Distributional assumptions in food and feed commodities- development of fit-for-purpose sampling protocols

Area of Science:

• Food Science
• Analytical Chemistry
• Statistical Quality Control

Background:

• Current food and feed sampling guidelines often rely on unrealistic assumptions of randomness.
• Material heterogeneity within commodity lots is a significant factor influencing sampling effectiveness.
• Practical constraints frequently overshadow the need for representative sampling protocols.

Purpose of the Study:

• To demonstrate the effectiveness of the Theory of Sampling (TOS) in addressing material heterogeneity in food and feed sampling.
• To highlight the limitations of traditional statistical distributions in sampling protocols for heterogeneous materials.
• To provide a scientifically sound basis for developing valid sampling protocols.

Main Methods:

• Application of the Theory of Sampling (TOS) principles to analyze sampling errors.
• Review and summarization of empirical data from a European Union study on genetically modified soybean heterogeneity (Kernel Lot Distribution Assessment).
• Evaluation of compositional and distributional heterogeneity in commodity lots.

Main Results:

• The Theory of Sampling (TOS) provides an effective framework for managing both compositional and distributional heterogeneity.
• Traditional sampling methods based on random distribution assumptions are inadequate for heterogeneous materials.
• Empirical data supports the universal applicability of TOS principles in the food and feed sectors.

Conclusions:

• The Theory of Sampling (TOS) is essential for developing accurate and reliable sampling protocols for food and feed commodities.
• Ignoring material heterogeneity leads to biased assessments and ineffective quality control.
• Valid sampling protocols must be based on TOS principles to overcome distributional constraints and ensure representativeness.