Abstract
We investigate the effects of alternate process models on the economic design of X andi? charts employed simultaneously. These process models comprise a memoryless or Markov model, the geometric model, and two models with memory, the Poisson and logarithmic series. Numerical results for a large experimental design are presented. These results indicate that both the Markov assumption and the shape of the distribution of the process model are important determinants of the quality of X and R control chart designs.