Journal Search

Korean Journal of Construction Engineering and Management

Korean Journal of Construction Engineering and Management

ISO Journal Title : Korean J. Constr. Eng. Manag.
Open Access Journal Bimonthly
  • ISSN (Print) : 2005-6095
  • ISSN (Online) : 2465-9703
Online Submission

Current Issue : Vol.26 No.4, 2025-07

more

Development of Generative AI-based System for Extracting and Analyzing Construction Accident Cases

Su Hyeon Choe ; Han Soo Kim

https://dx.doi.org/10.6106/KJCEM.2025.26.4.003

The construction industry faces a higher severity and risk of accidents compared to other industries. With growing societal concern over construction accidents and the strengthening of legal responsibilities for construction companies, systematic and proactive safety management has become even more crucial. Analyzing past accident cases is essential for preventing the recurrence of similar incidents and establishing an effective safety management system. However, identifying relevant cases from the vast accumulation of past accident records that suit specific project conditions requires substantial time and effort. One potential solution is applying generative AI technology with natural language processing capabilities to extract and analyze construction accident cases. The objective of this study is to develop and propose a system that uses ChatGPT, a generative AI, to extract and analyze construction accident cases provided by the Korea Occupational Safety and Health Agency. This system is expected to contribute to proactive accident response and improve safety management systems in construction companies.

Full Text PDF Abstract

Development of a Support Tool for the Task Schedule Plan of PHC Pile Head Cutting Robot

Woo Jin Lee ; Se Jin Park ; Ju Ha Suk ; Jun-Sang Kim ; Young Suk Kim

https://dx.doi.org/10.6106/KJCEM.2025.26.4.015

The conventional PHC pile head cutting method has several problems with regard to labor safety, productivity, and quality. The PHC pile head cutting robot was created in earlier research to address these problems. The purpose of this study is to develop an algorithm and software that optimize the PHC pile head cutting robot’s deployment timing and task scheduling by taking into account the automated work process as well as the preliminary and post PHC pile head cutting procedures. By simply inputting the pile penetration schedule and the quantity of piles for each facility on the construction site, the development algorithm and software can calculate the head cutting work schedule and deployment timing. Furthermore, through a task allocation procedure, the algorithm and software are made to produce even when more than two facilities need head cutting on the same day. In comparison to conventional task schedule plans, the case study result shows that the developed task schedule plan algorithm is anticipated to plan for the deployment of a single robot on the construction site, minimizing idle time, mobilization & demobilization frequency, and associated costs. In the future, it is expected that the software would eventually allow construction site managers to predict the robot’s deployment schedule in advance, facilitating accurate cost estimation for the head cutting task using robots.

Full Text PDF Abstract

A Study on Displacement Visualization Technology for Retaining Walls Using LiDAR

Jun-Sang Kim ; Young Suk Kim

https://dx.doi.org/10.6106/KJCEM.2025.26.4.026

Retaining wall displacement is a critical measurement factor for intuitively assessing the stability of retaining walls, making its measurement essential. However, traditional inclinometer-based measurement methods face several challenges: 1) difficulties in installation and dismantling, 2) labor-intensive processes, and 3) limitations in displacement measurement range. To address these issues, a 2D-LiDAR-based retaining wall displacement monitoring system has recently been developed. This system provides a cost-effective solution for monitoring retaining wall displacements compared to conventional terrestrial laser scanners. The objective of this study is to develop displacement visualization technology and software for application in the 2D-LiDAR-based retaining wall displacement monitoring system and to evaluate their performance. The developed LiDAR-based displacement visualization technology consists of two key features: 1) a region of interest (ROI) defining function and 2) a multidimensional displacement visualization function. Performance evaluation of the software showed that the average time for updating the ROI boundary was 0.1 seconds, while the average completion times for the front view visualization function and cross-sectional visualization function were 1.47 seconds and 0.24 seconds, respectively. These results demonstrate that the developed system can efficiently process irregular point cloud data and visualize displacement data in real time for field managers. In the future, if the visualization technology is applied in field settings, it is expected to enable field managers to instantly assess retaining wall displacements and make prompt and effective decisions regarding retaining wall stability.

Full Text PDF Abstract

A Study on the Effect of Digital Transformation on Project Performance in the Construction & Plant Project Industry

Chang Yeong Choi ; Sinbong Kang

https://dx.doi.org/10.6106/KJCEM.2025.26.4.036

In this study, a new research model and hypothesis were proposed to systematically introduce digital transformation (DX, DT, hereinafter referred to as "DX") to domestic plant construction industry, which has been struggling with management due to recent weakening price competitiveness, lack of project management capabilities, and lack of technology and innovation. In the research model setup, digital competency was set as the independent variable, digital maturity as the mediator variable, and CEO's willingness to utilize DX and client's demand intensity to utilize DX as the moderator variables, allowing for the verification of project performance as the dependent variable. The survey is meaningful as an empirical study because it targets samples with experience in the plant industry and DX. As a result of this study, it was verified that the introduction of DX into plant construction industry has a significant impact on project performance and is a strategic tool that can bring about innovative changes in business.

Full Text PDF Abstract

Estimation of Utility Index in Design Value Engineering Considering Owner’s Preference

Sungwoo Moon ; Changsik Kim

https://dx.doi.org/10.6106/KJCEM.2025.26.4.049

Design Value Engineering (VE) enhances desing values by exploring alternatives to the existing design. During the design VE process, VE experts evaluate the superiority of design alternatives by estimating the value index of both the existing design and the alternatives. However, when estimating the value index, an excessively large cost reduction can lead to a rapid increase in the value index. This phenomenon occurs because the owner's preference for the cost reduction is not sufficiently reflected. Therefore, to avoid this result, the owner’s preferences should be analyzed and taken into consideration. The objective of this paper is to calculate a realistic design VE value index by reflecting the owner's preference to the cost reduction. To achieve this research objective, this paper 1) developed an owner's utility curve, 2) used the utility curve to correct the cost reduction and 3) calculated value index to consider the owner's preference. The research results demonstrated that applying the owner’s preference can better represent the improvement of desing value and effectively reflect the complex and diverse characteristics of construction operation in design VE.

Full Text PDF Abstract

Development of an Smart Airports for Enhancing BIM Base on Simulation Verification System

KiHyun Park ; ChangKu Woo ; KISIg Han ; Yugil Kim ; JaeSeok Kang ; JungKi Kim

https://dx.doi.org/10.6106/KJCEM.2025.26.4.058

Airport simulation is a program that performs predictive analysis of airport operational environments by constructing a digital twin of the airport within a virtual space. It supports the maximization of operational efficiency and the establishment of optimal facility investment plans and design strategies. One of the primary applications of simulation in the airport design process is to optimize the passenger terminal area and determine the appropriate number of service counters, enabling visualization of operations accordingly. Airport passenger simulations are conducted by modeling passenger terminals using collected and applied national airport data. These models help stakeholders better understand terminal operations and system behaviors. To apply the model to real-world systems, a simulation is executed using the Ministry of Land, Infrastructure and Transport’s passenger service level standard algorithm (developed in-house by the Korea Airports Corporation) to analyze performance results. The research model selected for this study was the expansion project of the international passenger terminal at Gimhae Airport. The simulation yielded significant results, notably reducing average maximum passenger waiting time in the arrival hall by approximately 27%, demonstrating an optimal simulation outcome. These results were directly incorporated into the design-stage BIM model, enhancing design appropriateness and accuracy. By integrating the KAC-PERT system, the simulation operated at half the resource consumption compared to conventional commercial software. Additionally, the Work Breakdown Structure (WBS) method was integrated, improving both usability and system efficiency. Design verification was conducted using the simulation analysis outcomes, and technological linkages were established with the Da:Bom PMIS platform. BIM methodology was also used to assess the suitability of construction methods, which contributed to a 20% reduction in the constructoin period and a 29% decrease in construction costs.

Full Text PDF Abstract

A Study on the Effect of Construction PM Competency on the Construction Project Quality Performance - Focus on General Construction Project -

Hyunseong Shin ; Sinbong Kang

https://dx.doi.org/10.6106/KJCEM.2025.26.4.068

This study is based on the theoretical frameworks of PMBOK and ISO 21500 and aims to analyze the impact of construction PM competencies on project quality performance in construction projects. The core competencies of construction PMs include project goal setting, stakeholder management, WBS development, risk management, and execution process control. Additionally, the study examines whether the construction management and communication competencies of subcontractor site managers have a moderating effect on the relationship between PM competencies and project quality performance. The results indicate that the construction management competency of subcontractor site managers significantly influences project execution. This finding highlights the need for a structured training system and the implementation of mandatory education programs. Furthermore, enhancing construction PM competencies through education and awareness is expected to improve productivity and reduce the occurrence of defective construction. Finally, improving the competencies of subcontractor site managers may strengthen the negotiation power of specialty contractors and contribute to the overall quality enhancement of domestic construction sites.

Full Text PDF Abstract

An Analysis of Construction Workers‘ Ages, Accident Types, and Injured Body Parts Using Correspondence Analysis in Korea

Jiseon Lim ; Jaehong Cho ; Sungjin Choi ; Sanghyeok Kang

https://dx.doi.org/10.6106/KJCEM.2025.26.4.081

The aging population of skilled construction workers is increasing. As the proportion of elderly workers with reduced physical and cognitive functions increases compared to young workers, accidents occurring at construction sites will inevitably increase. Additionally, it has been revealed that there are differences in the injured body parts by age group. Therefore, understanding the worker’s age, type of accident, and injury body parts is considered important for effective safety management. In this study, a correspondence analysis method was used to investigate the relationship between the ages of the injured, the injured body parts, and accident types using accident case data that occurred in the construction industry. As a result, it was confirmed that there were differences depending on the category of the variables. By age group, those over 60 years old showed a high correlation with torso injuries, those in their 50s with face and head injuries, and those in their 10s and 20s with foot and ankle injuries. By type of disaster occurrent, falls and falling accidents were highly associated with workers in their 50s and 60s and older, while being struck by objects and cuts were highly associated with workers in their 30s. The results of this study will be helpful in establishing more detailed management strategies and measures when performing safety management in the field.

Full Text PDF Abstract

Predicting Construction Schedule Delays Using Large Language Models

Saruul Ishdorj ; Jongho Lee ; Young hun Jun ; Yongseok Choi ; Kyu Heong Kim ; Junghon Jeon ; Jaeyoon Kim ; Jinwoo Kim

https://dx.doi.org/10.6106/KJCEM.2025.26.4.089

This study investigates the performance of Large Language Models (LLM), specifically generative pretrained transformer, in predicting construction schedule delays through in-context learning approaches. Predictions were conducted both on unstructured data and on data transformed into structured formats, allowing for a comparative analysis of prediction accuracy across different levels of data preprocessing. The results showed that the prediction accuracy of the zero-shot approach was limited to 38.3%. In contrast, the few-shot approach achieved a prediction accuracy of up to 48.8% when trained on unstructured data and up to 73.9% when trained on structured data. These findings demonstrate the critical role of data structuring and prompt design in enhancing model performance. This study contributes to advancing the use of LLM in construction delay analysis and offers practical implications for improving schedule management and decision-making processes in the construction industry.

Full Text PDF Abstract