Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Control charts for health care monitoring under intermittent Out-of-Control behavior

Summary: 1
Control charts for health care monitoring under
intermittent Out-of-Control behavior
Willem Albers,Department of Applied Mathematics, University
of Twente, P.O.Box 217, 7500 AE Enschede,The Netherlands
Abstract. Health care monitoring typically concerns attribute data with very low failure rates. Efficient
control charts then signal if the waiting time till r (e.g. r 5) failures is too small. An interesting
alternative is the MAX-chart, which signals if all the associated r waiting times for a single failure are
sufficiently small. In comparing these choices, the usual change point set-up has been used, in which
going Out-of-Control (OoC) means that the failure rate suddenly jumps up and then stays at this higher
level. However, another situation of interest is intermittent OoC behavior. In industrial settings, an OoC
process can be adjusted to return to In-Control (IC), but with health care monitoring this usually is no
option and stretches of OoC and IC behavior may alternate. Comparison of such intermittent
alternatives to the change point situation shows that the former can be characterized as tail alternatives,
in the sense that the difference w.r.t. the IC-distribution becomes more concentrated in the lower tail.
This suggests to generalize the MAX-chart as follows: now signal if all but 1 (or 2) out of r individual
waiting times are too small. A numerical study shows that this approach indeed works well.
Keywords: Statistical Process Control, high-quality processes, tail alternatives, average run length
2000 Mathematics Subject Classification: 62P10, 62C05, 62G15
1. Introduction and motivation


Source: Al Hanbali, Ahmad - Department of Applied Mathematics, Universiteit Twente


Collections: Engineering