Optimal and Practicable Designs for Measuring Plaque pH-Profiles

In this paper we will show that the theory of optimal experimental design leads to quite different patterns of measurement. These patterns take into account (a) the structure of the ongoing process and (b) what parameters or characteristics are object of inference. We will demonstrate that the use of optimal designs or of designs that are close to optimal ones allows to reduce the sample size without decreasing the accuracy of the results. In Section 2 we present a model that is suitable to describe the pH-profile and we give some justification on an empirical basis. Section 3 discusses optimal experimental designs for estimating the model parameters or other characteristics of the pH-profile. Section 4 gives some numerical illustrations of designs that are optimal or that are derived from the optimal design by taking practicability into account. In the final section, recommendations are given for the practical use of the suggested designs. (author's abstract)