A Comparative Study of VR and 2D Plans in Concrete Structure Evaluation for Engineering Education: Does Perceived Usability Affect Performance?

Traditional engineering education often struggles to effectively teach practical construction inspection skills through conventional 2D drawings. This study investigates the effectiveness of different visualization methods-traditional 2D plans, Desktop-Based Virtual Reality (DB-VR), and Head-Mounted Display Virtual Reality (HMD-VR)-in helping civil engineering students identify construction mistakes in concrete structures. Sixty-nine second-year undergraduate students were randomly assigned to three groups and tasked with identifying twelve predetermined structural mistakes using their assigned visualization method. Performance metrics (accuracy, and completion time), system usability, and prior experience were evaluated. Results showed that both VR methods significantly outperformed traditional approaches in accuracy (57.5% vs 24.2% identification rate) and efficiency (7.56-11.13 vs 15 minutes). HMD-VR users demonstrated the lowest wrongly identified mistakes (mean = 1.91 vs 3.17 and 3.00) and fastest completion times. Both VR methods achieved “Good” usability ratings, significantly higher than traditional methods. Notably, performance and usability were not correlated with prior experience, suggesting broad applicability across different user groups. The findings support the integration of VR technologies, particularly HMD-VR, into engineering education as an effective tool for developing practical inspection skills in a risk-free environment.