• Journal of the Korea Society of Computer and Information
• /
• v.23 no.2
• /
• pp.63-70
• /
• 2018

Safety management agent manages the risk behavior of the worker with the naked eye, but there is a real difficulty for one the agent to manage all the workers. In this paper, IoT device is attached to a harness safety belt that a worker wears to solve this problem, and behavior data is upload to the cloud in real time. We analyze the upload data through the deep learning and analyze the risk behavior of the worker. When the analysis result is judged to be dangerous behavior, we designed and implemented a system that informs the manager through monitoring application. In order to confirm that the risk behavior analysis through the deep learning is normally performed, the data values of 4 behaviors (walking, running, standing and sitting) were collected from IMU sensor for 60 minutes and learned through Tensorflow, Inception model. In order to verify the accuracy of the proposed system, we conducted inference experiments five times for each of the four behaviors, and confirmed the accuracy of the inference result to be 96.0%.

• Journal of the Korea Safety Management & Science
• /
• v.16 no.4
• /
• pp.257-262
• /
• 2014

Risk information may be one of the most important factor for worker's safe behavior because that safe behavior can be oriented by attitude based on risk information. Traditionally KAP(knowledge, attitude and practice) model was useful frame for the change of human behavior. Knowledge is formed by information through experience and education. Worker's health may be prevented by his or her own active safe behavior based on risk information. This paper is to investigate the effect of labor union on the provirion of risk information by labor union. Data for analysis is the third Korean Working Conditions Survey done by Occupational Safety and Health Research Institute in 2011. The sample size is 50,032 economic active person surveyed by household interview survey with structure questionnaire by trained interviewer. The difference of risk information provision among employees was tested by mean difference test. The level of risk information of employees of companies with labor union is higher than that with non labour union. This paper has some implication for the promotion of safe behavior of employees through risk information provision mediated by labor union. Some limitation of this study may be considered because of using the cross sectional survey data.

• Journal of the Korean Society of Safety
• /
• v.21 no.5 s.77
• /
• pp.60-71
• /
• 2006

In recent years, the most safety studies have been focused on identifying the relationship between safety climate and safe work behavior in accordance with the awareness that prime causes of accidents are organizational, managerial, and human factors rather than pure technical failures across the world. This study focused empirical research aimed at finding out the safety climate factors and examining the relationships of the safety climate and safe work behavior in construction site in Korea, too. A 10 numbers of safety climate factors were adapted from the Mohamed's study(2002) and tested using PLS-GRAPH 3.0 expecting same results as Mohamed' study. And then the relationship between safety climate and safe work behavior was examined. Only two safety climate factors-personal risk appreciation and worker's safety competence-were found and a strong positive relationship between safety climate and safe work behavior was identified in this study. Discussed in details about the results and implications and suggested further studies.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.247-248
• /
• 2019

본 논문에서는 블루투스 비콘을 이용해 고소 작업장 등의 위험지역에서 작업자 추적 및 확인과 안전 벨트고리를 체결했는지 여부와 작업자의 행동에 따른 데이터를 추가로 수집하여 작업자의 행동 패턴을 분석하였다. IoT 국제 표준인 oneM2M을 기반으로 IoT Device와 Application을 연결하는 중간 매개체로 모비우스 플랫폼을 사용해 시스템을 구축하였다. 또한, 본 연구팀의 선행 연구에서 작업자 위험 행동분류 시스템을 개선할 수 있는 연구 결과를 비교하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.63-64
• /
• 2020

본 논문에서는 딥러닝을 이용한 작업자 위험 행동 모니터링 선행 연구에 기반해, 엣지 컴퓨팅 기반 딥러닝을 사용하여 클라우드에 대한 의존성 문제를 해결하였다. 작업자는 IoT 안전벨트와 영상 전송 안전모를 통해 정보를 수집, 처리한다. 또한 LSTM 방식에서 개량된 필터를 통한 FFNN 딥러닝 방법을 사용하여 작업자 위험 행동 패턴 분석을 하며 선행 연구의 작업자 행동 모니터링 시스템을 엣지 컴퓨팅 기반 위에서 구현하였다.

• Journal of the Korean Society of Safety
• /
• v.23 no.2
• /
• pp.57-64
• /
• 2008

The Semiconductor industry in Korea has achieved a great contribution to the growth of its economy for the last 20 years with its product export ranked to #1 nowadays. However, the working environment in semiconductor industry is always exposed to a potential risk of critical safety issue for workers with many kinds of utilities used such as toxic chemicals, gases, high vacuum and high voltages of electricity. This study is focused on empirical research to find out the factors on safety climate and to examine the relationships of safety climate and safe work behaviour in the work site of Korea semiconductor industry. As a result of simulation, a strong positive relationships and safe work behaviour have been identified in this study. Its result and implications has been discussed and suggested further studies on its limitation from this study.

• Journal of the Korea Safety Management & Science
• /
• v.26 no.3
• /
• pp.19-27
• /
• 2024

This study analyzes the impact of occupational health risk assessments on the safety and health levels and the safety behaviors of workers in manufacturing workplaces. An online survey was conducted among 3,172 companies, yielding 637 responses. The statistical analysis on the collected responses revealed three key findings. First, the safety and health levels (safety importance, safety comprehension, safety awareness) positively influence the outcomes of occupational health risk assessments(safety practice, safety management, safety improvement) and safety behaviors (activity change, safety check). Second, the effectiveness of occupational health risk assessments has a positive impact on safety behaviors. Lastly, the effectiveness of occupational health risk assessments partially mediate the relationship between safety and health levels and safety behaviors. These findings are expected to contribute to the promotion of risk assessments in the field of industrial health and to enhancing safety performance by improving workplace safety, health levels, and safety behavior.

• Safety and Health at Work
• /
• v.14 no.3
• /
• pp.279-286
• /
• 2023

Background: This study aimed to evaluate the association between exposure to occupational hazards and the metabolic syndrome. A secondary objective was to analyze the additive and multiplicative effects of exposure to risk factors. Methods: This retrospective cohort was based on 31,615 health examinees at the Pusan National University Yangsan Hospital in Republic of Korea from 2012-2021. Demographic and behavior-related risk factors were treated as confounding factors, whereas three physical factors, 19 organic solvents and aerosols, and 13 metals and dust were considered occupational risk factors. Time-dependent Cox regression analysis was used to calculate hazard ratios. Results: The risk of metabolic syndrome was significantly higher in night shift workers (hazard ratio = 1.45: 95% confidence interval = 1.36-1.54) and workers who were exposed to noise (1.15:1.07-1.24). Exposure to some other risk factors was also significantly associated with a higher risk of metabolic syndrome. They were dimethylformamide, acetonitrile, trichloroethylene, xylene, styrene, toluene, dichloromethane, copper, antimony, lead, copper, iron, welding fume, and manganese. Among the 28 significant pairs, 19 exhibited both positive additive and multiplicative effects. Conclusions: Exposure to single or combined occupational risk factors may increase the risk of developing metabolic syndrome. Working conditions should be monitored and improved to reduce exposure to occupational hazards and prevent the development of the metabolic syndrome.

• Journal of The Korea Institute of Healthcare Architecture
• /
• v.24 no.2
• /
• pp.37-44
• /
• 2018

Purpose: This study investigates the influences of coughing direction and healthcare worker's location on the transport characteristics of coughed particles in airborne infection isolation room (AIIR), which is commonly called negative pressure isolation room, with a downward ventilation system. Methods: Computational Fluid Dynamics (CFD) was used to simulate the airflow and for tracing the behavior of particles. Results: The results show that the airflow pattern and coughing direction have a significant influence on the characteristics of particle dispersion and deposition. When healthcare workers are in the isolation room with the patient who is lying on the bed, it is recommended to be located far from the anteroom to reduce the exposures from infectious particles. And when the patient is lying, it is more effective in removing particles than when the patient is in Fowler's position. Although it is an isolation room that produces unidirectional flow, coughing particles can spread to the whole room and a large number of particles can be deposited onto patient, bed, side rails, healthcare worker, ceiling, floor, and sidewall. Implications: Following the patients' discharge or transfer, terminal cleaning of the vacated room, furniture, and all clinical equipment is essential. Also, it is necessary to establish detailed standard operating procedure (SOP) in order to reduce the risk of cross-contamination.

• Health Policy and Management
• /
• v.31 no.1
• /
• pp.74-90
• /
• 2021

Background: It is getting important to improve the oral health status of the elderly because oral health status may affect their health status of the whole body. In this respect, we aimed to explore the association of oral health status and behavior factors with self-rated health status by sex. Methods: Using the data from the 7th Korea National Health and Nutrition Examination Survey for health surveys and oral examinations (2016-2018), we analyzed a total of 3,070 people aged 65 or older (men: 1,329; women: 1,741). Our dependent variable, self-rated health status, was divided into two groups: not good (bad and very bad) and good (very good, good, and fair), whereas our independent variables of interest were oral health status and behavior factors. In addition to descriptive analysis and the Rao-Scott chi-square test, reflecting survey characteristics, we conducted hierarchical multivariable logistic regression analyses adjusted for socio-demographics and health status and behavior factors. All analyses were stratified by sex. Results: The proportion of people having 'not good' self-rated health was 36.5% in women but 24.5% in men. In a model adjusted for all covariates, the self-rated health status showed significant association with the self-rated oral health status. For example, in men, the risk of having 'not good' self-rated health was high in people having 'poor' (odds ratio [OR], 5.31; 95% confidence interval [CI], 2.34-12.03) self-rated oral health status and in those having 'fair' (OR, 4.03; 95% CI, 1.68-9.70) in comparison with those having 'good' self-rated oral health status. Dental status regarding speaking difficulty seemed to be very important in influencing self-rated health status. For instance, in women, compared to people having 'no discomfort' speaking difficulty, the risk of having 'not good' self-rated health was high in people having 'not bad' (OR, 1.60; 95% CI, 1.14-2.24) and 'discomfort' (OR, 1.79; 95% CI, 1.30-2.47) speaking difficulty. The covariates significantly associated with the risk of having 'not good' self-rated health were: physical activity, chronic disease, stress, and body mass index in both sexes; health insurance type and drinking only in men; and economic activity only in women. Conclusion: Oral health status and behavioral factors were associated with self-rated health status among the elderly, differently by sex. This suggests that public health policies toward better health in the elderly should take their oral health status and oral health behaviors into account in a sex-specific way.