Semi-supervised CAPP Transformer Learning via Pseudo-labeling

2026-02-01

AI Summary

This study addresses the challenge of data scarcity in high-level computer-aided process planning (CAPP), where limited annotated industrial data hampers model generalization. To overcome this limitation, the authors propose a semi-supervised learning approach that introduces, for the first time in CAPP, an Oracle-guided pseudo-labeling mechanism. Specifically, an Oracle model trained on existing behavioral data evaluates the confidence of Transformer-based predictions on unlabeled parts, selectively retaining high-confidence pseudo-labels for a single round of model retraining. The method is evaluated on a small-scale simulated dataset designed to cover the full data distribution, demonstrating significant performance gains over baseline models. This approach effectively mitigates the data scarcity problem without requiring additional manual annotation, thereby enhancing the practical applicability of CAPP systems in real-world industrial settings.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

High-level Computer-Aided Process Planning (CAPP) generates manufacturing process plans from part specifications. It suffers from limited dataset availability in industry, reducing model generalization. We propose a semi-supervised learning approach to improve transformer-based CAPP transformer models without manual labeling. An oracle, trained on available transformer behaviour data, filters correct predictions from unseen parts, which are then used for one-shot retraining. Experiments on small-scale datasets with simulated ground truth across the full data distribution show consistent accuracy gains over baselines, demonstrating the method's effectiveness in data-scarce manufacturing environments.

Problem

• Computer-Aided Process Planning
• semi-supervised learning
• data scarcity
• model generalization
• limited dataset

Innovation

• semi-supervised learning
• pseudo-labeling
• CAPP
• Transformer
• data scarcity