Hardness and Surface Roughness of 3D-printed ASA Compo-nents Subjected to Acetone Vapor Treatment and Different Production Variables: A Multi-estimation Work via Machine Learning and Deep Learning

This dataset accompanies the research article titled “Hardness and Surface Roughness of 3D-printed ASA Components Subjected to Acetone Vapor Treatment and Different Production Variables: A Multi-estimation Work via Machine Learning and Deep Learning.” The study investigates the combined effects of acetone vapor post-processing and fused deposition modeling (FDM) printing parameters on the hardness and surface roughness of acrylonitrile styrene acrylate (ASA) components. Three process parameters—layer thickness, infill rate, and vaporizing duration—were experimentally varied, and their influence on two key quality indicators (Shore-D hardness and surface roughness Ra) was examined. The dataset contains 108 experimental observations, each representing a unique combination of layer thickness, vaporizing time, and infill rate. For each configuration, the corresponding hardness and surface roughness values are reported. These measurements were obtained using ASTM-compliant hardness testing and stylus-based profilometry. The dataset was used to train and evaluate a broad range of machine learning (ML) and deep learning (DL) models. A hybrid approach integrating 1D-CNN feature extraction with Support Vector Regression (SVR) demonstrated the best predictive performance, achieving an average R² of 0.9614 with a mean squared error of 2.0941 in five-fold cross-validation. This framework successfully enabled simultaneous multi-target prediction of hardness and surface roughness. This repository provides the experimental dataset in Excel format (hardness_roughness.xlsx), enabling replication, comparative analysis, and future methodological extensions in additive manufacturing research, predictive modeling, and data-driven process optimization. Files Included: hardness_roughness.xlsx — Experimental dataset (input parameters + hardness + roughness) Keywords: ASA, FDM, additive manufacturing, surface roughness, hardness, acetone vapor smoothing, machine learning, deep learning, CNN, SVR, multi-target regression