Confidence and Reliability based sampling plans in the Medical Device Industry

Patrick Giuliano, Abbott Structural Heart

02/13 2020 @ 6:15 p– 8:00 pm

Qualcomm Inc. Building-B Cafeteria

3165 KifeRoad

Santa Clara, CA 95051


Acceptance sampling in the medical device industry typically takes it’s direction from ANSI Z1.4 (AQL attribute) and sometimes even ANSI Z1.9 (variable) sampling plan tables, of which, ~ 99% of US medical device companies today who perform IQC inspection employ some version of ANSI Z1.4. ANSI Z1.4 and its associated zero acceptance number (c=0) sampling plan derivative (Squeglia) are widely used because of their explicit endorsement by the FDA, in its Medical Device Quality Systems Manual: “[Sampling] Plans should be developed by qualified mathematicians or statisticians or be taken from established standards such as ANSI Z1.4.” A common misconception is held throughout the industry that this AQL inspection plan ensures that a certain level of lot product quality is being shipped to the customer, and this is simply not true!

In fact, these plan’s stated purpose “is not intended as a procedure for estimating lot quality or for segregating lots” … but rather to “induce a supplier [the producer] to maintain a process average… [and to] control consumers’ risk….” So what? Without this understanding, it can lead to significant risk mis-estimation, resulting in a much higher proportion defective being shipped to the customer than anticipated and otherwise mandated by risk-management procedures. Simply put, AQL plans don’t work “as-advertised” and they don’t focus on consumer protection!

A new kind of sampling plan is needed, one that specifies sampling levels based on consumer’s risk and not on producer’s risk. When we develop and establish an inspection plan that has an associated limiting quality level (LQL) that supports Risk-Management statements, we are focusing on the risk to the consumer: and this is crucial in high-risk patient-centered treatment environment. Several “Confidence and Reliability-based” sampling plans will be described and developed using the Operating Characteristic (OC) curve framework, and their practical utility will be presented.


Patrick Giuliano is an ASQ-certified manager of quality and organizational excellence (CMQ/OE) and hold a B.S. in Biomedical Engineering and an M.S. in Mechanical Engineering from the University of California, San Diego. He has over a decade of experience in medical device product development, manufacturing, quality and testing – in devices ranging from treatment of coronary/cerebral/peripheral vascular disease to heart valve dysfunction, and breast tissue reconstruction after mastectomy. Patrick is currently a Senior Quality Engineer at Abbott Structural Heart in Menlo Park Ca, where he acts as a site subject-matter-expert (SME) and trainer in the application of statistics and the use of JMP software for the effective management and interpretation of engineering data using quality process methods.

IEEE SCV REL sponsored by EAG

Attendance to this seminar will count towards professional development hours for IEEE, ASQ. Please feel free to forward this message to your friends and colleagues.

Check-in and pizza at 6:15PM – 6:45 PM.
6:45 to 8:00 – Conclusion with Q&A

Confidence and Reliability based sampling plans in the Medical Device Industry

Thursday, February 13, 2020

Sponsored by IEEE SCV Reliability Chapter

Time: 6:15 pm – 8:00 pm

Register here





Innovative Circuits Engineering, Inc.

EAG Laboratories | Eurofins Engineering Science