The objective of this article is to carry out a systematic review of the literature on multivariate statistical process control (MSPC) charts used in industrial processes. The systematic review was based on articles published via Web of Science and Scopus in the last 10 years, from 2010 to 2020, with 51 articles on the theme identified. This article sought to identify in which industry the MSPC charts are most applied, the types of multivariate control charts used and probability distributions adopted, as well as pointing out the gaps and future directions of research. The most commonly represented industry was electronics, featuring in approximately 25% of the articles. The MSPC chart most frequently applied in the industrial sector was the traditional T2 of Harold Hotelling (Hotelling, 1947), found in 26.56% of the articles. Almost half of the combinations between the probabilistic distribution and the multivariate control graphs, i.e., 49.4%, considered that the data followed a normal distribution. Gaps and future directions for research on the topic are presented at the end.

The manufacturing and characterization of polymer nanocomposites is an active research trend nowadays. Nonetheless, statistical studies of polymer nanocomposites are not an easy task since they require several factors to consider, such as: large amount of samples manufactured from a standardized procedure and specialized equipment to address characterization tests in a repeatable fashion. In this manuscript, the experimental characterization of sensitivity, hysteresis error and drift error was carried out at multiple input voltages (����) for the following commercial brands of FSRs ( force sensing resistors): Interlink FSR402 and Peratech SP200-10 sensors. The quotient between the mean and the standard deviation was used to determine dispersion in the aforementioned metrics. It was found that a low mean value in an error metric is typically accompanied by a comparatively larger dispersion, and similarly, a large mean value for a given metric resulted in lower dispersion; this observation was held for both sensor brands under the entire range of input voltages. In regard to sensitivity, both sensors showed similar dispersion in sensitivity for the whole range of input voltages. Sensors’ characterization was carried out in a tailored test bench capable of handling up to 16 sensors simultaneously; this let us speed up the characterization process.

Increasing productivity is currently the biggest challenge for manufacturing industries in terms of implementation of Industry 4.0 technologies. This article deals with the widely used methods of measuring of overall equipment effectiveness that in combination with statistical approaches confirms the growth in productivity and seems to be simple and novel technique particularly in the field of printing industries. The aim of the present study is to determine quantitatively the productivity, effectiveness, utilization, risk factor and sigma level of some machines in a printing company that are validated by the selected statistical approaches such as six sigma and analysis of variance techniques. Machine operating time, machine downtime and machine idle-time of different machines in a printing house are considered as main variable parameters for analysis of variance and six-sigma analysis. The results show that the proposed methodology can be a promising development towards improvement of productivity parameters of machines in the printing house.