A Variable-Selection-Based Multivariate EWMA Chart for Process Monitoring and Diagnosis
نویسندگان
چکیده
Fault detection and root cause identification are both important tasks in Multivariate Statistical Process Control (MSPC) for improving process and product quality. Most traditional control charts, including Hotelling’s T 2 chart and the Multivariate Exponential Weighted Moving Average (MEWMA) chart, separate the two tasks into independent and successive procedures by signaling the existence of process faults followed by auxiliary methods to locate root causes. This paper proposes an integrated procedure, a Variable-Selection-based MEWMA (VS-MEWMA) chart, for multivariate process monitoring and fault diagnosis by utilizing dimensionality reduction techniques. The VS-MEWMA chart first locates potentially out-of-control variables via variable selection and then deploys such information in the monitoring statistics with the reduction in dimensionality providing increased sensitivity to out-of-control conditions. When a signal is given, the algorithm also identifies the suspected variables for further root cause diagnosis. Both numerical simulations and real examples are presented to illustrate the performance of the proposed chart, as well as design guidelines.
منابع مشابه
Online Monitoring and Fault Diagnosis of Multivariate-attribute Process Mean Using Neural Networks and Discriminant Analysis Technique
In some statistical process control applications, the process data are not Normally distributed and characterized by the combination of both variable and attributes quality characteristics. Despite different methods which are proposed separately for monitoring multivariate and multi-attribute processes, only few methods are available in the literature for monitoring multivariate-attribute proce...
متن کاملA Generalized Linear Statistical Model Approach to Monitor Profiles
Statistical process control methods for monitoring processes with univariate ormultivariate measurements are used widely when the quality variables fit to known probabilitydistributions. Some processes, however, are better characterized by a profile or a function of qualityvariables. For each profile, it is assumed that a collection of data on the response variable along withthe values of the c...
متن کاملMonitoring Fuzzy Capability Index $widetilde{C}_{pk}$ by Using the EWMA Control Chart with Imprecise Data
A manufacturing process cannot be released to production until it has been proven to be stable. Also, we cannot begin to talk about process capability until we have demonstrated stability in our process. This means that the process variation is the result of random causes only and all assignable or special causes have been removed. In complicated manufacturing processes, such as drilling proces...
متن کاملMonitoring Lognormal Reliability Data in a Two-Stage Process Using Accelerated Failure Time Model
The reliability data is getting used to monitor and improve the quality of products or services. Nowadays, most of products or services are the results of processes with dependent stages referred to as multi-stage process. In these processes, the quality characteristics are affected by the quality characteristics in the previous stages, called as cascade property. In some cases, it is not possi...
متن کاملPhase II logistic profile monitoring
In many industrial and non-industrial applications the quality of a process or product is characterized by a relationship between a response variable and one or more explanatory variables. This relationship is referred to as profile. In the past decade, profile monitoring has been extensively studied under the normal response variable, but it has paid a little attention to the profile with the ...
متن کامل