Recently many attentions has been placed to process transitions, startups and restarts, since these abnormalities often arrives to the loss of production time, undesirable and not detectable variability increasing with production of off-grade materials. The present research deal with the modeling a continuous soft-drinks bottling process by a multivariate approach based on the 3-Way PLS. Process was modeled, filtering the high grade of autocorrelation and cross-correlation within the studied variables. The responsible variables for that behavior were detected by 3-Way PLS which could be useful to separate autocorrelated sources of variability enable to control this kind of industrial process. (C) 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of 11th International Congress on Engineering and Food (ICEF 11) Executive Committee.