There is a requirement in both the 21 CFR 820 and ISO 9001 to utilize statistical techniques. According to the FDA, “Where appropriate, each manufacturer shall establish and maintain procedures for identifying valid statistical techniques required for establishing, controlling, and verifying the acceptability of process capability and product characteristics. Sampling plans, when used, shall be written and based on a valid statistical rationale. Each manufacturer shall establish and maintain procedures to ensure that sampling methods are adequate for their intended use and to ensure that when changes occur the sampling plans are reviewed.”
The FDA is looking for manufacturers to create sampling plans which are linked to the risk management program. When trying to establish the appropriate level of sampling that is to be used for a component or device, perform an evaluation to base that number on the actual risk of failure involved with that item.
Statistical techniques will typically fall into one of three aspects of the quality system:
- Incoming Inspection – This is an evaluation performed upon receipt of material or component from a supplier. The purpose of this step is to confirm the product meets the established specifications. Applying a sampling plan here would allow you to randomly select parts for inspection to represent the whole lot. The sampling plan is based on your desired confidence level, along with the risk involved with the supplier and/or component.
- Process Validation – A sampling plan used here would be helpful in proving the consistency of your process, along with allowing you to establish your process capability along with your ability to repeatedly take reliable measurements.
- Sample Size – When establishing the quantity that will be required to perform your tests, consider the indirect cost of completing the tests. A destructive test should take into consideration the price of the unit. For example, if a part costs millions of dollars, you are not going to want to destroy multiple units to prove conformance.
When trying to identify the appropriate and relevant areas to evaluate, consider the following criteria: selection of important characteristics, selection of measurement methods, selection of statistical techniques, record and report the results, establish control methods, perform ongoing verification, and build specific work instructions.
Let’s talk data loggers!
Talk with one of our experts to get your questions answered and see how we can help you solve your data logger pain points.