ECONOMIC DESIGN OF (X)over-bar CONTROL CHART UNDER MEASUREMENT ERROR

In most of the industrial situations, data follow normal distribution. We may be confronted with an industrial situation where the assumption of normality and measurement errors are achievable or desirable. Thus, there is a need for a procedure which enables us to deal with measurement errors of the data. The design of a control chart requires the engineer or analyst to decide a sample size, a sampling interval and the control limit, to design control charts accordingly and to continue our search for the assignable causes of variation. The objective of this paper is to determine the design parameters, namely, sample size (n) and sampling interval (h) between successive samples. A numerical illustration has been supported to investigate the effects of cost parameters on the solution of the design. It may be inferred that measurement errors affect considerably the optimum value of the sample size and optimum sampling interval. It is necessary to point out that the measurement errors of the population should be taken into account while designing a control chart as the optimum values of the control chart parameters are affected by the measurement errors of the population. Visual comparisons of OC and ARL curves have been drawn in support of the problem.