MEASURE FOR MEASURE
High Risk, No Reward
Reducing calibration costs could be an expensive mistake
by Christopher L. Grachanen
The troubled economic environment has caused many companies to reexamine their calibration policies and to question the need for calibration and frequency of calibration for much of their inspection, measurement and test equipment (IM&TE). This is often the direct result of mandates from senior management to reduce operating expenses.
Nondiscretionary expenses—in terms of required head count, occupancy and raw materials—are quickly reduced to levels of absolute necessity to achieve the biggest cost savings. These reductions are frequently laid out as marching orders delivered by senior management, which leaves little room for exceptions or alterations.
For frontline managers further pressured to reduce operating expenses, this generally leaves only a few possible areas available for additional cost reductions: employee incentives, capital acquisitions or upgrades, and services (internal and external).
The first two areas are typically frozen by a company’s finance group after essential, product-driven expenses have been optimized. This leaves services as the only area frontline managers have to make any real cost-reduction decisions.
Against all odds
For many technical groups, IM&TE calibration costs account for the lion’s share of service expenses. As a result, frontline managers often have no choice but to reduce their active IM&TE inventory to the bare essentials to trim calibration costs. This is to be expected and makes good business sense.
Unfortunately, the pressure to further reduce operating expenses (calibration costs) may put IM&TE users at odds with a company’s ISO 9001 requirements, as well as good laboratory shop practices based on fundamental metrological concepts.
Essentially, the ISO 9000 series of quality standards requires IM&TE to be calibrated if it is used to help decide a product’s acceptability (conformance to applicable specification criteria). IM&TE users may want to deactivate or change the active calibration status of an IM&TE piece to no calibration required (NCR) to avoid calibration costs if they still want to use it for what they consider to be nonessential, non-critical measurements.
Determining product-essential measurements is subject to wide interpretation for measurements that are not obviously used to evaluate a product’s functionality, quality and performance. To illustrate, we’ll look at a case involving a product prototype powered by a DC power supply whose output is monitored by a calibrated, handheld DC digital multimeter (DMM).
On the surface, you could argue the power supply is only a source and does not need to be calibrated, because its output is set up and monitored by the DMM.
Upon closer examination, however, power supply attributes (specifications)—such as periodic and random deviation (PARD) and transient response—can have an adverse effect on product performance if they are out of tolerance. The monitoring DMM cannot readily detect these DC power supply attributes and, if found to be out of tolerance, their adverse influences on product performance are unknown.
Another potential sticky situation involves a DC shunt whose initial resistive value is assumed by users not to change because it is constructed of solid metal and may not be subject to excessive wear or abuse.
This notion may be the case for users who employ shunts only to determine the presence of current (go/no go) and may not be applicable for other users trying to make accurate current measurements (all current shunt resistive values change to some degree over time). Normally, IM&TE is used by different users for different applications, and it is a safe bet that an application’s accuracy requirements are not the same as those of other applications.
Another calibration cost-cutting posture IM&TE users often propose is increasing calibration intervals.
At a recent metrology conference, I learned that many of my calibration colleagues have been requested by IM&TE users to arbitrarily increase calibration intervals to reduce calibration costs during these tough economic times.
I will not belabor the merits of statistical vigor used to predict the likelihood of acceptable IM&TE performance within a calibration interval period. I will note, however, that calibration interval changes that do not reflect analysis of past performance or other reliability-based information impose an unnecessary risk for consumers and producers.
Business decisions based on flawed measurements that stem from out-of-tolerance IM&TE can lead to increased field failures and longer product development cycles. The costs to the organization would be substantial, and in the worst-case scenario, customers would be put in life-threatening situations. A company’s calibration-related measurement risk exposure can be greatly reduced by adhering to a sound calibration program based on industry-accepted best practices.
An ailing economic environment often requires drastic cost-cutting actions if an organization is going to remain viable. Cost savings realized from circumventing industry-accepted calibration practices are rarely significant when compared to the costs of a product recall or inability to sell product because of the suspension of ISO accreditation.
Without rigorous root cause analysis, the contributions of calibration-related measurement risk to faulty business decisions often go unrecognized. Because of a lack of information and inadequate understanding or appreciation of the ramifications, companies seldom take into consideration calibration-related measurement risk exposure when making calibration cost-cutting decisions—a short-sighted approach that costs everyone in the end.
Christopher L. Grachanen is a master engineer and operations manager at Hewlett-Packard Co. in Houston. He earned an MBA from Regis University in Denver. Grachanen is a co-author of The Metrology Handbook (ASQ Quality Press), a senior member of ASQ, an ASQ-certified calibration technician and the Measurement Quality Division Certified Calibration Technician chairman.