ABSTRACT In industry, processes with multiple streams or gauges in parallel are common. We discuss monitoring such processes to detect changes in both the overall process mean and changes in the individual stream or gauge means. We propose two new control chart statistics based on an F test and a likelihood ratio test. One appealing aspect of these approaches is that they can be implemented either with or without process parameter estimates obtained from previous data (i.e., from phase 1 implementation of the control chart). These proposals are shown to compare favorably to available methods. The article is motivated by a truck assembly process in which wheel alignment characteristics are measured on every truck by one of four alignment machines, arranged in parallel within the overall process.
[1]
H. Leon Harter,et al.
THE PROBABILITY INTEGRALS OF THE RANGE AND OF THE STUDENTIZED RANGE. PROBABILITY INTEGRAL AND PERCENTAGE POINTS OF THE STUDENTIZED RANGE; CRITICAL VALUES FOR DUNCAN'S NEW MULTIPLE RANGE TEST
,
1959
.
[2]
楊素芬.
Statistical process control for multiple stream processes
,
2001
.
[3]
Fred Spiring,et al.
Introduction to Statistical Quality Control
,
2007,
Technometrics.
[4]
Roger M. Sauter,et al.
Introduction to Statistical Quality Control (2nd ed.)
,
1992
.
[5]
Lloyd S. Nelson,et al.
Control Chart for Multiple Stream Processes
,
1986
.
[6]
Tee-Chin Chang,et al.
Monitoring linearity of measurement gauges
,
2006
.