Q: I know that calibration labs holding the ISO/IEC 17025:2005 standard are considered to be in compliance with ISO 9001:2000. But what about the most recent release, ISO 9001:2008? Do the labs need to make some adjustments to their quality systems?
Jeddah, Saudi Arabia
A: First, let’s clarify a few things. Calibration laboratories are accredited to the ISO/IEC 17025:2005 standard. Accredited laboratories are not compliant to ISO 9001, but they meet the intent of the ISO 9001 requirements.
ISO/IEC 17025:2005 was released to ensure its compatibility with the requirements of ISO 9001:2000—Quality management systems—Requirements. This became necessary because of the generalized adoption of quality management systems (QMS)conforming to ISO 9001:2000, including many of the organizations that testing and calibration laboratories serve.
The International Organization for Standardization also clarified that while compatible, ISO/IEC 17025 and ISO 9001 are not interchangeable. Although both standards can be used by laboratories as a framework for providing their customers with confidence that the laboratories are managing their activities, only ISO/IEC 17025 can be used to demonstrate the technical competence specific to laboratories.
Laboratories may choose to be accredited to ISO/IEC 17025, certified to ISO 9001 or both, but the processes of accreditation to ISO/IEC 17025 and certification to ISO 9001 would still be two separate actions—although highly facilitated for the laboratories and the assessors by the consistency between the two standards.
There are no essential changes to the technical requirements in ISO/IEC 17025:2005 from the 1999 version. The modifications relate mainly to the management requirements in the document to reflect the content of ISO 9001:2000, especially a greater emphasis on:
- The responsibilities of top management.
- The need to demonstrate a commitment to continually improve the effectiveness of the management system.
- Customer satisfaction.
- Internal and customer communication about the management system.
Laboratories that implemented their processes as required by the 1999 edition of ISO/IEC 17025 should have only needed to make minor adjustments to ensure their existing procedures comply with the new management requirements in ISO/IEC 17025:2005.
The changes to ISO 9001 from the 2000 version to the 2008 version clarified some requirements in the older edition and will not affect laboratories accredited to ISO/IEC 17025:2005. Also note that this year, ISO/IEC 17025:2005 was reaffirmed for another five years. Therefore, accredited laboratories will not need to modify their QMS.
If you still need some clarification on the differences between accreditation and certification, the International Laboratory Accreditation Cooperation has a publication, "Laboratory Accreditation or ISO 9001 Certification," that may be of help.1
- International Laboratory Accreditation Cooperation, "Laboratory Accreditation or ISO 9001 Certification," www.ilac.org/documents/Bro_english/Laboratory_accred_or_cert.pdf (case sensitive).
For More Information
- Shah, Dilip, "Calibration Evaluation," Quality Progress, September 2010, pp. 44–45.
- Shah, Dilip, "Supplier Demand,"
Progress, May 2010, pp. 48–50.
Q: One of the key requirements of any measurement system is that it is fit for purpose. This is left as vague as possible in standards from the International Organization for Standardization (ISO) and the Code of Federal Regulations from the U.S. Food and Drug Administration (FDA).
Over time, the approach has been to apply a test accuracy ratio (TAR) or test uncertainty ratio (TUR) of 4:1. This has become the accepted norm but hasn’t been specified anywhere.
I lecture on metrology to part-time students who primarily work at U.S. multinationals based in Ireland. Some of my students recently told me that corporate decisions have been made to remove any mention of specific TARs or TURs from their procedures.
When I asked what would be the criteria for fit-for-purpose measurement systems, they told me it was to be a vague as possible. I know fit for purpose is ultimately the vendor’s call, but surely we need some criteria.
A: Here in the United States, we discuss calibration, traceability and uncertainty as applicable to quality calibration programs. The 4:1 rule comes from ANSI/NCSL Z540.3-2006, released by the American National Standards Institute.
If an organization wants to eliminate any mention or use of a TUR equal to or greater than 4:1, it must use and show its uncertainty budget—usually the root of the sum of squares plus combined uncertainty with a k = 2 for a 95% confidence level—for all parameters of the calibration being performed.
If calibration is being performed inside a company—such as biotech, pharmaceutical, medical device and healthcare—and the standards being used are far more accurate than the unit under test, it is usually not economically viable to perform complex uncertainty budgets during all calibrations.
But if this is corporate policy, it would seem to me that somebody is getting the wrong information.
As far as fit for purpose, I have not heard that term applied to a quality calibration program. A quality calibration program should be designed to meet the ISO standards and FDA regulations within the parameters of the company’s goals for a traceable calibration program.
Bucherview Metrology Services
De Forest, WI
For more information
Bucher, Jay L., "Where Does It Say That?" Quality Progress, November 2010, pp. 50–51.