• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.3
• /
• pp.56-63
• /
• 2015

Several fields of science have demanded large-scale workflow support, which requires thousands of CPU cores or more. In order to support such large-scale scientific workflows, high capacity parallel systems such as supercomputers are widely used. In order to increase the utilization of these systems, most schedulers use backfilling policy: Small jobs are moved ahead to fill in holes in the schedule when large jobs do not delay. Since an estimate of the runtime is necessary for backfilling, most parallel systems use user's estimated runtime. However, it is found to be extremely inaccurate because users overestimate their jobs. Therefore, in this paper, we propose a novel system for the runtime prediction based on workload-aware clustering with the goal of improving prediction performance. The proposed method for runtime prediction of parallel applications consists of three main phases. First, a feature selection based on factor analysis is performed to identify important input features. Then, it performs a clustering analysis of history data based on self-organizing map which is followed by hierarchical clustering for finding the clustering boundaries from the weight vectors. Finally, prediction models are constructed using support vector regression with the clustered workload data. Multiple prediction models for each clustered data pattern can reduce the error rate compared with a single model for the whole data pattern. In the experiments, we use workload logs on parallel systems (i.e., iPSC, LANL-CM5, SDSC-Par95, SDSC-Par96, and CTC-SP2) to evaluate the effectiveness of our approach. Comparing with other techniques, experimental results show that the proposed method improves the accuracy up to 69.08%.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.9 no.2
• /
• pp.109-115
• /
• 2014

Analyzing large scale data has become an important activity for many organizations. Since MapReduce is a promising tool for processing the massive data sets, there are increasing studies to evaluate the performance of various algorithms related to MapReduce. In this paper, we first develop a simulation framework that includes MapReduce workload model, data center model, and the model of data access pattern. Then we propose two algorithms that can reduce the energy consumption of MapReduce systems. Using the simulation framework, we evaluate the performance of the proposed algorithms under different application characteristics and configurations of data centers.

• Journal of the Ergonomics Society of Korea
• /
• v.29 no.2
• /
• pp.225-239
• /
• 2010

A high prevalence of farmer's work-related musculo-skeletal disorders (MSDs) have been reported in precedent studies. Risk factors - physical, chemical, biological, ergonomic risk factor etc. - in farm-work were identified by previous studies. To prevent farmers from exposing to them and reduce, Agricultural Safety Technology were introduced and used on farm. The purpose of this study is to review and summarize the agricultural safety technology on farm in Korea. According to strategy principles, this technology is classified as four criteria - workload reduction by reducing awkward posture, workload reduction by using manual material handling aids, Comfort improvement by protecting worker from hot, cold environment and work efficiency improvement by workplace redesign and cultivation modification. This technology includes transporting tools, working cloth, personal protective equipment, and pesticide-proof Clothing. Anthropometric study of Korean farmers was also carried out to support the development of equipment and clothes. In long terms, it needs to precaution activity from risk factors by using cultivation modification and working improvement. Although working condition gets better and workload reduce by this technology, it required customized safety technology in each crops - rice cultivation, horticulture and livestock farming. And farmers should use the assistive technology and participate voluntarily in ergonomic intervention program.

• Journal of the Ergonomics Society of Korea
• /
• v.25 no.2
• /
• pp.43-50
• /
• 2006

The purpose of this study was to investigate applicability of NIOSH lifting equation(NLE) to analysis of workload for patient transferring. In principle, the NLE is not applied to analyzing workload of patient transferring, because 1) the task is generally performed by two or more persons; 2) unlike ordinary objects, human body of patients is basically unstable load with their location of the center of mass significantly varying during lifting activity; and 3) the task is done in a restricted work space. This study was conducted through comparison of NIOSH lifting indexes(LIs) and L5/S1 compressive forces by 3DSSPP for patient transferring tasks performed by 2~6 persons. The results showed that LIs are linearly correlated with L5/S1 compressive forces with correlation coefficient of 0.92, which resulted in a significant simple linear regression equation for LIs and L5/S1 compressive forces. Consequently, it was concluded that the NLE is applicable to transferring patient only with slight modification. Based on the results, instead of 1.0 originally used by NIOSH, the LI of 1.5 was proposed as a gauge to estimate whether or not the task needs corrective action to reduce risk for developing lifting-related low back pain.

• Korea Journal of Hospital Management
• /
• v.19 no.1
• /
• pp.54-67
• /
• 2014

Purpose: The aim of this study was to explore the relationship between nursing work environment and nursing outcome of clinical nurses. Also, the purpose of this study was to identify the nurses' workload. Method: The participants in this study were 246 nurses working at one hospital in a province. Data was collected using a structured questionnaire from October 2011 to November 2011. Open questions were subject to content analysis; closed questions were subject to descriptive statistics and Pearson correlation coefficient with SPSS 18.0 program were used. Result: The number of patients per nurse in day-duty and overtime work had a positive correlation. Nursing performance and the overtime work in night duty nurses had a negative correlation. The results of this study, with a much higher number of patients per nurse, showed an increase in overtime hours. In the nursing delivery system experience of nurses, 13 themes were identified and they were divided into five categories. Conclusion: Therefore, in order to improve nursing performance and increase job satisfaction, efforts must be made to reduce overtime, and nursing personnel placement to secure a substantial nursing staff.

• Journal of the Korean Society of Safety
• /
• v.23 no.2
• /
• pp.51-56
• /
• 2008

In these days, work-related musculoskeletal disorders(WMSDs) is one of the issues in the shipbuilding industry. As the number of injured workers and demands for worker's compensation have rapidly increased, improvement of work conditions and environments to prevent WMSDs has been more demanded. To reduce WMSDs' hazards in the shipbuilding industry, simulation technique which showed it's ability of increasing the manufacturing productivity was applied, because simulation technique has the evaluation ability for a worker's danger level of production process by RULA(Rapid Upper Limb Assesment). In this research, worker's altitude had modeled and worker's action has simulated. After the caution level was evaluated, we pointed out clues which had high workload. To reduce work-load, we applied ergonomic principles for improving working conditions and environments. Improved working conditions and environments were simulated using human modelling and simulation and their workload were evaluated again.

• Korean Journal of Computational Design and Engineering
• /
• v.19 no.3
• /
• pp.272-280
• /
• 2014

Most of the car navigation systems pzrovide 2D or 3D virtual map-based driving guidance. One of the important issues is how to reduce cognitive burden to the driver who should interpret the abstracted information to real world driving information. Recently, an augmented reality (AR)-based navigation is considered as a new way to reduce cognitive workload by superimposing guidance information into the real world scene captured by the camera. In particular, head-up display (HUD) is popular to implement AR navigation. However, HUD is too expensive to be set up in most cars so that the HUD-based AR navigation is currently unrealistic for navigational assistance. Meanwhile, smartphones with advanced computing capability and various sensors are popularized and also provide navigational assistance. This paper presents a research on cognitive effect and responsiveness of an AR navigation by a comparative study with a conventional virtual map-based navigation on the same smartphone. This paper experimented both quantitative and qualitative studies to compare cognitive workload and responsiveness, respectively. The number of eye gazing at the navigation system is used to measure the cognitive effect. In addition, questionnaires are used for qualitative analysis of the responsiveness.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.26 no.3
• /
• pp.40-47
• /
• 2018

Reducing aircraft speed is the important task in the Rejected Takeoff and/or landing process. It is known that the effect of the Speedbrake is most important factor during the rejected takeoff maneuver in particular near V1 on the critical field length runway. The B747 designer created Automatic Speedbrake Control System to relieve pilot workload, improves brake operation and ensures proper Speedbrake operation for rejected take off. However, those who make the Rejected Takeoff procedure ignored the Automatic function and made it does all manual operations. This lets procedures difficult, complicated, and a cause of confusion and pilot error. This study was conducted to commentary the mechanism and function of the Automatic Speedbrake Control System of B747-8 and to propose appropriate B747-8 Rejected Take off procedures for its function to reduce the workload of pilots and contribute to reduce the possibility of pilot error during Rejected Takeoff.

• Safety and Health at Work
• /
• v.13 no.3
• /
• pp.350-356
• /
• 2022

Background: Although sick leave is a complex phenomenon, it is believed that there is potential for prevention at the workplace. However, little is known about this potential and what specific measures should be implemented. The purpose of the study was to identify perceived reasons to take work-related sick leave and to suggest preventive measures. The study was completed before the COVID-19 pandemic emerged, and the risk factors identified may have been amplified during the pandemic. Methods: An in-depth cross-sectional survey was conducted across a randomly selected sample of hospital nurses in Norway. The national sample comprised 1,297 nurses who participated in a survey about their sick leave during the previous 6 months. An open-ended question about perceived reasons for work-related sick leave was included to gather qualitative information. Results: Among hospital nurses, 27% of the last occurring sick leave incidents were perceived to be work-related. The most common reasons were high physical workload, high work pace, sleep problems, catching a viral or bacterial infection from patients or colleagues, and low staffing. Conclusions: Over a quarter of the last occurring sick leave incidents among Norwegian hospital nurses are potentially preventable. To retain and optimize scarce hospital nursing resources, strategies to reduce work-related sick leave may provide human and financial benefits. Preventive measures may include careful monitoring of nurses' workload and pace, optimizing work schedules to reduce the risk of sleep problems, and increasing staffing to prevent stress and work overload.

• Nuclear Engineering and Technology
• /
• v.42 no.1
• /
• pp.115-124
• /
• 2010

According to wide-spread experience in many industries, a procedure is one of the most effective countermeasures to reduce the possibility of human related problems. Unfortunately, a systematic framework to evaluate the complexity of procedural tasks seems to be very scant. For this reason, the TACOM measure, which can quantify the complexity of procedural tasks, has been developed. In this study, the appropriateness of the TACOM measure is investigated by comparing TACOM scores regarding the procedural tasks of high speed train drivers with the associated workload scores measured by the NASA-TLX technique. As a result, it is observed that there is a meaningful correlation between the TACOM scores and the associated NASA-TLX scores. Therefore, it is expected that the TACOM measure can properly quantify the complexity of procedural tasks.