Scholarlink

<p>The rapid advancement of additive manufacturing technologies, particularly VAT photopolymerization (VPP), has transformed how we fabricate complex geometries and custom parts. However, the lack of standardized metrics for design complexity creates significant challenges in manufacturability, cost optimization, and performance enhancement. This study tackles these issues by introducing a hybrid framework for the Additive Manufacturing Complexity Index (AMCI) that integrates the Analytical Hierarchy Process (AHP) with machine learning (ML). The framework combines expert-driven AHP weighting with ML-driven validation, addressing the limitations of traditional single-method approaches. A case study involving 100 diverse parts demonstrates the framework’s effectiveness, with the Random Forest Regressor achieving an <i>R</i><sup>2</sup> of 0.93 and RMSE of 3.45. Geometric attributes (e.g., volume, surface area) and manufacturing parameters (e.g., exposure time, part orientation) significantly influence complexity. Key findings point out significant influences of the geometric attributes, i.e., volume and surface area, on complexity. At the same time, machine settings, such as bottom exposure time and part orientation angle, could influence manufacturability factors. The AMCI offers a robust, data-informed tool for optimizing production.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hybrid framework for assessing additive manufacturing complexity index: integration of analytical hierarchy process and machine learning for VAT photopolymerization

  • 期刊论文

  • 发表日期：

    2025年6月29日

查看全文
  • Dhal A. Matoc,
  • Nikunj Maheta,
  • Bhavesh Kanabar,
  • Amit Sata

摘要

The rapid advancement of additive manufacturing technologies, particularly VAT photopolymerization (VPP), has transformed how we fabricate complex geometries and custom parts. However, the lack of standardized metrics for design complexity creates significant challenges in manufacturability, cost optimization, and performance enhancement. This study tackles these issues by introducing a hybrid framework for the Additive Manufacturing Complexity Index (AMCI) that integrates the Analytical Hierarchy Process (AHP) with machine learning (ML). The framework combines expert-driven AHP weighting with ML-driven validation, addressing the limitations of traditional single-method approaches. A case study involving 100 diverse parts demonstrates the framework’s effectiveness, with the Random Forest Regressor achieving an R2 of 0.93 and RMSE of 3.45. Geometric attributes (e.g., volume, surface area) and manufacturing parameters (e.g., exposure time, part orientation) significantly influence complexity. Key findings point out significant influences of the geometric attributes, i.e., volume and surface area, on complexity. At the same time, machine settings, such as bottom exposure time and part orientation angle, could influence manufacturability factors. The AMCI offers a robust, data-informed tool for optimizing production.