Abstract The nitrogen content of meat, seafood and poultry products is frequently used in checking and enforcement of quantitative ingredient content declarations. Use of nitrogen content for checking ingredient declarations requires a good knowledge of the nitrogen content of unprocessed ingredients, typically expressed as a ‘nitrogen factor’. Nitrogen factors have generally been determined using extensive surveys of meat from particular species. Such surveys often need to assess a number of possible effects on nitrogen content, including, for example, breed, gender, age at slaughter, seasonal variation, carcase weight or size, and geographical origin. Full exploration of all pertinent factors can lead to large studies and high study costs. Here, we proposed the use of a number of alternative study designs, including fractional designs and algorithmic optimal designs, with a view to reducing survey costs with minimal impact on the information gained. The application of a reduced cost design is demonstrated using data from a previous study executed as a full factorial design. It is shown that broadly similar conclusions are reached with as few as one third of the samples originally taken.
[1]
D. Bates,et al.
Fitting Linear Mixed-Effects Models Using lme4
,
2014,
1406.5823.
[2]
Meat Products Sub-Committee..
Nitrogen factors for pork. Nitrogen content of rusk filler.
,
1961
.
[3]
J. S. Hunter,et al.
The 2 k—p Fractional Factorial Designs Part I
,
2000,
Technometrics.
[4]
J. Stuart Hunter,et al.
The 2 k—p Fractional Factorial Designs Part I
,
2000,
Technometrics.
[5]
Ulrike Grömping,et al.
R Package DoE.base for Factorial Experiments
,
2018
.
[6]
George E. P. Box,et al.
The 2 k — p Fractional Factorial Designs Part II.
,
1961
.
[7]
Per B. Brockhoff,et al.
lmerTest Package: Tests in Linear Mixed Effects Models
,
2017
.
[8]
Christopher J. Nachtsheim,et al.
A Class of Three-Level Designs for Definitive Screening in the Presence of Second-Order Effects
,
2011
.
[9]
Christopher J. Nachtsheim,et al.
Definitive Screening Designs with Added Two-Level Categorical Factors*
,
2013
.