The Cost of Quality: Where Calibration Pays for Itself

When budgets are tight, calibration is often one of the first costs questioned. It appears as a regular invoice, and its benefits are invisible when everything is working. The cost of quality model gives a better way to look at it. It asks not what calibration costs, but what it costs not to measure correctly.

What “cost of quality” means

The concept was developed by Joseph Juran and Armand Feigenbaum in the 1950s and later popularized by Philip Crosby as the “price of conformance” versus the “price of nonconformance”. It splits all quality-related costs into two groups:

• Cost of good quality (COGQ): what an organization spends to get things right
• Cost of poor quality (COPQ): what it loses when things go wrong

The prevention appraisal failure (PAF) model

Category Group What it includes Calibration examples

Prevention | Good quality | Activities that stop problems | Instrument selection, planned calibration, | occurring | procedures, training

Appraisal | Good quality | Activities that detect | Inspection, testing, audits, intermediate checks | problems

Internal failure | Poor quality | Problems found before | Scrap, rework, re-testing, product held for | delivery | investigation

External failure | Poor quality | Problems found after delivery | Complaints, returns, recalls, penalties, lost | customers

The model’s central principle is that money invested in prevention and appraisal reduces failure costs. External failures are almost always the most expensive, because they add damage to reputation and customer relationships on top of the direct cost.

A practical way to picture it: the “1-10-100” principle

Quality professionals often explain the cost of quality with a simple illustration known as the 1-10-100 principle. It is commonly attributed to George Labovitz and Yu Sang Chang in the early 1990s. It is not a formal standard or

a scientific law, and the numbers are not measured ratios. It is a practical illustration of one idea: the later a problem is caught, the more it costs.

A widely used informal rule of thumb puts this in simple terms:

• 1 to prevent a problem
• 10 to correct it once it is found inside the organization
• 100 to deal with it once it reaches the customer

The ratios are illustrative, not exact, but the direction holds in almost every industry. A calibration costs the “1”. The out of tolerance instrument it would have caught costs the “10” or the “100”.

Worked example 1: product giveaway when the instrument reads low

Illustrative figures for a food plant filling 100 g packs:

Step Figure

Checkweigher drift Reads 0.5 g low

Effect The line overfills every pack by 0.5 g to stay above the declared weight

Daily production 20,000 packs

Product given away per day 10 kg

Per year (300 working days) 3,000 kg (3 tons)

At an illustrative product value of EGP 100 per kg EGP 300,000 a year from one instrument

The calibration of that checkweigher costs a small fraction of this. The drift would have shown up in the as found data long before anyone noticed it in the yield figures.

Worked example 2: underfilling when the instrument reads high

Now reverse the drift. The checkweigher reads 0.5 g high:

• Packs appear correct on the line but are underweight.
• Consumers receive less than the declared quantity.
• Pre-packaged goods are subject to legal metrology controls, so inspections by market surveillance authorities can lead to penalties, product withdrawal and public complaints.
• Major retailers and export customers may reject consignments or remove the supplier from approved lists.

In this case there is no visible loss on the production floor at all, until the failure reaches the customer or the regulator. That is the “100” in the 1-10-100 rule.

Where hidden costs appear

The cost of poor quality is like an iceberg: only a small part is visible. The visible costs, such as scrap and rework, customer returns and calibration or repair invoices, appear in reports and budgets, so they get attention. The larger share sits below the surface.

When an instrument is found out of tolerance, every batch measured since its last good calibration has to be reviewed, and product is often placed on hold while the investigation runs. Management time is spent closing audit nonconformities instead of improving the business. Disputes over quantities and invoices strain relationships with customers and suppliers. An unreliable instrument can trip a process and cause unplanned downtime. Most costly of all, and hardest to measure, is the loss of customer trust and the future orders that go with it.

None of these hidden costs appears on a calibration invoice, yet together they usually outweigh the visible ones many times over.

How to measure your own cost of quality

• List the four categories and assign each relevant cost to one of them.
• Collect the data from finance, production and quality records over a defined period, for example 12 months.
• Include hidden costs. Estimate staff time on investigations, holds, re-testing and complaint handling.
• Express COPQ as a percentage of sales or production cost, so it can be tracked over time.
• Target the biggest failure costs with prevention, then measure again.

ISO 10014, Quality management — Guidance for realizing financial and economic benefits, gives organizations a framework for linking quality activities like these to financial results.

Spending calibration budget where it matters

Reducing cost of quality does not mean calibrating everything more often. It means matching effort to risk:

• Critical instruments that affect product safety, legal compliance or customer acceptance: accredited calibration at evidence-based intervals.
• Working instruments: calibrated in-house against accredited references, where competence allows.
• Indication only instruments: clearly labelled and excluded from calibration.

Calibration history then shows where intervals can safely be extended and where they need to be shortened. See How to Set Calibration Intervals.