ISO 18315:2018

Overview - ISO 18315:2018 (measurement uncertainty for impurities in uranium solution)

ISO 18315:2018 provides guidance for evaluating measurement uncertainty when impurity concentrations in uranium solutions are determined using a linear calibration fitted by the method of least squares. The standard introduces and applies a reversed inverse regression approach (also called pseudo-basic regression) in which the reference solution values (y) are treated as the response and instrument measurements (x) as inputs. It is intended for chemical analyzers and laboratories performing quantitative analysis in the nuclear fuel cycle.

Key topics and technical requirements

• Regression line fitting: Fit a linear calibration y = a + b x using ordinary least squares (OLS) with n calibration points.
• Reversed inverse regression: Distinguishes this approach from classical/inverse regression; assumes xi vary (approximately normally) while yi are fixed.
• Calibration uncertainty: Estimate the variance of the predicted y value from the fitted line and treat its square root as the calibration uncertainty (degrees of freedom = n − 2).
• Random measurement uncertainty: Account for repeatability and sample measurement variability separately from calibration uncertainty.
• Combined uncertainty and degrees of freedom: Combine calibration and random uncertainties by propagation rules and compute effective degrees of freedom using the Welch–Satterthwaite approximation.
• Expanded uncertainty and coverage: Apply a coverage factor k (based on effective degrees of freedom) to obtain expanded uncertainty (typically ~95 % confidence).
• Reference solution uncertainties & bias correction: Reflect uncertainties of calibration standards in the final result and apply bias correction where needed.
• Adequacy checks and weighting: Check calibration quality; Annex C explains weighting factors for handling non‑uniform variances in reversed inverse regression.
• Supporting material: Annex A offers a practical worked example; Annex B provides a flowchart of the uncertainty-evaluation process.

Practical applications and users

• Target users: analytical chemists, calibration laboratories, QA/QC teams and instrumentation engineers working in nuclear energy, uranium processing, safeguards and radiochemistry.
• Use cases: Determining impurity concentrations in uranium solutions (e.g., process control, compliance testing, material accountancy) where linear calibration and reliable uncertainty budgets are required.
• Practical benefits: Provides a reproducible statistical method for expressing measurement confidence, integrating calibration and measurement variability, and ensuring traceable reporting of impurity results.

Related standards and guidance

• ISO 18315 complements general measurement uncertainty frameworks and calibration best practices (for example, the GUM-type approaches).
• It is specifically tailored to linear-regression calibration contexts in nuclear chemistry and should be used alongside laboratory quality-management and accreditation requirements.

Keywords: ISO 18315, measurement uncertainty, uranium solution, impurity determination, linear regression, reversed inverse regression, calibration uncertainty, Welch–Satterthwaite, weighting factors.