WEARABLE IOT DEVICES WITH AI FOR OCCUPATIONAL HEALTH: REAL-TIME WORKER MONITORING AND SAFETY ANALYTICS

Abstract

Background: Wearable IoT devices with artificial intelligence are set to change occupational health by allowing near real-time worker tracking and safety assessment. These technologies can enhance workplace safety, decrease the rate of accidents, and assess the health of workers who are involved in the construction, manufacturing, or logistics sectors. However, the levels of adherence to these devices as well as the perceived utility of the devices differ between one user and another. Objectives: The aim and objectives of the work include a quantitative analysis of the effectiveness, satisfaction, and impact of wearable IoT devices with AI for real-time health monitoring and risk assessment. Instead, it focuses on examining precursors to device usage and how ergonomic technologies are perceived by workers in the area of safety enhancement. Methods: An online self-administered was done on 250 participants in industries that adopt wearable IoT devices. The key variables of frequency, satisfaction, perceived usefulness of health monitoring, and safety analytics generated from AI were captured using a Likert scale questionnaire. Data analysis was done by use of descriptive statistics, inferential statistics, Shapiro-Wilk normality test, Cronbach’s Alpha reliability coefficient test, and linear regression analysis. Results: Descriptive analysis showed that the age ranged from 17–56 years and age was skewed, W(821) = 0.960, P value = 0.000, Shapiro and Wilk's test showed that age was non-normal shaped which indicates different levels of adoption among age groups. The test-retest reliability of Cronbach’s Alpha was 0.014, for the Likert scale items of the survey, indicating a low internal consistency because of the multi-faceted nature of the worker experience. Those who reported how often they used devices differed in their responses with over half claiming that they used devices occasionally, whereas the perceived efficacy of health monitoring was relatively lower, with almost a quarter of the respondents stating that they found health monitoring to be only somewhat effective or not at all effective. Conclusion: The study shows how occupational health and safety can be enhanced through wearable IoT devices with AI but also clarifies some limitations in the usability of the device, the training of the end-users, and how the survey should be designed. Albeit, adoption has risen over time, daily usage is still low, and they do not have one view of it as being effective. Professoring and resolving these challenges may also help improve the design and bring forth better integration including safety measures in different workplaces and advance the effectiveness of these devices.