• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.487-492
• /
• 2009

A large-scaled research project titled "Intelligent Excavating System (IES)" sponsored by Korean government has launched in 2006. An issue of real-time 3D terrain modeling has become a crucial point for successful implementation of IES due to many application limitations of state-of-the-art techniques developed in various high-technology fields. Many feasible technologies such as laser scanning, structured lighting and so on were widely reviewed by professionals and researchers for one year. Various efforts such as literature reviews, interviews, and indoor experiments make us select a structural light technique and stereo vision technique as appropriate techniques for accomplishment of real-time 3D terrain modeling. It, however, revealed that off-the-shelf products of structural light and stereo-vision technique had many technical problems which should be resolved for practical applications in IES. This study introduces diverse methods modifying off-the-shelf package of the structural light method, one of feasible techniques and eventually allowing this technique to be successfully utilized for achieving fundamental research goals. This study also presents many efforts to resolve practical difficulties of this technique considering basic characteristics of excavating operations and particular environment of construction sites. Findings showed in this study would be beneficial for other researchers to conduct new researches for application of vision techniques to construction fields by provision of detail issues about practical application and diverse practical methods as solutions overcoming these issues.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.77-84
• /
• 2023

The utilization of GPUs on general-purpose computers is currently on the rise due to the increase in its programmability and performance requirements. The utility of tools like NVIDIA's CUDA have been designed to allow programmers to code algorithms by using C-like language for the execution process on the graphics processing units GPU. Unfortunately, many of the performance and correctness bugs will happen on parallel programs. The CUDA tool support for the parallel programs has not yet been actualized. The use of a dynamic analyzer to find performance and correctness bugs in CUDA programs facilitates the execution of sophisticated processes, especially in modern computing requirements. Any race conditions bug it will impact of program correctness and the share memory bank conflicts to improve the overall performance. The technique instruments the programs in a way that promotes accessibility of the memory locations accessed by different threads well as to check for any bugs in the code of a program. The instrumented source code will be used initiated directly in the device emulation code of CUDA to send report for the user about all errors. The current degree of automation helps programmers solve subtle bugs in highly complex programs or programs that cannot be analyzed manually.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.603-622
• /
• 2023

Human error (HE) is an important concern in safety-critical systems such as nuclear power plants (NPPs). HE has played a role in many accidents and outage incidents in NPPs. Despite the increased automation in NPPs, HE remains unavoidable. Hence, the need for HE detection is as important as HE prevention efforts. In NPPs, HE is rather rare. Hence, anomaly detection, a widely used machine learning technique for detecting rare anomalous instances, can be repurposed to detect potential HE. In this study, we develop an unsupervised anomaly detection technique based on generative adversarial networks (GANs) to detect anomalies in manually collected surveillance data in NPPs. More specifically, our GAN is trained to detect mismatches between automatically recorded sensor data and manually collected surveillance data, and hence, identify anomalous instances that can be attributed to HE. We test our GAN on both a real-world dataset and an external dataset obtained from a testbed, and we benchmark our results against state-of-the-art unsupervised anomaly detection algorithms, including one-class support vector machine and isolation forest. Our results show that the proposed GAN provides improved anomaly detection performance. Our study is promising for the future development of artificial intelligence based HE detection systems.

• Journal of Wellbeing Management and Applied Psychology
• /
• v.6 no.2
• /
• pp.9-13
• /
• 2023

Purpose: In this study, we would like to make a technical proposal to solve the odor problem in pig houses. Through this, we would like to suggest an effective way to reduce the odor generated in the pig house as a solution to civil complaints. Research design, data and methodology: Conduct direct visits to pig farms where many civil complaints about bad odor occur, and identify the problems of each farm. Identify elements related to odor control, such as structure, facility, equipment, odor management method, and ventilation type. Through this, the technology to be applied to reduce odor and the solution to the odor problem are presented. Results: The results of major improvements are as follows: 1. Improvement of the structure of the barn or composting shed to an airtight type 2. Improvement of the pig manure treatment structure using the slope inside the barn 3. Establishment of ventilation and cooling systems 4. Automation of the mist spray system. Conclusions: As a result, as practical measures, sealing of facilities using winch curtains, construction of air conditioning systems using negative pressure ventilation, and management systems using AIoT systems were presented. It is judged that this study can be helpful in determining the grievances caused by civil complaints of tenant livestock farms and the direction of facility improvement in the future.

• International journal of advanced smart convergence
• /
• v.11 no.4
• /
• pp.68-80
• /
• 2022

Since the launch of AI, technology development to implement complete and sophisticated AI functions has continued. In efforts to develop technologies for complete automation, Machine Learning techniques and deep learning techniques are mainly used. These techniques deal with supervised learning, unsupervised learning, and reinforcement learning as internal technical elements, and use the Big-data Analysis method again to set the cornerstone for decision-making. In addition, established decision-making is being improved through subsequent repetition and renewal of decision-making standards. In other words, big data analysis, which enables data classification and recognition/recognition, is important enough to be called a key technical element of AI function. Therefore, big data analysis itself is important and requires sophisticated analysis. In this study, among various tools that can analyze big data, we will use a Python program to find out what variables can affect addiction according to smartphone use in a decision tree environment. We the Python program checks whether data classification by decision tree shows the same performance as other tools, and sees if it can give reliability to decision-making about the addictiveness of smartphone use. Through the results of this study, it can be seen that there is no problem in performing big data analysis using any of the various statistical tools such as Python and R when analyzing big data.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.78.2-78.2
• /
• 2019

The temperature control of a scientific device is essential because extreme temperature conditions can cause hazard issues for the operation. We developed a software which can interact with the temperature sensor using the GMT SDK(Giant Magellan Telescope Software Development Kit) version 1.6.0. The temperature sensor interacts with the EtherCAT(Ethernet for Control Automation Technology) slave via the hardware adapter, sending and receiving data by a packet. The PDO(Process Data Object) and SDO(Service Data Object), which are the packet interacts with each EtherCAT slave, are defined on the TwinCAT program that enables the real-time control of the devices. The user can receive data from the device via grs(GMT Runtime System) tools and log service. Besides, we programmed the software to print an alert message on the log when the temperature condition changes to certain conditions.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.404-406
• /
• 2021

In modern society, artificial intelligence and automation are being applied in various fields due to the development of the 4th industrial revolution and IoT technology. In particular, systems with a high proportion of image processing, such as automated processes, intelligent CCTV, medical industry, robots, and drones, are susceptible to external factors noise. In this paper, we propose a digital filter based on noise estimation and weights to reconstruct an image in a complex noise environment. The proposed algorithm classifies the types of noise using noise judgment, and determines the noise level of the filtering mask to switch the filtering process to obtain the final output. In order to verify the performance of the proposed algorithm, simulation was conducted, compared with the existing filter algorithm, and the results were analyzed.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.492-499
• /
• 2022

A large number of employees shifted to Work from home (WFH) due to the COVID-19 pandemic, including the construction workforce. The changes in workload and productivity due to WFH impact the work performance and economic outputs of companies. However, there are mixed results about the impacts of WFH on workload and productivity. In particular, limited studies focused on specific types of work of different occupations in the construction workforce. This study aims to explore the impacts of WFH on workload and productivity considering different types of work for the construction workforce in the U.S. After identifying three main occupations and five types of work, an online survey (N = 69) was distributed. Descriptive analysis showed that participants had less workload (0.82 hours/week) and lower productivity (9.69%) during WFH. Three occupations had varied changes due to the different types of work. Analysis of Variance (ANOVA) indicated that there was no significant difference in workload, while productivity was decreased during WFH. In particular, the productivity of project-related work and communication and documentation decreased significantly. Overall, participants finished 2.85% less workload per week during WFH. The findings provide an insight into WFH in the construction workforce, which improves future remote or hybrid work arrangements in the construction industry.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.500-507
• /
• 2022

COVID-19 pandemic forces college education to be rapidly switched from face-to-face education into remote education. Two inconsistent findings exist in previous study about remote learning. First, studies before COVID-19 pandemic found remote learning is an effective method, which provided students with higher achievement and improved their work-life balance. However, studies showed remote learning during COVID-19 pandemic is not as effective as expected because of technical issues, lack of motivations and even mental health issues. Second, findings from studies about remote learning impacts on workload and productivity during COVID-19 are also inconsistent. Therefore, this study aims to quantitatively measure college students' workload and productivity during COVID-19 of different types of tasks to provide a comprehensive and latest evaluation on remote learning. The findings of this study show remote learning slightly increases college students' total listening and speaking tasks workload, total reading and writing tasks workload. Furthermore, phone call, in-person meeting, online meeting and email workload increase significantly in remote learning. However, productivity for both listening and speaking, reading and writing tasks decreases after remote learning but no significant changes of productivity are found.

• International Journal of High-Rise Buildings
• /
• v.11 no.2
• /
• pp.87-102
• /
• 2022

In Japan, the population progresses to the extreme aging society and it is entering the phase of the population decrease while the population increase is continuing in the world. The construction market is expected to shrink accordingly, however the situation of labor shortage is expected to continue at a faster rate, because the aging of construction workers is progressing and new younger labor force cannot be secured. In order to supplement the labor shortage, it is required to progress mechanization, automation, labor saving, and efficiency improvement by utilizing the information well in each stage in a series of flow of planning, design, construction, operation, and disassembly in one building. The measures to maintain and expand the construction market by the new efficiency improvement techniques which enhance the utilization degree of building information are required. Currently, the elemental technologies which utilized BIM (Building Information Modeling) are accumulated by advancing digitization in each phase. DX (Digital transformation) in the construction industry can be achieved by the technology maturing and having a series of continuities. It is anticipated that this will evolve to a new method which is unprecedented. Present status of BIM and mechanized constructions in Taisei Construction are introduced, and future prospect is described.