Liu Liu

Dr. Liu Liu


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 February 2023 in Sensors
Verified
Journal Article
Open access
PDF file
LLLiu Liu
Liu Liu
FGFeng Gao
Feng Gao
YLYan Li
Yan Li
Owing to the remarkable development of deep learning algorithms, defect detection techniques based on deep neural networks have been extensively applied in industrial production. Most existing surface defect detection models assign equal costs to the classification errors among different defect categories but do not strictly distinguish them. However, various errors can generate a great discrepancy in decision risk or classification costs and then produce a cost-sensitive issue that is crucial to the manufacturing process. To address this engineering challenge, we propose a novel supervised classification cost-sensitive learning method (SCCS) and apply it to improve YOLOv5 as CS-YOLOv5, where the classification loss function of object detection was reconstructed according to a new cost-sensitive learning criterion explained by a label–cost vector selection method. In this way, the classification risk information from a cost matrix is directly introduced into the detection model and fully exploited in training. As a result, the developed approach can make low-risk classification decisions for defect detection. It is applicable for direct cost-sensitive learning based on a cost matrix to implement detection tasks. Using two datasets of a painting surface and a hot-rolled steel strip surface, our CS-YOLOv5 model outperforms the original version with respect to cost under different positive classes, coefficients, and weight ratios, but also maintains effective detection performance measured by mAP and F1 scores.

14 August 2025 in Mathematics
Verified
Journal Article
Open access
PDF file
LLLiu Liu
Liu Liu
YLYan Li
Yan Li
FGFeng Gao
Feng Gao
Discrete manufacturing environments face increasing challenges in managing work-in-process (WIP) inventory due to growing product customization and demand volatility. While Value Stream Mapping (VSM) has been widely used for process improvement, traditional approaches lack the ability to dynamically control WIP levels while optimizing multiple performance dimensions simultaneously. This research addresses this gap by developing an integrated framework that synergizes Multi-Dimensional Value Stream Mapping (MD-VSM) with multi-objective optimization, functioning as a specialized digital twin for dynamic WIP control. The framework employs a four-layer architecture that connects real-time data collection, multi-dimensional modeling, dynamic WIP monitoring, and execution control through closed-loop feedback mechanisms. A mixed-integer optimization model is used to balance time, cost, and quality objectives. Validation using a high-fidelity simulation, parameterized with real-world industrial data, demonstrates that the proposed approach yielded up to a 31% reduction in inventory costs while maintaining production throughput and showed a 42% faster recovery from equipment failures compared to traditional methods. Furthermore, a comprehensive sensitivity analysis confirms the framework’s robustness. The system demonstrated stable performance even when key operational parameters, such as WIP upper limits and buffer capacity coefficients, were varied by up to ±30%, underscoring its reliability for real-world deployment. These findings provide manufacturers with a validated methodology for enhancing operational efficiency and production flexibility, advancing the integration of lean principles with data-driven, digital twin-based control systems.