Abstract: A
multivariate exponentially weighted moving average (MEWMA) control chart is used
for fast detection of small shifts in multivariate statistical quality control.
Recently, it was found, the use of variable parameter scheme could further
enhance the efficiency of the traditional MEWMA control charts. The control
charts using variable parameter policy have been shown to give substantially
faster detection of most process shifts than the traditional control charts.
Previously, economic design of variable sampling intervals control
charts using genetic algorithms, was used in the consideration (Chou et al.
[11], Chen [41]). The objectives of this work are to extend to minimizing the
hourly cost of MEWMA control charts with the variable parameter using genetic
algorithm involves determining the optimum values of the control chart
parameter. We develop the economic design of variable parameters MEWMA control
chart to determine the value of test parameters of the chart (i.e., the sample
size, the sampling interval, exponential weights for past observation, the
warning limit and the control limit) such that the expected hourly loss is
minimized. Also the expected cost per hour in constructed and regarded as an
objective function for optimally determining through the genetic algorithm (GAs)
is given. Sensitivity analysisof the effects of model parameters on the
optimal design of MEWMA control chart with the variable parameter has also been
performed.
Keywords and phrases: a multivariate exponentially weighted moving average (MEWMA) control chart, variable parameter, genetic algorithm, sensitivity analysis.
