RULA, REBA and OWAS real-time software model in an academic environment

Resumen

Manual ergonomic assessment can be slow and highly dependent on the evaluator's criteria. This study developed and tested computer vision software with two cameras to analyze postures in real time using the RULA, REBA, and OWAS methods. The system's results were compared with those of an expert in 75 measurements of five postures. The program worked stably at 30 FPS and achieved a 68.0% to 76.0% match in risk levels. In addition, it was found that the software almost always overestimates risk, which helps prevent important problems from being overlooked. In conclusion, this low-cost ergonomic tool based on computer vision supports the rapid assessment of postures in schools and companies, providing effective support for ergonomic prevention.