Related Experiment Videos
A structured expert judgment study for a model of Campylobacter transmission during broiler-chicken processing
H J Van der Fels-Klerx1, Roger M Cooke, Maarten N Nauta
1National Institute for Public Health and the Environment (RIVM), Microbiological Laboratory for Health Protection (MGB), Bilthoven, The Netherlands. Ine.vanderFels@wur.ne
Summary
This study quantified uncertainty in campylobacter contamination during broiler processing using expert judgment. Optimized weighting schemes for expert data significantly improved risk assessment model accuracy.
Area of Science:
- Food safety
- Microbial risk assessment
- Veterinary public health
Background:
- Campylobacter contamination in broiler chickens is a significant food safety concern.
- Accurate microbial risk assessment models are crucial for effective control strategies.
- Quantifying uncertainty in expert-derived data is essential for robust risk modeling.
Purpose of the Study:
- To obtain input data for a microbial risk-assessment model of Campylobacter transmission during broiler chicken processing.
- To quantify uncertainty in model input parameters, specifically focusing on carcass contamination.
- To evaluate different weighting schemes for aggregating expert judgments.
Main Methods:
- A structured expert judgment study was conducted following established protocols.
- Expert assessments were elicited individually as subjective probability distributions.
- The classical model aggregated expert distributions, with three weighting schemes applied: equal, performance-based, and optimized performance-based.
Main Results:
- All three weighting schemes demonstrated adequate performance.
- Optimized combined distributions significantly outperformed equal weight and non-optimized combined distributions.
- The chosen optimized weighting scheme exhibited adequate robustness for further analysis.
Conclusions:
- Structured expert judgment is a viable method for quantifying uncertainty in microbial risk assessment.
- Optimized weighting schemes enhance the accuracy and robustness of expert-derived data for risk models.
- The findings support the use of optimized expert judgment aggregation in food safety risk assessments.