An alternative procedure for the estimation problem in 2n factorial experimental models

Factorial experimental designs are widely used as ‘off-line’ statistical procedures for product and process quality improvement. Analysis of variance is an appropriate method for analyzing the resulting data from such experiments, and enables engineers to estimate both magnitude and direction of the effects of factors. But magnitudes estimated by using some conventional side conditions do not reflect the realistic numerical contributions of levels of factors and interactions to the factorial model. In this article, some alternative side conditions are proposed to tackle this problem, along with the results of a case study. In this case study, a full factorial experiment demonstrating the superiority of the proposed side conditions over other alternatives has been carried out.