The 2017 revision of ISO/IEC 17025 brought decision rules to the forefront. When a customer requests a statement of conformity (e.g., pass, fail, in-tolerance), the laboratory must now apply a decision rule that accounts for the measurement uncertainty. This guide explains how to select and apply an appropriate decision rule.
What is a Decision Rule?
A decision rule is a documented rule that describes how measurement uncertainty will be accounted for when making a statement of conformity. In simpler terms, it's the rule you use to decide if a result is a pass or a fail, considering that every measurement has a degree of doubt.
Common Types of Decision Rules
- Simple Acceptance (Shared Risk): The most common rule. A result is considered a pass if it is within the specification limits, regardless of the measurement uncertainty. The risk of an incorrect decision is shared between the laboratory and the customer.
- Guard Banding (Stringent Acceptance): To reduce the risk of accepting a non-conforming item (false accept), the acceptance zone is made smaller than the specification zone. This is achieved by using a guard band.
Choosing a Decision Rule
The choice of decision rule should be agreed upon with the customer. If the customer does not specify a rule, the laboratory can choose a default rule, but this must be communicated to the customer. The decision should be based on the level of risk that is acceptable to the customer and the laboratory.
Documenting the Decision Rule
The selected decision rule must be documented. This includes the rule itself and the justification for choosing it. This information should be included in the test report when a statement of conformity is made.