The use of experimental design methodology for investigating a lipid oxidation rate assay

The effect of five variables on the rate of lipid oxidation in fish oil was investigated. Two approaches for analysing the data were compared. The first approach was based on detailed chemical knowledge about the system, e.g. kinetics. The second approach was based on experimental design and empirical response surface modelling. The second approach gave slightly better predictions than the first. The measurement error in the experimental area of interest was estimated by repeated measurements, and based on this the signal to noise ratio was calculated. The SNR turned out to be the same in all of the measured levels, indicating that the measurement error of the system is proportional to the signal level.