In the additive manufacturing laser powder bed fusion (L-PBF) process, the optimization of the print process parameters and the development of conduction zones in the laser power (P) and scanning speed (V) parameter spaces are critical to meeting production quality, productivity, and volume goals. In this paper, we propose the use of a machine learning approach during the process parameter development to predict the melt pool dimensions as a function of the P/V combination. This approach turns out to be useful in speeding up the identification of the printability map of the material and defining the conduction zone during the development phase. Moreover, a machine learning method allows for an accurate investigation of the most promising configurations in the P-V space, facilitating the optimization and identification of the P-V set with the highest productivity. This approach is validated by an experimental campaign carried out on samples of Inconel 718, and the effects of some additional parameters, such as the layer thickness (in the range of 30 to 90 microns) and the preheating temperature of the building platform, are evaluated. More specifically, the experimental data have been used to train supervised machine learning models for regression using the KNIME Analytics Platform (version 4.7.7). An AutoML (node for regression) tool is used to identify the most appropriate model based on the evaluation of R² and MAE scores. The gradient boosted tree model also performs best compared to Rosenthal’s analytical model.

Industry 4.0 has introduced a data-driven model of production and management of goods and services. This manufacturing paradigm leverages the potential of the Internet of Things (IoT), but finding the information necessary to drive manufacturing processes can be challenging. In this context, the authors propose an innovative approach based on axiomatic design to design RDF knowledge graphs from which to extract the information needed by decision makers. This approach derives from the possibility of providing RDF knowledge graphs with an equivalent matrix representation based on axiomatic design. It allows the selection of the most reliable data sources, thereby optimizing the knowledge graph construction process using matrix algebra, minimizing redundancy and improving the efficiency of query response. The goal of the presented methodology is to address the five critical aspects of Big Data (volume, velocity, variety, value, and veracity) by preordering the knowledge graph according to the information needs of business decision makers, thereby optimizing the use of the immense wealth of information made available by the Web in design.