Arduino-Based Development of Automatic Body Posture Measuring Device for Rapid Entire Body Assessment Method

Abstract

Workers from various industries can be exposed to musculoskeletal disorders (MSDs) risks in the workplace. Most cases of MSDs happen because of the accumulation of repetitive bad postures. Postural analysis can be done to implement changes that minimize MSDs risks in the workplace. The rapid entire body assessment (REBA) method provides an easy and fast measurement to analyze various postures. The accuracy of REBA assessments from observations is not good enough due to the limitations of human vision. The input of photo-based postural analysis also has a lot of error potential and takes up even more time. Therefore, this study aims to design a tool to do body posture measurements accurately and quickly. The Arduino-based postural analysis tool measures the position of several body segments, which are then used as the input of REBA in real time. The acceleration of each body segment recorded by the MPU6050 will be processed as displacement data and used to analyze the posture. The posture measurements detected by the tool is then used as input for REBA assessment. The device's accuracy was tested by comparing the photo-based postural analysis result to the postural analysis result of the tool. The reliability of the device was tested by Cronbach's Alpha method. The Cronbach's Alpha value of normal standing posture is 0.692, and the extreme posture's Cronbach's alpha value is 0.537. Multiple postures can be recorded at once using the tool. The tool can capture and assess multiple postures in approximately 6 minutes.