Enhancing efficiency and environmental performance of laser-cutting machine tools - An explainable machine learning approach

Aug 3, 2026

Upload by Riley

This paper investigates the integration of sustainability and environmental performance in the development of machine tools. It uses a literature review and a case study on a fully automated solid-state laser-cutting system to develop a data-driven, explainable machine learning approach for optimizing operations. The study models resource consumption, selects key factors using correlation analysis and stepwise linear regression, and predicts outcomes with random forest regression. SHAP is applied to interpret model behavior, highlighting operational duration and part contour length as critical drivers, and motivating sustainability-focused optimization in design and production planning.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Introduction

Literature Review

Data-driven Sustainability in Product Generation Engineering

[Available online at www.sciencedirect.com](Available online at www.sciencedirect.com) ScienceDirect Procedia CIRP 130 (2024) 1674–1679 57th CIRP Conference on Manufacturing Systems 2024 (CMS 2024) Enhancing efficiency and environmental performance of laser-cutting machine tools: An explainable machine learning approach Artur Krausea*, Tobias Dannerbauerb, Steffen Wagenmannc, Greta Tjadend, Robin Ströbelb, Jürgen Fleischerb, Albert Albersc, Nikola Bursaca, aISEM, Hamburg Institute of Technology, 21073 Hamburg, Germany bwbk, Karlsruhe Institute of Technology, 76131 Karlsruhe, Germany cIPEK, Karlsruhe Institute of Technology, 76131 Karlsruhe, Germany dLFU, Technical University Dortmund, 44227, Dortmund, Germany * Corresponding author. Tel.: +49-40-42878-6110 . E-mail address: [artur.krause@trumpf.com](artur.krause@trumpf.com) Abstract This paper investigates the integration of sustainability and environmental performance in the development of machine tools. Through a literature review and a case study involving a fully automated solid-state laser-cutting machine tool, this research explores the potential of a data-driven, explainable machine learning (XML) approach to optimize machine tool operations for sustainability. Specifically, it addresses modeling resource consumption in laser-cutting machines, identifies the most significant factors influencing these models, and examines their contributions toward sustainable machine operation practices. Employing a combination of correlation analysis and stepwise linear regression for feature selection, and utilizing random forest regression models for predictive analysis, this study reveals that operational duration significantly impacts resource consumption levels, more than the effects of machine and laser configuration settings. The utilization of Shapely Additive Explanations (SHAP) further elucidates the predictive behaviors of these models, emphasizing the critical role of part contour length in program run duration and resource consumption. The findings suggest that optimizing machine speed and production planning, as well as incorporating part contour length as a sustainability Key Performance Indicator (KPI) during the design phase, can enhance the environmental performance of laser-cutting machines. Further, the analysis of resource consumption patterns of machine tools also offers actionable strategies to improve their sustainability during operation. It underscores the importance of integrating sustainability considerations into the development and operational phases of machine tools, contributing valuable insights to the field. © 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([https://creativecommons.org/licenses/by-nc-nd/4.0](https://creativecommons.org/licenses/by-nc-nd/4.0)) Peer-review under responsibility of the scientific committee of the 57th CIRP Conference on Manufacturing Systems 2024 (CMS 2024) Keywords: Data-driven Sustainability; Industrial Energy; Explainable Machine Learning; Machine Tool Optimization 1. Introduction The integration of Industry 4.0 technologies, particularly laser-cutting systems, has significantly advanced manufacturing efficiency through high-precision and optimal utilization of material. However, these technologies also pose environmental challenges due to their high energy consumption and therefore emissions. In Germany, the industrial sector accounts for approx. 29% of total energy use in 2021, where machinery contributes significantly to CO2 emissions [1] . This underscores the need for sustainable manufacturing practices, especially in energy-intensive sectors like laser cutting. The European Corporate Sustainability Reporting Directive (CSRD) further emphasizes the need for transparent environmental impact reporting, pushing industries toward sustainable practice. Leveraging Industry 4.0's data analytics could optimize resource use an

FAQs

What sustainability goal does the paper target in laser-cutting machine tools?

It targets minimizing resource consumption to improve environmental performance in laser-cutting machine tool operations, aligning efficiency gains with sustainability needs.

How are key factors selected for modeling resource consumption?

The study applies correlation analysis and stepwise linear regression for feature selection, then builds predictive models to relate influential factors to resource consumption.

Which explainability method helps interpret the predictive models?

Shapely Additive Explanations (SHAP) is used to clarify how model features contribute to predictions, emphasizing drivers such as part contour length and operational duration.

Aug 3, 2026

Upload by Riley