XAI–LCA random forest models and datasets for PLA material extrusion (MEX)

This deposition contains the code, trained random forest models, processed datasets, and STL geometries used in the study: “Towards Sustainable Material-Extrusion Additive Manufacturing: Explainable AI Analysis of Printing Parameter Effects on Life Cycle Assessment”Yanlong Zhu et al.,. The materials are organised into three main components: 1. 01_model_building - 134-sample training datasets for the Direct-RFM, EC-RFM and MC-RFM models (Excel). - Trained random forest models (Direct-RFM.pkl, EC-RFM.pkl, MC-RFM.pkl). - The 434-sample dataset (predictions_434_samples.xlsx) combining 134 semi-real prints and 300 virtual LHS samples with model predictions for four ReCiPe 2016 Midpoint (H) indicators (GWP, FRS, OFHH, TA). - Python scripts (ecmc_model.py, predict_lca.py) that implement the EC/MC decomposition model and a command-line interface for running predictions. 2. 02_figure_generation - Processed data tables (CSV/XLSX) and Python plotting scripts used to reproduce Figures 6–12 in the paper, including: predicted vs. actual plots, Direct-RFM vs. EC/MC-RFM comparison, SHAP importance, response surfaces, PDP-style curves, Bayesian optimisation convergence, optima scatter plots, perturbation analysis, ratio-to-BO-optimum heatmaps, and cost-per-print and material-transfer bar charts. - The scripts rely on standard Python packages (pandas, numpy, matplotlib, seaborn, scipy, scikit-learn, joblib) and load the trained models from 01_model_building where required. 3. geometries (STL files) - rect_specimen_180x15x2p4mm.stl: standard rectangular tensile specimen used in the study. - 3DBenchy_calibration_boat.stl: the “3DBenchy” calibration geometry used for benchmarking and illustration. The raw printer G-code files and full SHAP analysis scripts are not included due to size and machine-specific dependencies, but can be shared upon reasonable request. This dataset is intended to support full reproducibility of the XAI–LCA workflow and to facilitate reuse of the models and figures in related research on sustainable material-extrusion additive manufacturing.