The potential of implementing AI-driven quality control in Ukrainian investment casting facilities

Abstract

The purpose of this study is to assess how AI can reduce energy consumption, labor intensity, and scrap rates, thereby improving yield and long-term operational efficiency of investment casting foundries. The methodology includes a literature review and feasibility analysis conducted using recent academic studies and industry case reports from 2013 to 2024. Additionally, the study conducted a basic cost-benefit analysis comparing implementation expenses with potential annual savings in scrap reduction, labor optimization, and material efficiency. Findings indicate that key AI applications include process-parameter modeling and machine learning prediction, and automated defect detection through deep learning-based visual and radiographic inspection. Research shows that AI systems can reduce casting defects by 30–50%, with substantial savings in labor and material costs. The study highlights low-cost and open-source options for AI deployment, increasing accessibility for resource-constrained facilities. The originality of the paper is its emphasis on the practical implementation of AI-driven quality control solutions for Ukrainian foundries, investment casting facilities in particular. The practical value of the study lies in a structured, actionable roadmap, including software and hardware requirements, and cost and ROI estimates, that can assist local foundries in beginning their Industry 4.0 transition with a focus on quality optimization.