top of page

Global Manufacturer of Pressure Safety Products - Quality Testing and Forecasting

Analysis was done on product quality to ensure it met industry standards. Recommendations we're made based on the analysis of how to improve product quality.

Background:

A global manufacturer of pressure control devices was looking for assistance in quality assurance for a specific product line that narrow tolerance on the product performance.


It is critical that the device reliably relieve pressure at predetermined pressure levels for safety purposes. But releasing at pressures too low would unnecessarily disrupt the industrial process and result in both product waste and process downtime. AAARL team worked directly with the Engineering Manager and Product team for this project.


Challenge:

The expectation of the exercise was twofold:


1) To have an independent agent witness testing and verify the results of a statistically significant sample of devices and confirm 95% compliance of pressure relief at +/- 5% of the specified pressure measured in Pounds per Square Inch (PSI) when operating at the expected temperature. This was required to meet the applicable industry certification.


2) To obtain other insights to assist with manufacturing and quality assurance beyond the immediate requirement of witness testing.


The Approach:

The approach was to have AAARL witness test for multiple days and draft a compliance report for internal and marketing use. This was done in the presence of another certifying authority to confirm the calibration of the testing equipment (pressure and temperature gauges), the methodology and reliability of output. AAARL would provide independent confirmation of compliance (or failure) with the 95% confidence of relief at specific pressure at constant temperatures.


The testing included multiple lots of production for various pressure levels, with measurement of the mean, minimum and maximum point of failure along with the standard deviation. This was analyzed to predict the impact of the product variance on the probability of compliance.


Based on the observed failures, Monte Carlo simulation of 100,000 scenarios was run to extrapolate the probability of failure.


Results:

The product batches successfully passed the witness testing and the products were granted certification for use in the prescribed industrial process. The resulting publication was the first of its kind in the industry. The products performed beyond expectation and the company was able to showcase this “third party verified” result in their marketing material, which is something not done by any other competitors.

AAARL was further able to provide new insight into the degree of precision required in the product standardization to exceed the 95% confidence. This would allow for easier review of product performance by measuring variance from design as part of standard quality control.

bottom of page