Statistical And Economic Study Of Dynamic T~2 Control Charts With Switching Rules

Quality has always been a core element of an enterprise's competitiveness.In today's production,it is the primary task to vigorously develop quality and efficiency.In the report of the 19th Congress,the concept of "quality first" and "quality power" have been clearly put forward.Control charts are the most direct and effective means of monitoring quality characteristics,and at present,the most widely used and studied is T2 control charts.(1)In this paper,aiming at the dynamic control chart of variable sample size and variable sampling interval,the warning area is increased and the SVSSI control chart is proposed.The statistical indicators and economic indicators are analyzed and compared with FSR,VSI,VSSI,VP T2 control charts.The results show that the statistical performance and economic performance of the dynamic control chart are much better than the traditional static control chart,that is,the dynamic control chart has more ability to monitor the abnormal factors and at the same time save more costs.Different dynamic control charts have different advantages over different offset ranges.(2)The statistical and economic indexes of the control charts with different deviations of the VSSI control chart under unknown covariance conditions are studied.Combined with full factorial experimental design,the effect of model parameters on the EL value is analyzed.Under the condition of unknown covariance,the law of the control chart conforms to the general rule.Through factor analysis,it is found that each parameter has a significant impact on the loss of profits.When controlled,the profit has a positive correlation with it,and the profit and offset size in the event of loss of control are negatively correlated with each other.(3)Based on the effect of conversion times on economic performance,a Markov chain model of VP control chart after the addition of conversion rules is established to study the influence of different conversion parameters on the statistical and economic performance of control charts.Particle swarm optimization algorithm is used to optimize the economic model.Adding conversion rules does effectively reduce the number of transitions to the sampling strategy and increases its ability to monitor anomalies.But at the same time,it will also increase the size of the sample,increase the cost of sampling.Therefore,in actual production,it is necessary to choose the appropriate parameter value according to the demand.