This paper presents YOLO-FIX, an improved intelligent detection model based on YOLOv11, designed to identify glue line defects in mobile phone frames. The model addresses the challenges of complex glue line morphology, background interference, and illumination transformation. YOLO-FIX enhances the extraction of local and global features to optimize the detection accuracy by integrating advanced attention mechanisms and multi-scale feature fusion modules specifically Deformable Large-Kernel Attention (De-formable-LSKA) and Mamba-Like Linear Attention (MLLA). Experimental evaluations demonstrate that YOLO-FIX achieves a mean Average Precision (mAP50) of 95.2%, an 8.6% improvement over the baseline YOLOv11 model while maintaining a real-time detection speed of 189 FPS. It effectively identifies five common defect types: broken glue, wall climbing, glue dropping, single-tip wall climbing, and collapsed glue, showcasing exceptional robustness and generalization across varying production environments. These results affirm YOLO-FIX as a highly accurate and efficient solution for automated defects in industrial applications.

The flexible job-shop scheduling problem with worker flexibility (FJSPW) has been a frequently researched topic in recent years. In FJSPW, the number of workers is generally fewer than the number of machines, and workers are flexible enough to perform operations on multiple machines, thereby improving productivity. However, the consideration of workers' well-being is generally overlooked. Worker well-being is crucial, particularly in manufacturing processes where frequent manual handling of tasks in awkward positions is involved, to prevent fatigue and work-related musculoskeletal disorders. This paper proposes an FJSPW model that considers ergonomic factors (FJSPW-Ergo) to improve workers' productivity and well-being simultaneously. The ergonomic load is measured using the rapid entire body assessment (REBA) score. An improved non-dominated sorting genetic algorithm II (NSGA-II) is employed to solve the FJSPW-Ergo model with a three-layer encoding method, two crossover operators and two mutation methods to minimize the makespan and maximum REBA score (i.e., the highest ergonomic load experienced by any worker). Different combinations of jobs and machines with a variable number of workers are used to demonstrate the effectiveness of the proposed FJSPW-Ergo approach. The results show that FJSPW-Ergo is an effective method for simultaneously achieving productivity and ergonomic relief.