The Industry 4.0 paradigm, since its initial conception in Germany in 2011, has extended its scope and adoption to a broader set of technologies. It is being considered as the most vital mechanism in the production systems lifecycle. It is the key element in the digital transformation of manufacturing industry all over the world. This scenario imposes a set of major unprecedented challenges which require to be overcome. In order to enable integration in horizontal, vertical, and end-to-end formats, one of the most critical aspects of this digital transformation process consists of effectively coupling digital integrated service/products business models with additive manufacturing processes. This integration is based upon advanced AI-based tools for decentralized decision-making and for secure and trusted data sharing in the global value. This paper presents the FASTEN IIoT Platform, which targets to provide a flexible, configurable, and open solution. The platform acts as an interface between the shop floor and the industry 4.0 advanced applications and solutions. Examples of these efforts comprise management, forecasting, optimization, and simulation, by harmonizing the heterogeneous characteristics of the data sources involved while meeting real-time requirements.

The naïve Bayes classifier is considered one of the most effective classification algorithms today, competing with more modern and sophisticated classifiers. Despite being based on unrealistic (naïve) assumption that all variables are independent, given the output class, the classifier provides proper results. However, depending on the scenario utilized (network structure, number of samples or training cases, number of variables), the network may not provide appropriate results. This study uses a process variable selection, using the chi-squared test to verify the existence of dependence between variables in the data model in order to identify the reasons which prevent a Bayesian network to provide good performance. A detailed analysis of the data is also proposed, unlike other existing work, as well as adjustments in case of limit values between two adjacent classes. Furthermore, variable weights are used in the calculation of a posteriori probabilities, calculated with mutual information function. Tests were applied in both a naïve Bayesian network and a hierarchical Bayesian network. After testing, a significant reduction in error rate has been observed. The naïve Bayesian network presented a drop in error rates from twenty five percent to five percent, considering the initial results of the classification process. In the hierarchical network, there was not only a drop in fifteen percent error rate, but also the final result came to zero.

In this article, we present a comprehensive survey on time-sensitive applications implemented in fog computing environments. The goal is to research what applications are being implemented in fog computing architectures and how the temporal requirements of these applications are being addressed. We also carried out a comprehensive analysis of the articles surveyed and separate them into categories, according to a pattern found in them. Our research is important for the area of real-time systems since the concept of systems that respond in real time has presented various understandings and concepts. This variability of concept has been due to the growing requirements for fast data communication and processing. Therefore, we present different concepts of real-time and near real-time systems found in the literature and currently accepted by the academic-scientific community. Finally, we conduct an analytical discussion of the characteristics and proposal of articles.

No abstract

No abstract