Journal of Engineering, Project, and Production Management, 2026, 16(3), 2026-199
Assessing Vocational Students’ Gemstone Operation Using Internet of Things and ST-GCN Graph Modeling
Lecturer, Anhui
Technical College of Industry and Economy, Hefei, 230051, China, E-Mail:
zhouzhengzzz22@163.com
Project Management
Received February 7, 2026; revised March 20, 2026; accepted April 18, 2026
Available online April 25, 2026
Abstract: A skeleton spatiotemporal graph-based modeling and training skill evaluation management system for the industrial Internet of Things (IoT) is proposed to address subjectivity and the lack of quantitative analysis in skill evaluation during vocational jade processing training. Firstly, a perceptual environment that integrates vision and inertia is constructed, and an improved High Resolution Network (HRNet) model that combines L-Basicblock and Convolutional Block Attention Module (CBAM) is proposed to address hand occlusion and extract skeletons. Subsequently, a skill assessment model is designed that combines the improved Spatial Temporal Graph Convolutional Network (ST-GCN) with the Gated Recurrent Unit (GRU). The system uses joint spatial position deviation in motion prediction and the ergonomic angle error of the hand as evaluation criteria, forming a quantifiable skill-scoring system. The results show that the Average Precision (AP) of the improved HRNet model reaches 92.8%, achieving a balance between accuracy and efficiency. The average joint position error of the evaluation model in the 400ms prediction time domain is as low as 28.1mm, which is significantly better than the 38.6mm of the Simple Multi-Layer Perceptron (SiMLPe) model. In addition, the median scores for the model in experts and novices are 95.5 and 54.0, respectively, with clear hierarchical discrimination. In actual teaching verification, the system sets a score decision threshold, allowing teachers to intuitively locate student’s specific deduction points, effectively reducing teaching communication costs and consumables consumption. The system proposed by the research achieves objective quantification of gemstone operation behavior, which not only provides effective support for standardized skill certification and teaching decision optimization of precision manufacturing majors in higher vocational education, but also provides scalable management tools for the production preparation of high-quality labor under the background of Industry 4.0.
Keywords: Gemstone operation, ST-GCN, GRU, HRNet, higher vocational talents, wireless perception, skill assessment, workforce training, vocational education management, digital manufacturing training.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Zhou, Z. (2026). Assessing Vocational Students' Gemstone Operation Using Internet of Things and ST-GCN Graph Modeling. Journal of Engineering, Project, and Production Management, 16(3), 2026-199.
