Industry-Guided Optimized Rebar Spacing Quality Control using Augmented Reality

Accurate rebar placement is essential for ensuring the structural performance and durability of reinforced concrete components, particularly in construction industry. In practice, rebar inspection is commonly performed using manual measurements and two-dimensional drawings, resulting in subjective interpretation, inefficiencies, and physically demanding inspection workflows. This study presents an augmented reality (AR)–based rebar inspection and correction system developed to support practical inspection and decision-making tasks. The system integrates RGBD sensing, automated Building Information Modeling (BIM)–based spacing analysis, greedy correction sequencing, and motion-capture-based ergonomic assessment within a unified workflow. As-built rebar geometry is captured using an Azure Kinect depth camera and registered with the design BIM model to identify spacing deviations. Inspection results are presented in situ through a Microsoft HoloLens 2 headset using spatially anchored, color-coded AR overlays, while a greedy optimization algorithm generates step-by-step correction guidance intended to minimize unnecessary bar adjustments and physical effort during inspection. Real-time TCP/IP communication enables low-latency transfer of analytical results to the AR interface, supporting interactive inspection without offline processing. A controlled laboratory study compares conventional manual inspection procedures with AR-assisted workflows using a Vicon motion capture system to quantify differences in task execution and inspector movement. The results demonstrated descriptive evidence of differences in task duration, movement patterns, direction changes, and trunk and neck postures between inspection approaches. The findings demonstrate the practical potential of an integrated digital workflow to enhance quality control in construction, addressing industry-identified needs for objective measurement, reduced physical effort, and guided decision-making in repetitive inspection tasks.