\name{pois.cusum.crit} \alias{pois.cusum.crit} \title{Compute alarm thresholds and randomization constants of Poisson CUSUM control charts} \description{Computation of the CUSUM upper limit and, if needed, of the randomization probability, given mean \code{mu0}.} \usage{pois.cusum.crit(mu0, km, A, m, i0=0, sided="upper", rando=FALSE)} \arguments{ \item{mu0}{actual in-control mean.} \item{km}{enumerator of rational approximation of reference value \code{k}.} \item{A}{target in-control ARL (average run length).} \item{m}{denominator of rational approximation of reference value.} \item{i0}{head start value as integer multiple of \code{1/m}; should be an element of \code{0:100} (a more reasonable upper limit will be established soon). It is planned, to set \code{i0} as a fraction of the final threshold.} \item{sided}{distinguishes between different one- and two-sided CUSUM control chart by choosing \code{"upper"}, \code{"lower"} and \code{"two"}, respectively.} \item{rando}{Switch for activating randomization in order to allow continuous ARL control.} } \details{ The monitored data follow a Poisson distribution with \code{mu} (here the in-control level \code{mu0}). The ARL values of the resulting EWMA control chart are determined via Markov chain calculations. With some grid search, we obtain the smallest value for the integer threshold component \code{hm} so that the resulting ARL is not smaller than \code{A}. If equality is needed, then activating \code{rando=TRUE} yields the corresponding randomization probability \code{gamma}. More details will follow in a paper that will be submitted in 2020. } \value{Returns two single values, integer threshold \code{hm} resulting in the final alarm threshold \code{h=hm/m}, and the randomization probability.} \references{ J. M. Lucas (1985) Counted data CUSUM's, \emph{Technometrics} 27(2), 129-144. C. H. White and J. B. Keats (1996) ARLs and Higher-Order Run-Length Moments for the Poisson CUSUM, \emph{Journal of Quality Technology} 28(3), 363-369. C. H. White, J. B. Keats and J. Stanley (1997) Poisson CUSUM versus c chart for defect data, \emph{Quality Engineering} 9(4), 673-679. G. Rossi and L. Lampugnani and M. Marchi (1999), An approximate CUSUM procedure for surveillance of health events, \emph{Statistics in Medicine} 18(16), 2111-2122. S. W. Han, K.-L. Tsui, B. Ariyajunya, and S. B. Kim (2010), A comparison of CUSUM, EWMA, and temporal scan statistics for detection of increases in poisson rates, \emph{Quality and Reliability Engineering International} 26(3), 279-289. M. B. Perry and J. J. Pignatiello Jr. (2011) Estimating the time of step change with Poisson CUSUM and EWMA control charts, \emph{International Journal of Production Research} 49(10), 2857-2871. } \author{Sven Knoth} \seealso{later.} \examples{ ## Lucas 1985 mu0 <- 0.25 km <- 1 A <- 430 m <- 4 #cv <- pois.cusum.crit(mu0, km, A, m) cv <- c(40, 0) # Lucas reported h = 10 alias hm = 40 (in Table 2, first block, row 10.0 .25 .0 ..., column 1.0 # Recall that Lucas and other trigger an alarm, if the CUSUM statistic is greater than # or equal to the alarm threshold h print(cv) } \keyword{ts}