Answers are grounded in this paper and cite their sources.
Contemporary manufacturing environments have shifted from intuition-based decisions to data-driven strategies, particularly those related to quality control. Quality remains central to manufacturing competitiveness; therefore, early detection of quality issues is important for minimizing rework, delays, and costs. This article proposes a modular data-driven quality management system (DD-QMS) that integrates 3-D scanning, machine learning, and near real-time decision support to enable proactive quality intervention during manufacturing. The framework follows a modular workflow in which high-resolution 3-D scans are converted into structured deviation profiles and analyzed through neural network models to classify components into acceptable, reworkable, and reject categories with real-time decision support. The DD-QMS is demonstrated through a case study of crankshaft manufacturing, in which the system detects dynamic imbalances and dimensional deviations early in the process. This workflow enables early visibility into deviations and complements existing inspection practices. This pilot implementation demonstrated that the DD-QMS improved root cause traceability, reduced rejection costs, and enabled faster decision cycles. The framework supports a strategic transition from reactive inspection to real-time self-regulating quality systems, aligning with the goals of smart manufacturing and Quality 4.0. Its modular design allows scalable implementation, offering practical value for both advanced and digitally transitioning manufacturing setups.
No takes yet. Share an insight, caveat, or question.
Kulkarni et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: