ABSTRACT
In Chapters 2 and 3, robust first-and second-order designs are examined assuming the response (y) follows Normal distribution. In the present chapter, robust first-and secondorder regression designs are considered for two non-normal distributions (i.e., log-normal and exponential). For example, in quality engineering, the most commonly used lifetime response distributions are log-normal and exponential. For lifetime improvement experiments, generally, design of experiments are mostly used in locating the optimal operating conditions of the process parameters. In the present chapter, D-optimal robust first-order and robust second-order rotatable designs are described for correlated lifetime responses having log-normal and exponential distributions. Robust designs are described for some well-known error variance covariance structures as mentioned in Chapter 2. It is shown that robust first-order D-optimal designs for these two lifetime distributions are always robust rotatable, but the converse is not true.
