An approach to automated detection of structural failure using chronological image analysis in temporary structures

Jung, Younghan; Oh, Hyounkyun; Jeong, M. Myung

doi:10.1080/15623599.2017.1411457

Cited by 13 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The objectives of this research are to improve scaffold safety monitoring methods and overcome limitations presented in past studies-mainly, the classification of a small number of failure cases for small scaffolds [11][12][13]. To accomplish these, this research designs a method to detect both the global and local member failures of scaffolding structures, as well as complex combinations of member failures.…”

Section: Research Objective and Scopementioning

confidence: 99%

“…To assist in the monitoring of safety problems during construction, a few research studies have sought real-time monitoring approaches [10][11][12][13]. Xue et al [10] proposed the use of radio frequency identification (RFID), together with scaffold safety management, in order to warn safety managers in real-time.…”

Section: Introductionmentioning

confidence: 99%

“…Yuan et al [11] developed a cyber-physical system, which integrates the temporary structure with its virtual model in order to automate potential hazard detection. Jung et al [12] explored the capability of sequential image processing to automate the possible failure detection of temporary structures. Cho et al [13] demonstrated the applicability of machine learning (ML) to automate the classification of scaffold safety conditions, using the strain measurements of the scaffold members.…”

Section: Introductionmentioning

confidence: 99%

“…Cho et al [13] demonstrated the applicability of machine learning (ML) to automate the classification of scaffold safety conditions, using the strain measurements of the scaffold members. However, these efforts to automate scaffold safety assessments are based on a limited number of safety conditions, which are significantly simpler with respect to the mode of failure corresponding to scaffolds used at most construction sites [11][12][13]. As past investigations have identified hazards by checking against threshold values [11], or accounting for only a small number of failure cases [11][12][13] associated with structural configuration, there may be difficulty in establishing their applicability for generic, temporary structures and addressing their potential risks.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multi-Level-Phase Deep Learning Using Divide-and-Conquer for Scaffolding Safety

Sakhakarmi

Park

2020

IJERPH

View full text Add to dashboard Cite

A traditional structural analysis of scaffolding structures requires loading conditions that are only possible during design, but not in operation. Thus, this study proposes a method that can be used during operation to make an automated safety prediction for scaffolds. It implements a divide-and-conquer technique with deep learning. As a test scaffolding, a four-bay, three-story scaffold model was used. Analysis of the model led to 1411 unique safety cases for the model. To apply deep learning, a test simulation generated 1,540,000 datasets for pre-training, and an additional 141,100 datasets for testing purposes. The cases were then sub-divided into 18 categories based on failure modes at both global and local levels, along with a combination of member failures. Accordingly, the divide-and-conquer technique was applied to the 18 categories, each of which were pre-trained by a neural network. For the test datasets, the overall accuracy was 99%. The prediction model showed that 82.78% of the 1411 safety cases showed 100% accuracy for the test datasets, which contributed to the high accuracy. In addition, the higher values of precision, recall, and F1 score for the majority of the safety cases indicate good performance of the model, and a significant improvement compared with past research conducted on simpler cases. Specifically, the method demonstrated improved performance with respect to accuracy and the number of classifications. Thus, the results suggest that the methodology could be reliably applied for the safety assessment of scaffolding systems that are more complex than systems tested in past studies. Furthermore, the implemented methodology can easily be replicated for other classification problems.

show abstract

Section: Research Objective and Scopementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Level-Phase Deep Learning Using Divide-and-Conquer for Scaffolding Safety

Sakhakarmi

Park

2020

IJERPH

View full text Add to dashboard Cite

show abstract

“…Besides proximity issues, researchers have also attempted to resolve other types of hazard on construction sites, such as the use of sensor data to automate the monitoring of temporary structures [18,19] and prevent fall hazards [20]. There are many studies that have used computer vision techniques to identify the unsafe behavior of workers [21] and failure of temporary structures [22,23].…”

Section: Introductionmentioning

confidence: 99%

Investigation of Tactile Sensory System Configuration for Construction Hazard Perception

Sakhakarmi

Park

2019

Sensors

View full text Add to dashboard Cite

The application of tactile-based wearable devices to assist in navigation for people with low sight/low memory has demonstrated the feasibility of using such devices as a means of communication. Accordingly, a previous study in construction research investigated various parameters of tactile signals to develop a communicable system for potential application in construction hazard communication. However, the nature of construction limits the application of such devices to the body of construction workers, and it is important to understand sensor design parameters for improved communication, which has not been given significant attention yet. Therefore, this study aims to determine key design factors such as the number of motors, spacing between sensors and the layout of a tactile sensory system to be used for communicating construction hazards to workers. For this purpose, this study focused on identifying the number of motors based on extensive literature and the problem of construction safety as to hazard communication, determining the arrangement that allowed for effective delivery and perception of information with minimum effort. The researchers conducted two experimental studies: First, to determine the minimum spacing between vibration motors that allows for the identification of each individual motor with high accuracy; and second, to determine the layout of motors that is suitable for effective communication of multiple types of information. More importantly, the tactile-sensor configuration identified from this study allows the workers to learn the signal patterns easily in order to identify multiple types of information related to hazards. Using such a communication system on construction sites will assist in transmitting hazard-related information to workers, and thus, protect the lives of workers. Such wearable technologies enable the detection of individual-level hazards and prevent worker fatalities and severe injuries.

show abstract

From Smart Construction Objects to Cognitive Facility Management

Anumba

et al. 2020

Cyber-Physical Systems in the Built Environment

View full text Add to dashboard Cite

An approach to automated detection of structural failure using chronological image analysis in temporary structures

Cited by 13 publications

References 8 publications

Multi-Level-Phase Deep Learning Using Divide-and-Conquer for Scaffolding Safety

Multi-Level-Phase Deep Learning Using Divide-and-Conquer for Scaffolding Safety

Investigation of Tactile Sensory System Configuration for Construction Hazard Perception

From Smart Construction Objects to Cognitive Facility Management

Contact Info

Product

Resources

About