Introduction

In Korea, the Serious Accident Punishment Act, implemented in January 2022, holds business managers legally accountable for industrial accidents resulting in fatalities. Despite this strengthened legal framework, the construction industry continues to face high accident rates. In 2022, construction-related accidents accounted for 402 fatalities, representing 46.0% of all industrial accident deaths [1]. Alarmingly, the annual fatality rate per 10,000 construction workers increased for both general and specialized construction companies after the Act’s implementation [2], and no significant improvement was observed (Table 1). Particularly, recent major accidents at apartment construction sites have drawn significant public attention due to the risks they pose not only to workers but also to surrounding communities.

Traditionally, safety management at construction sites has been primarily the responsibility of contractors. However, international best practices, such as the American Society of Civil Engineers’ Policy Statement 350, emphasized that owners should play a proactive role in safety management [3]. Although owners are central to safety management, they often lack the technical expertise required to monitor and control the safety activities conducted by contractors and other involved parties. Designers and contractors tend to focus narrowly on safety aspects relevant to their specific responsibilities, leading to inconsistencies in safety practices across the project. This highlights the need for owners, who oversee the entire construction process, to ensure comprehensive safety management. While several studies highlight the importance of the owner’s role in safety management [4, 5, 6], most focus on the necessity of owner involvement without addressing the development of a structured safety management system tailored specifically for owners. The Construction Safety Management Manual issued by Korea’s Ministry of Land, Infrastructure and Transport outlines the owner’s responsibilities in implementing safety measures at different stages of construction work [7]. However, the owner’s role is often limited to reviewing construction plans and confirming supervision documents. Given that most owners lack the technical knowledge required for construction and safety management, this level of involvement proves insufficient for effective safety oversight. Although owners are primarily responsible for granting approvals, they often do so without fully understanding the safety implications due to their limited expertise. Nonetheless, they remain legally accountable for the project’s safety outcomes. Therefore, there is an urgent need to develop tools that support owners in managing safety risks effectively.

One of the critical challenges is the fragmentation of safety-related standards and data. Existing safety guidelines are often extensive, varied, and categorized by construction type, but fail to align with the actual sequence of construction activities. This misalignment creates practical difficulties for owners, particularly those less experienced in safety management. To address this, a system is needed to organize and present safety data in a way that reflects the chronological workflow of construction, ensuring it is intuitive and actionable.

This study proposes a time-series work type (TSWT) classification system, which systematically categorizes construction activities in chronological order. The TSWT classification system bridges the gap between safety guidelines and on-site workflows, allowing owners to access and apply relevant safety information at each stage of construction. Building on the TSWT framework, the study also presents a novel dashboard specifically for owners. Unlike traditional tools focused on contractors, this dashboard integrates fragmented safety data into a cohesive platform tailored to the unique responsibilities of owners. Key features include intuitive data visualization, integration with existing safety protocols, and predictive analytics for proactive risk management. These innovations enable owners to make informed safety decisions, enhancing both practicality and efficiency.

By combining the TSWT classification system with an intuitive, data-driven dashboard, this study introduces a pioneering approach to safety management in apartment construction. The proposed platform is expected to transform the utilization of safety data, fostering a proactive and resilient safety culture within the industry. Ultimately, this integrated approach aims to reduce accident rates, enhance workplace safety, and establish a new standard for owner-led safety management practices.

Literature review

Safety management in the construction industry

Safety management aims to prevent accidents—recognized as inefficiencies—and enhance productivity by minimizing losses. It is a planned and systematic activity designed to protect human life and property from hazards [8]. According to the U.S. Occupational Safety and Health Administration (OSHA), construction safety management is a systematic and preventive approach aimed at identifying hazards and managing associated risks [9]. It encompasses strategies and activities designed to create an organizational environment in which all personnel—from managers to workers—are trained and motivated to perform construction tasks safely and efficiently [10].

The most effective way to prevent accidents is to eliminate or reduce potential hazards in advance [11]. Construction sites are inherently complex and hazardous environments, filled with dangerous equipment, sensitive materials, and diverse teams of participants working toward various objectives. If not managed properly, these factors can compromise the health and safety of all involved.

Heinrich’s law, a statistical principle formulated by Herbert William Heinrich during his tenure in the engineering and loss control department of an American insurance company, is based on an analysis of numerous industrial accident cases encountered through his professional activities [12, 13]. This law establishes that the ratio of occurrences among ‘accidents causing major injuries,’ ‘accidents causing minor injuries,’ and ‘no-injury incidents’ (such as near misses) at worksites is approximately 1:29:300 (Figure 1). The law empirically demonstrates that major accidents do not happen randomly or suddenly; instead, they typically emerge through a process where minor accidents are repeatedly observed beforehand. Furthermore, it highlights that several warning signs and precursors usually precede a major accident. In other words, significant accidents tend to occur when minor risk factors are overlooked. Consequently, reducing the risk by managing or eliminating these factors at construction sites is likely associated with a reduction in the frequency of major incidents [14].

Figure 1.

Heinrich’s pyramid.

Fundamentally, construction safety management is a process that involves the flow of safety-related information, including its collection, transmission, storage, analysis, assessment, visualization, and response [15]. Various innovative technologies can be utilized to support the management of this information flow. While efforts by researchers and practitioners have led to a steady decline in construction accidents [4, 16], construction projects remain inherently complex, difficult, unstructured, and subject to constant change, making the adoption of innovative technologies challenging [15, 17, 18]. Furthermore, the heavy workloads of safety managers and workers’ resistance to safety monitoring present additional barriers to the implementation of advanced technologies. As such, there is a pressing need to develop and apply technologies that align with the practical requirements of owners and field personnel, who are the primary stakeholders in construction site safety management.

User needs analysis for platform dashboard design

A dashboard is a visual interface designed to present information in a clear and easily understandable manner, facilitating users’ ability to perform their tasks effectively. It serves as a visual representation tool that allows users to access relevant information at a glance [19]. For a dashboard to be effective, it must demonstrate both high usability and usefulness. While these two terms are sometimes used interchangeably, they represent distinct concepts [20]. Usability refers to how easy and convenient the tool is to use, but a tool with high usability may not necessarily contribute to achieving a specific goal or solving a particular problem. In contrast, a tool may offer high usefulness, meaning it is essential for goal achievement, but it may lack usability, making it difficult to operate efficiently. Thus, it is crucial to develop a dashboard that excels in both usability and usefulness. A dashboard with these characteristics increases efficiency, user satisfaction, and task effectiveness, as users are able to achieve their objectives more comfortably and enjoyably. However, dashboards designed with a sole focus on functionality can make it challenging for users to intuitively understand the information presented. Therefore, the design must adopt a user-centered perspective [21]. To achieve a user-centered dashboard design, an accurate understanding of the users’ needs is essential. By reflecting user requirements in the design phase, a platform can be created that aligns more closely with the purposes and tasks users intend to accomplish [22]. Accordingly, this study aims to analyze the needs of platform users—specifically safety managers and site supervisors—to design a safety management platform dashboard for apartment construction sites.

Research process and methodology

The research framework is divided into three main parts: (1) the establishment of a TSWT classification system for apartment construction sites; (2) the derivation of necessary platform functions based on an analysis of safety management platform user needs; and (3) the proposal of a prototype for an apartment construction safety management database linkage platform. The details of each step are as follows. First, a TSWT classification system is proposed as the foundation for developing a platform that provides safety standards and related data to site supervisors according to the phases of apartment construction projects. Current safety standards and data are mainly organized by machinery, equipment, or work processes, which hinders the timely retrieval of information and effective on-site safety management. To address this issue, a classification system serving as the basis for developing a platform to offer systematic safety standards and data must first be established. The Delphi method was employed to verify the validity of the proposed TSWT classification system, derived from an analysis of apartment construction project schedules. Second, the needs of site supervisors at apartment construction sites, as platform users, were analyzed to identify the functions required for the platform based on the results. Finally, a prototype dashboard for the safety management platform was developed, designed to provide project owners with the necessary safety management information at each stage of the construction process. The platform is intended to be both accessible and practical for real-world application.

Delphi method for the establishment of a time-series work type classification system

The Delphi method was used for the establishment of a TSWT classification system for apartment construction sites. It is a research technique that derives collective consensus by employing an iterative process, controlled feedback, anonymity, and statistical techniques to elicit and synthesize expert opinions, particularly when relevant data is lacking. It is used to objectify and predict expert opinions [23, 24, 25]. The Delphi survey is commonly utilized in research where key decisions must be informed by expert input, and thus requires careful consideration of the expertise, qualifications, sincerity, representativeness, and appropriateness of the participants [26].

Previous Delphi survey studies have commonly employed career experience as a key criterion for selecting experts, aiming to enhance the reliability of the expert panel and ensure the inclusion of qualified professionals in the field [27, 28]. Typically, experts with more than 10 years of experience are selected; however, adopting a 15-year threshold serves as a more rigorous criterion, further reinforcing the reliability of the survey. In this study, the selection criteria for experts contributing to the development of a safety criteria classification system for apartment construction projects were as follows: First, experts must have at least 15 years of experience in apartment construction sites. Second, for those with less than 15 years of experience, they must possess at least one of the following certifications: Engineer Construction Safety, Industrial Engineer Construction Safety, Engineer Industrial Safety, or Industrial Engineer Industrial Safety. Prior to forming the expert panel, we presented these selection criteria to potential panel members, explaining their alignment with the research objectives. The experts unanimously agreed that the criteria were appropriate and reasonable, ensuring their validity for the study.

The experts who participated in the Delphi survey to develop the TSWT classification system for various types of work in apartment projects are outlined in Table 2. The expert panel had an average of 14.17 years of experience (SD = 11.05) in safety management and construction management at apartment construction sites. A total of 30 experts from 7 organizations participated in the expert panel, comprising 3 large construction companies, 3 mid-sized construction companies, and 1 public ordering agency.

The Delphi survey conducted in this study was conducted over three rounds, and the TSWT classification system for apartment projects was established through a process of collecting, analyzing, and adjusting expert opinions in each round (Figure 2). A Delphi survey was conducted among experts to evaluate the three draft of TSWT classification system derived from the analysis of planned construction schedules for apartment projects. In the first round of the Delphi survey, open-ended questions were used to identify relevant elements, and the validity of the data collected, as well as expert opinions, were reviewed. In the second round, the validity of the TSWT classification system, revised based on the first round’s feedback, was assessed. Items were ranked and classified according to their importance, and participants were asked to suggest additional or modified items. The third Delphi survey used a 5-point Likert scale to assess the validity of the refined work classification system. To verify the system’s validity, the mean (M), standard deviation (SD), content validity ratio (CVR), and coefficient of variation (CV) were calculated for each item.

Figure 2.

Flow chart of the three rounds of Delphi interview.

CVR is mainly used as an indicator to measure the appropriateness of items [29] and is calculated using following formula (1). Here, ne is the number of panelists who responded ‘essential’ to the item, and N is the total number of panelists.

Lawshe (1975) proposed minimum CVR values based on the number of expert panelists in Delphi surveys. For a panel size of 30, the minimum acceptable CVR value is 0.33 (Table 3). Items with CVR values below this threshold were either removed or revised in accordance with expert feedback. When two or more panelists highlighted significant modifications to certain items, the content was revised and supplemented through additional literature review, as deemed appropriate by the researcher.

The level of agreement among respondents is assessed using the CV, which is calculated by dividing the standard deviation by the mean. A CV of 0.5 or less indicates strong agreement among experts, suggesting that sufficient stability has been achieved, and the Delphi survey can be concluded [30]. In the third round of the Delphi analysis, items with a CVR of 0.33 or less, or a CV above 0.5, are removed. The final TSWT classification system for apartment construction projects was then established after incorporating expert advice and making necessary revisions and adjustments.

Analysis of construction safety management platform user needs

To support safety management operations at construction sites, interviews were conducted with industry practitioners to gain detailed insights into the current status and challenges of safety management. Seven newly constructed apartment sites were visited, and feedback was collected from 18 experts (Table 4). Additionally, platform design requirements were obtained from seven site supervisors representing public ordering agencies (Table 5).

Based on the findings from the survey on safety management practices and the expert interviews, common user needs for a platform linked to the safety management database for apartment construction sites were identified and organized as follows:

1.Construction contractors (safety managers and construction managers from private companies)

ㆍProvide information on relevant laws, construction methods, risk factors, and safety measures according to the work sequence within each construction work type.

ㆍOffer guidance on safety management priorities that adjust according to milestone updates during construction.

ㆍEnsure compliance with safety standards prior to the commencement of each construction work type.

ㆍProvide safety training materials based on real- world disaster case studies.

ㆍEnable integrated management of work types with overlapping risk factors.

2.Owner (public ordering agency construction site supervisor)

ㆍPresent and notify submissions and review checklists for safety compliance at each stage of construction.

ㆍCurate safety management materials and organize content in a manner accessible to non-expert users, such as new site managers.

ㆍSupport the review of critical safety documents, including the harmfulness and hazard prevention plan and the safety management plan, ensuring alignment with site-specific conditions.

ㆍProvide periodic updates on work-related information to meet owner obligations, in accordance with legal and regulatory guidelines.

ㆍInclude seasonal safety inspection content to address accident vulnerabilities during high-risk periods (e.g., rainy season, typhoon season, winter, and thawing seasons).

Results and Discussion

Time-series work type classification system

To classify work types in apartment construction projects according to the construction sequence, the planned work schedules from 10 apartment construction sites were analyzed. Based on this analysis, three draft classification systems were developed, categorizing the work types in chronological order. The three draft versions of the classification system differ slightly in the order of work and in the division into major and minor categories, as shown in Figure 3.

Figure 3.

Three drafts of the TSWT classification system.

In the first Delphi survey, experts identified the third draft as the optimal TSWT classification system. Feedback for revision and supplementation was collected through an open-ended questionnaire, and the consolidated opinions are summarized as follows:

ㆍSince most safety management practices for finishing work are general and uniform, detailed classification of finishing work is deemed unnecessary for effective safety management. Work types with similar or tolerable risk factors related to finishing work should be managed in an integrated manner.

ㆍUnderstanding construction methods is essential for preventing safety accidents. Therefore, it is necessary to organize this information based on standard specifications, which serve as national construction standards, to enhance ease of information accessibility.

The revised TSWT classification system for apartment construction projects, incorporating feedback from the first Delphi survey panel, is presented in Table 6.

To verify the validity of the TSWT classification system for apartment construction established through the results of the first Delphi survey, a second Delphi survey was conducted. Feedback for revision and supplementation was gathered through open-ended questions and in-depth interviews, and the common opinions were summarized as follows:

ㆍFrom a safety management perspective, the overall work type classification for construction is appropriately organized into five categories: 1) earthwork, 2) foundation and groundwork, 3) temporary work, 4) reinforced concrete work, and 5) finishing work. Consequently, some items will be revised based on this middle classification.

ㆍWork types that account for a very small proportion of the overall process, such as window ventilation and miscellaneous works, will be excluded.

A revised TSWT classification system for apartment construction was developed by incorporating the feedback from experts in the second Delphi survey panel. To verify the validity of the TSWT classification system for apartment projects established through the results of the second Delphi survey, a third Delphi survey was conducted. Questions were designed using a 5-point Likert scale to assess the appropriateness of each construction work type. The analysis revealed no items deemed inappropriate, as all items met the criteria of having a CVR above 0.33 and a CV below 0.5 (Table 7). Consequently, all derived items were ultimately selected.

The TSWT classification system addresses the limitations of traditional construction schedules, which primarily focus on quantity estimation and payment processes. By organizing work types for apartment construction according to the construction sequence, this classification system provides a structured framework for safety standards. It serves as a foundation for developing a platform dashboard tailored for apartment construction site owners, integrating a safety management database based on time-series work types.

The analysis of construction safety management platform user needs

An investigation into the status of safety management at apartmZent construction sites revealed several issues, including a lack of expertise due to the frequent rotation of supervisors, and field staff being unaware of changes in relevant laws and regulations. These shortcomings in expertise and technical skills among safety managers can lead to operational errors and undermine the reliability of safety management information systems, especially given the unique characteristics of construction work. Furthermore, there is a need for a system that clearly outlines the responsibilities of each stage and participant throughout the entire construction process, enabling easier tracking of administrative and safety management tasks.

Construction safety managers are required to submit various documents to assess risks associated with different types of work and to manage overall construction safety [31]. These include the harmfulness and hazard prevention plan (required by the occupational safety and health act), safety management plan (under the construction technology promotion act), construction plans, regular or occasional risk assessments, and tool-box meeting (TBM) reports. The number of safety documents required under the Industrial Safety and Health Act alone totals approximately 180 items, and with the enactment of the Serious Accident Punishment Act, the volume of documents requested by owners and general contractors has increased. This has significantly escalated the administrative burden on safety managers, negatively impacting the effectiveness of safety management and hindering the efficient operation of safety systems.

Based on the findings from a survey on safety management practices at apartment construction sites, as well as interviews with contractors (including safety managers and construction managers from private construction companies) and owners (construction site supervisors from public ordering agencies), key user needs for a safety management platform were identified and categorized. These findings informed the definition of essential elements to be incorporated into the platform design as follows:

1.Owner’s Safety Obligations Checklist (by Phase)

ㆍInclude mandatory safety obligations of owners for each construction stage, applicable regardless of work type.

ㆍProvide a comprehensive guide to documents required for submission, review, and approval by the owner, in compliance with construction safety- related laws and guidelines.

ㆍAttach relevant laws and reference materials as supporting data for each checklist item.

ㆍDetail the necessary implementation and review items for various time periods during the construction phase, including daily, weekly, monthly, quarterly, and annual tasks, as well as adjustments for seasonal changes such as rainy season or winter season and specific events such as design modifications or construction suspensions.

2.TSWT Safety Standards and Data Linkage

ㆍEnsure linkage of safety standards and related data based on the TSWT classification system for apartment construction.

ㆍProvide direct access to data on detailed standards, including construction methods and site- specific characteristics.

ㆍSupply information on major risk factors, supplemented by disaster cases.

ㆍPresent a summary and key details of each safety standard and associated safety management data.

Prototype for the platform dashboard

This study applied platform design reflection factors derived from user needs analysis to create a platform dashboard focused on usability and effectiveness. The platform dashboard, along with its underlying database of materials, was evaluated for appropriateness and validity through expert interviews.

The owner’s safety obligations checklist (by phase) offers a structured tool for managing essential safety tasks at each stage of construction: commencement, construction, and completion. It categorizes safety management measures by phase, supplemented by relevant legal frameworks to facilitate efficient safety oversight. This checklist not only centralizes safety documentation but also details responsible parties (Who), timelines for preparation and submission (When), procedures for drafting and approval (How), and designated submission locations (Where).

For instance, during the commencement phase, tasks such as submitting the harmfulness and hazard prevention plan and obtaining approval for the safety management plan are required before physical construction begins. Additionally, the checklist incorporates periodic safety measures for the construction phase, subdividing tasks into daily, weekly, monthly, quarterly, semi-annual, annual, seasonal, and other intervals. Seasonal conditions include extreme weather events, such as thawing season, typhoon season, and winter. Other intervals encompass factors such as extensions of construction period, design changes, and accident occurrences. In the construction phase, the content that needs to be performed on that date is automatically loaded based on the data previously entered data by the user. For example, if the platform database reflects the following entries: ‘daily’ checked one day ago, ‘weekly’ one week ago, ‘monthly’ two weeks ago, and ‘quarterly’ three months ago, then the checklist will prompt the user to verify daily, weekly, and quarterly tasks on the current date. Seasonal and other can be provided or manually selected as required. The completion stage further provides guidance on safety tasks needed to finalize the project.

The dashboard was specifically designed to improve user accessibility and streamline data review. It achieves this by highlighting information directly relevant to safety requirements, sourced from original legal data and reference materials, in alignment with expert feedback. Figure 4 illustrates an example of how this checklist is integrated into the platform’s dashboard.

Figure 4.

Example of a dashboard design in construction phase for the owner’s safety obligations checklist (by phase).

The TSWT safety standards and data linkage is a classification system developed to organize safety standards and related data according to construction work types. Safety-related materials are categorized into distinct groups: laws under ‘Laws/Regulations’, national standards under ‘Construction Standards’, guidelines for safety management under ‘Safety Guidelines’, owner-specific management tasks under ‘Guidelines/Manuals’, and risk assessments under ‘Checklists’. Accident cases and best practices are compiled under ‘Field Casebook’. These data have been systematized into a database, ensuring streamlined access for users based on work type classification. Scheduled works for a specific date are automatically loaded based on information entered by project managers. As shown in Figure 5, construction supervisors can select the ongoing work type and efficiently access necessary safety management data for that specific work type.

Figure 5.

Illustrative dashboard design for monitoring ongoing construction work.

Feedback from industry experts was gathered to assess the platform’s practical usability. The general response was favorable, indicating the platform’s potential for real-world application in apartment construction. However, several areas for improvement were identified. Key recommendations included: (1) providing summaries of linked data to help users quickly grasp key information; (2) distinguishing safety standards based on specific construction methods within each work type; and (3) updating key risk factors based on real-time accident data. Based on this feedback, the user experience (UX) design of the platform was refined to better align with on-site needs.

Figure 6 illustrates an example of data linkage for formwork. While the linked database provides comprehensive resources, users emphasized that summarizing essential data would enhance efficiency. In response, we provide ‘Summary & Key points’ so that users can quickly understand the main content as shown in Figure 6. Additionally, dashboard was designed to reflect expert feedback to differentiate safety standards according to specific construction methods. In case of formwork, it is divided into gang form, aluminum form, euro form, and other according to the construction method, so that users can easily access the relevant information for each type. Given the complexity of risk factors, they were derived based on construction work types (major) in TSWT classification system to enable integrated management across work types with overlapping risks. It provides information on key risk factors based on R/C, which are a construction work type (major) in Figure 6. Safety management priorities may vary depending on construction methods, necessitating a more tailored approach. The current classification of key risk factors is based on existing documents, but real-time updates reflecting actual accidents could lead to more proactive safety management.

Figure 6.

Example of a dashboard design in formwork for TSWT safety standards and data linkage after expert feedback.

Conclusions

This study developed a safety management database- linked platform dashboard, designed to assist construction owners with limited expertise in safety management. The platform enables efficient execution of safety management tasks by addressing user requirements identified through detailed analysis. The research achieved its goals by focusing on three key objectives. First, a time-series work type (TSWT) classification system was established based on expert surveys using the Delphi method, reflecting the temporal flow of apartment construction work. This system provides a structured framework for organizing safety management data according to the actual construction process. Given that the level of risk in the construction industry varies depending on the type of activity [32], it is essential to establish a standardized work type classification system that can categorize risk factors and safety management data effectively. Second, essential platform design elements were identified through an analysis of user needs, resulting in features tailored to the practical requirements of safety managers and field supervisors at construction sites. These features were designed to ensure that the platform offers applicable, on-site functionality. Finally, a prototype of the platform dashboard was developed and its usability was validated through expert evaluation. The findings suggest that the safety management platform developed in this study holds significant potential for practical implementation, with minor refinements needed for full operational readiness. It is anticipated that, with further enhancements, this platform could serve as an effective tool for improving safety management at apartment construction sites.

The findings of this study make significant contributions to various aspects of construction site management. The developed TSWT classification system offers a structured framework for organizing the work types involved in apartment construction, and its application could extend beyond safety management to areas such as quality and environmental management. Furthermore, the analysis of user needs helps bridge the gap between research outcomes and field application by identifying the functions required by managers and practitioners in real-world settings. Additionally, the proposed platform dashboard design demonstrates potential for broader application beyond apartment construction project owners, with possibilities for expansion to platforms serving contractors or supervisors. Moreover, if customized for workers with limited specialized knowledge, the platform could promote the development of more proactive and preventive safety management practices.

This study opens several avenues for future research. First, with the proposed prototype of a safety management database-linked platform, evaluating its real- world impact on the construction industry is essential. Future studies could assess the platform’s effectiveness in accident prevention and identify areas for enhancement. Second, as this study is based on safety-related data from South Korea, further research could explore applying safety management data from other countries or regions, allowing for platform customization to fit local safety practices. Lastly, the platform is designed to be expandable, capable of integrating diverse datasets beyond safety-related data, such as construction accident records or material information. Future research could utilize this flexibility to develop more comprehensive and effective safety management systems.