A Case Study in the Use of Statistical Experimental Design and Data Analysis in Lubricant Formulation 2000-01-1963
This case study illustrates the value of employing statistical experimental design and data analysis when formulating lubricants.
A fractional factorial experimental design enabled all main effects of 6 formulation variables, plus 6 important 2-factor interactions, to be estimated from a total of only 16 runs. The design also allowed the engine to be serviced during the test programme without biasing the conclusions.
The statistical analysis of the resulting data is also described. Data from a reference oil were used to augment the test data and clarify the conclusions.