corrected first. Typical for this tool is that it can summarise all types of data, however, it is mostly used to identify and determine nonconformities. To construct the Pareto diagram, you need to go through the following main steps:

• Determining the main approach to classify data (e.g. it is by cause, problem or nonconformity?)
• Deciding what to use to rank characteristics
• Collecting data for the appropriate time intervals
• Summarising the data and rank classifications from biggest to smallest
• Building the diagram and determining the significant few.

The Pareto diagram is an extremely useful tool for improving the quality of your medical device design.

2. Cause-and-Effect diagram (Fishbone diagram)

In 1943, Kaoru Ishikawa developed the Cause-and-Effect diagram, and because of this, it is also known as the Ishikawa diagram. This diagram resemblance the bone structure of a fish, showing the desirable or undesirable outcomes as effects and related causes as leading to, or potentially leading to, that effects. The right side of the Cause-and-Effect diagram (i.e. the “fish head”) denotes the effect, while the left side presents all possible causes linked to the central “fish” spine.

You can use this tool to investigate:

• A bad effect and consequently take appropriate measures to rectify the causes
• A good effect and learn about the causes that lead to it.

This tool can help you identify possible quality-related issues and inspection points.

The main steps you need to take to develop the Cause-and-Effect diagram are, as follows:

• Develop problem statement
• Brainstorm to identify possible causes
• Classify all identified causes into major categories
• Construct the diagram and refine the cause categories.

The use of this tool can bring many benefits, such as the fact that it’s useful to generate ideas and highlight the root cause or that it’s an effective tool to guide further inquiry. However, like any other tool, this tool can also be associated with some disadvantages, such as critical overlooking and complex interactions between causes.

3. Scatter diagram

This diagram can provide you with the simplest way possible to determine how two variables are connected and if a cause-and-effect relationship exists between them. However, please note that this tool cannot prove that the change in one of the variables is caused by the other, but only that there is a relationship existing between both of them and the strength of it.

Visually, the scatter diagram consists of two axes (horizontal and vertical). The horizontal axis presents the measurement values of one variable, while the vertical axis denotes the measurements of the other variable. To obtain a prediction equation, you can draw a line on the diagram using the least squares approach. If the line is extended beyond the plotted data points, it must be dashed as it indicates the lack of data for the concerned area.

4. Check sheets

The check sheets can help you collect data effectively and, afterwards, use and analyse it easily and efficiently. This tool can be very handy if you want to organize the data collected by classification and indicate the number of times that each specific value occurs. In order for this approach to be effective, you need to have at least 50 observations.

Some important points you need to keep in mind when using the check sheets are:

• They are useful for spotting problems when the frequency of a specific defect and how frequently it occurs in a particular location are shown.
• The check sheet form is individualized for each situation.
• It is usually designed by the project team.
• The check sheet should be user-friendly and includes data on time and location.
• Creativity is an important aspect of the check sheet design.

5. Histogram

The histogram plots data in a frequency distribution table. The main difference between the check sheets and the histogram is that here the data is classified into rows for the purpose of losing the identity of individual values. The histogram can be considered as the first “statistical” process control tool because it can describe the variation in the process. It can provide you with a satisfactory amount of information regarding a quality problem and consequently give you a basis for making decisions without additional analysis.

6. Flowcharts

This tool is great if you want to describe some processes in as much detail as possible by graphically showing the steps in proper order. Visually, the flowchart can be simple or complex and constructed of many boxes, symbols, etc. The complex flowchart shows the process steps in the proper sequence and related step conditions and the associated constraints by using various elements, such as arrows, if/then statements, and yes/no choices.

If built correctly, the flowchart usually:

• displays all process steps under consideration or analysis by the quality improvement team,
• highlights any crucial process points for control,
• suggests areas for improvement
• serves as a useful tool to explain and solve a problem.

7. Quality function deployment (QFD)

This technique was developed by professor Mizuno and first applied at Mitsubishi Heavy Industries Ltd. in 1972. The QMD a systematic mechanism to item design and manufacture, and it can provide you with an in-depth evaluation of a product or item. It is a planning tool that can help you satisfy your customers’ expectations. Its emphasis is on the customer requirements or expectations and because of this, it’s often called the voice of the customer. A set of matrices is used to relate customer expectations to counterpart characteristics expressed as technical specifications and process control requirements. A crucial element of the quality function deployment tool is the customer-need planning matrix.

The QFT approach consists of the following main steps:

• Identification of consumer expectations.
• Highlighting the product characteristics that will meet the consumer’s requirements.
• Association of consumer requirements and counterpart characteristics.
• Evaluation of competing products.
• Evaluation of the counterpart characteristics of competing products and developing objectives.
• Selection of counterpart characteristics that will be used in the remaining process.

Some of the main benefits when following this approach relate to:

• Improvement in engineering knowledge
• Engineering changes
• Quality and reduction in cost
• Product development time.

A major drawback is that the exact needs must be identified in complete detail.

8. Control charts

The control charts can help you monitor the state of the processes. They show statistically calculated upper and lower limits on either side of a process mean value. Their construction is based on statistical principles and distributions, and in particular, the normal distribution. The control charts are extremely useful for indicating trends and signals when the manufacturing process goes out of control. As overall, this is a great tool for determining the variability and its reduction as much as economically possible.

Source:

1. Dhillon, B.S. (2000). “Medical device reliability and associated areas”, CRC Press LLC