Today, in a smart manufacturing environment based on the Industry 4.0 paradigm, people, technological infrastructure and machinery equipped with sensors can constantly generate and communicate a huge volume of data, also known as Big Data. The manufacturing industry takes advantage of Big Data and analytics evolution by improving its capability to bring out valuable information and knowledge from industrial processes, production systems and sensors. The adoption of model-based frameworks in the Big Data Analytics pipeline can better address user configuration requirements (e.g. type of analysis to perform, type of algorithm to be applied) and also provide more transparency and clearness on the execution of workflows and data processing. In the current state of art, an application of a model-based framework in a manufacturing scenario is missing. Therefore, in this study, by means of a case study research focused on data from sensors associated with Computer Numerical Control machines, the configuration and execution of a Big Data Analytics pipeline with a Model-based Big Data Analytics-as-a-Service framework is described. The case study provides to theoreticians and managerial experts useful evidence for managing real-time data analytics and deploying a workflow that addresses specific analytical goals, driven by user requirements and developer models, in a complex manufacturing domain.
Model-based Big Data Analytics-as-a-Service framework in smart manufacturing: A case study
Corallo A.;Crespino A. M.;Lazoi M.;Lezzi M.
2022-01-01
Abstract
Today, in a smart manufacturing environment based on the Industry 4.0 paradigm, people, technological infrastructure and machinery equipped with sensors can constantly generate and communicate a huge volume of data, also known as Big Data. The manufacturing industry takes advantage of Big Data and analytics evolution by improving its capability to bring out valuable information and knowledge from industrial processes, production systems and sensors. The adoption of model-based frameworks in the Big Data Analytics pipeline can better address user configuration requirements (e.g. type of analysis to perform, type of algorithm to be applied) and also provide more transparency and clearness on the execution of workflows and data processing. In the current state of art, an application of a model-based framework in a manufacturing scenario is missing. Therefore, in this study, by means of a case study research focused on data from sensors associated with Computer Numerical Control machines, the configuration and execution of a Big Data Analytics pipeline with a Model-based Big Data Analytics-as-a-Service framework is described. The case study provides to theoreticians and managerial experts useful evidence for managing real-time data analytics and deploying a workflow that addresses specific analytical goals, driven by user requirements and developer models, in a complex manufacturing domain.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
