Postural assessment in occupational and sports contexts faces challenges due to the subjectivity of the observational methods like the Rapid Upper Limb Assessment (RULA), which helps identify and prevent injuries. Given the potential of inertial sensors for motion capture, integrating them with the RULA method offers an alternative for obtaining quantitative and repeatable assessments. This article proposes a fuzzy logic system based on three inference models using data acquired by a Bluetooth Low Energy (BLE) 5.0 network of six Inertial Measurement Units (IMUs) at a sampling frequency of 50 Hz. To validate the system, 25 participants were assessed on both sides simultaneously while performing three postures involving prolonged sitting in front of a computer. A total of 150 risk injury assessments were conducted. Results obtained by three experts using the traditional method were compared with those obtained using the Fuzzy Inference System (FIS). Agreement between the experts and the FIS was evaluated using Fleiss’ Kappa coefficient, showing good agreement (ranging from 0.57 to 0.63) in four of the six evaluated cases and excellent agreement (ranging from 0.75 to 0.78) in the remaining two cases.