Search

This study examines the effect of Industry 4.0 (I4.0) technologies on the learning process of operators. We collected data from the training of new operators in a quality inspection workstation. Two distinct scenarios were considered: before and after the adoption of I4.0 technologies. Data from 10 operators were collected in each scenario; the quality inspection cycle was repeated by each operator 30 consecutive times. A 2-parameter hyperbolic learning curve model was used to assess the learning process in the two groups. Results indicated that operators supported by I4.0 technologies had a significantly higher learning rate than those performing the same tasks without I4.0 support. No significant difference was found in the final performance level between groups. Our study bridges a theoretical gap in the relationship between I4.0 and learning by directly comparing the effect of digital support on the training of new employees in a manufacturing environment. We also offer arguments to support managerial decisions with regards to I4.0 adopti-on at an operational level. That allows organisations to prioritise their digitalisation efforts so that the training of operators in workstations can be expedited. [ABSTRACT FROM AUTHOR]

This study aimed at identifying applications of learning curve (LC) modelling at individual, group, and organisational levels in Industry 4.0 (I4.0) environments. For that, a scoping review on four databases was conducted using the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines. Our results indicated that LCs are more prominently adopted in I4.0 to model learning at the individual level using technologies oriented to sensing and communication (e.g. big data, IoT, wireless sensors, cloud computing, remote control, or monitoring). However, the effect of a few processing and actuation technologies, such as augmented/virtual reality, collaborative robots, and machine learning/AI, on learning seems promising. Further, despite the number of studies investigated, few explicitly described the LC model used to represent the impact of I4.0 technologies on learning. Our findings allowed the proposition of five research directions. Literature on both LC and I4.0 is still fragmented, poorly addressing their relationship. As I4.0 is an innovative approach that allows more extensive information exchange and processing, new ways of using I4.0 technologies to expedite data collection, which has always constrained LC practical applications, should be devised to close the gap between I4.0 and learning. [ABSTRACT FROM AUTHOR]

Spectral data describing product samples are typically composed of a large number of noisy and irrelevant wavelengths that tends to undermine the performance of multivariate predictive techniques. This paper proposes a two‐phase framework that integrates a preselection wavelength step oriented by wavelength clustering to a wrapper‐based strategy. The first phase performs a pruning process in the data that removes the less informative wavelengths relying on the spectral clustering, a technique deemed suitable to the Fourier transform infrared (FTIR) spectroscopy and near‐infrared (NIR) spectroscopy data at hand. The preselected wavelengths undergo a second phase of selection efforts based on the combination of different wavelength importance indices (i.e., Bhattacharyya distance, Chi‐square, ReliefF, and Gini) and classification techniques (i.e., support vector machine, k‐nearest neighbors, and random forest). When applied to 11 FTIR datasets from different domains, the recommended combination of importance index and classifier increased the average accuracy by 6.37% (from 0.863 to 0.918), while retaining average 3.84% of the original spectra. The framework also improved the selection process regarding computational time. This paper introduces a two‐phase framework merging wavelength preselection through clustering and a wrapper strategy. Initial spectral clustering eliminates less informative wavelengths. Subsequently, diverse wavelength importance indices and classification methods are integrated. Applied to 11 spectral datasets, the proposed combination (spectral clustering [SC]‐random forest [RF]+Gini [GI]) enhances average accuracy by 6.37%, retaining 3.84% of the original spectra, and reduces computational time. [ABSTRACT FROM AUTHOR]