On multivariate control charts

Frisén, Marianne

doi:10.1590/s0103-65132011005000010

Cited by 12 publications

(15 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One advantage of multivariate parallel surveillance is that the interpretation of alarms is clear. Other methods, like Hotelling T 2 control chart, are not capable of distinguishing a change in the mean vector from a change in the covariance structure for example [69]. …”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Methodological challenges to multivariate syndromic surveillance: a case study using Swiss animal health data

2016

View full text Add to dashboard Cite

BackgroundIn an era of ubiquitous electronic collection of animal health data, multivariate surveillance systems (which concurrently monitor several data streams) should have a greater probability of detecting disease events than univariate systems. However, despite their limitations, univariate aberration detection algorithms are used in most active syndromic surveillance (SyS) systems because of their ease of application and interpretation. On the other hand, a stochastic modelling-based approach to multivariate surveillance offers more flexibility, allowing for the retention of historical outbreaks, for overdispersion and for non-stationarity. While such methods are not new, they are yet to be applied to animal health surveillance data. We applied an example of such stochastic model, Held and colleagues’ two-component model, to two multivariate animal health datasets from Switzerland.ResultsIn our first application, multivariate time series of the number of laboratories test requests were derived from Swiss animal diagnostic laboratories. We compare the performance of the two-component model to parallel monitoring using an improved Farrington algorithm and found both methods yield a satisfactorily low false alarm rate. However, the calibration test of the two-component model on the one-step ahead predictions proved satisfactory, making such an approach suitable for outbreak prediction. In our second application, the two-component model was applied to the multivariate time series of the number of cattle abortions and the number of test requests for bovine viral diarrhea (a disease that often results in abortions). We found that there is a two days lagged effect from the number of abortions to the number of test requests. We further compared the joint modelling and univariate modelling of the number of laboratory test requests time series. The joint modelling approach showed evidence of superiority in terms of forecasting abilities.ConclusionsStochastic modelling approaches offer the potential to address more realistic surveillance scenarios through, for example, the inclusion of times series specific parameters, or of covariates known to have an impact on syndrome counts. Nevertheless, many methodological challenges to multivariate surveillance of animal SyS data still remain. Deciding on the amount of corroboration among data streams that is required to escalate into an alert is not a trivial task given the sparse data on the events under consideration (e.g. disease outbreaks).

show abstract

Section: Discussionmentioning

confidence: 99%

“…Multivariate SPC methods [69], such as MEWMA [70] (Lowry, Woodall, Champ, and Rigdon, 1992) and MCUSUM [19, 71] fall into this category. They are based on the assumption that the data or their residuals follow the normal distribution although a rank-based method or non-parametric scheme are discussed in [72, 73].…”

Section: Methodsmentioning

confidence: 99%

Methodological challenges to multivariate syndromic surveillance: a case study using Swiss animal health data

2016

View full text Add to dashboard Cite

show abstract

“…In order to monitor a multivariate process over time, the well-known Hotelling's T 2 control chart was established by Hotelling in his pioneering article (see Hotelling 1947). Many authors have focused on developing multivariate control charts for monitoring shifts in process mean and/or changes in a covariance matrix, for further details see for instance Lowry et al (1992), Chua and Montgomery (1992), Lowry and Montgomery (1995), Mason et al (1997), Young (1999, 2002), Mason, Chou, and Young (2001), Champ and Jones-Farmer (2005), Yeh, Lin, and McGrath (2006), Bersimis, Psarakis, and Panaretos (2006), Chen and Hsieh (2007), Frisen (2011).…”

Section: Introductionmentioning

confidence: 99%

Monitoring the ratio of two normal variables using Run Rules type control charts

Tran

Castagliola

Celano

2015

International Journal of Production Research

View full text Add to dashboard Cite

Recent studies show that Shewhart-type control charts monitoring the ratio of two normal random variables are useful to perform continuous surveillance in several manufacturing environments; anyway, they have a poor statistical sensitivity in the detection of small or moderate process shifts. The statistical sensitivity of a Shewhart control chart can be improved by implementing supplementary Run Rules. In this paper, we investigate the performance of Phase II Run Rules Shewhart control charts monitoring the ratio with each subgroup consisting of n > 1 sample units. A Markov chain methodology coupled with an efficient normal approximation of the ratio distribution is used to evaluate the statistical performance of these charts. We provide an extensive numerical analysis consisting of several tables and figures to discuss the statistical performance of the investigated charts for deterministic and random shift sizes affecting the in-control ratio. An illustrative example from the food industry is provided for illustration.

show abstract

“…With reference to multivariate SPC, when p ‐correlated variables should be monitored over time, the well‐known Hotelling's T 2 control chart is the counterpart to the Shewhart control chart; for further details, see Hotelling and Mason and Young . Review papers discussing multivariate charts monitoring the ( p × 1) process mean vector μ and/or the ( p × p ) variance–covariance matrix Σ are available from Alt, Lowry and Montgomery, Yeh et al , Bersimis et al , and Frisen . More recently, Hotelling's T c 2 control charts have been proposed by Vives‐Mestres et al for monitoring the product composition of a mixture: the authors explain how to correctly implement these charts in the presence of the restricted sum of components and suggest a method to interpret the out‐of‐control signals.…”

Section: Introductionmentioning

confidence: 99%