Functional control charts for multivariate quality characteristics

In this vignette we show how to use the funcharts package to apply the methods proposed in Colosimo and Pacella et al. (2010) to build control charts for monitoring a multivariate functional quality characteristic. Let us show how the funcharts package works through an example with the dataset air.

First of all, starting from the discrete data, let us build the multivariate functional data objects of class mfd, see vignette("mfd").

mfdobj <- get_mfd_list(air, grid = 1:24)
nobs <- dim(mfdobj$coefs)[2]
#> [1] 355

plot of chunk unnamed-chunk-2

In order to perform the statistical process monitoring analysis, we divide the data set into a phase I and a phase II dataset.

rows1 <- 1:300
rows2 <- 301:355
mfdobj1 <- mfdobj[rows1]
mfdobj2 <- mfdobj[rows2]

Multivariate functional principal component analysis

For class mfd we provide a function pca_mfd, which is a wrapper to pca.fd. It returns multivariate functional principal component scores summed over variables (fda::pca.fd returns an array of scores when providing a multivariate functional data object). Moreover, the eigenfunctions or multivariate functional principal components given in harmonics argument are converted to the mfd class. We also provide a plot function for the eigenfunctions (the argument harm selects which components to plot).

pca <- pca_mfd(mfdobj1)
plot_pca_mfd(pca, harm = 1:3)

plot of chunk unnamed-chunk-4

Control charts

The function control_charts_pca provides a data frame with all the information required to plot the desired control charts on multivariate functional data, based on the Hotelling’s \(T^2\) and the squared prediction error statistics (see Colosimo and Pacella (2010), Capezza et al. (2020)). Among the arguments, you can pass a tuning_data set, that is not used for model estimation/training, but only to estimate control chart limits. If this data set is not provided, control chart limits are calculated on the basis of the training data. The argument newdata contains the phase II data set of observations of multivariate functional covariates to be monitored. The function plot_control_charts returns the plot of the control charts.

cclist_pca <- control_charts_pca(pca = pca, newdata = mfdobj2)

plot of chunk plot_control_charts_pca

Fault detection

Once an out-of-control observation is identified in a control chart, one can obtain the contribution of the functional covariates to the monitoring statistics to understand which variable was possibly responsible of the anomalous observation.

We can get the out of control observations using which_ooc.

ooc_index <- which_ooc(cclist_pca)
#> $T2
#>    n  id
#> 1  2 302
#> 2  5 305
#> 3  9 309
#> 4 12 312
#> 5 23 323
#> 6 32 332
#> $spe
#>    n  id
#> 1  9 309
#> 2 25 325
#> 3 28 328
#> 4 34 334
#> 5 48 348
#> 6 52 352
#> 7 54 354
#> 8 55 355

From the control charts it is possible to see some phase II observations out of control. For example, let us consider observation 9. We can exploit the contribution plots using the function cont_plot to identify the variables responsible of the anomaly. The contributions are plotted as bar plots, against upper control limits estimated empirically on the training data.

cont_plot(cclist_pca, 9)

plot of chunk cont_plot Finally, we can also plot any multivariate functional observation against a reference data set by using the function plot_mon:

plot_mon(cclist_pca, fd_train = mfdobj1, fd_test = mfdobj2[9])

plot of chunk plot_mon This helps to understand which profiles were anomalous. Variables exceeding contribution plot limits are coloured in red.