Next Article in Journal
Developing Programs for Converting MIDAS GEN to ANSYS Models Based on Python
Next Article in Special Issue
Tower Crane Layout Planning: Multi-Optimal Solutions Algorithm
Previous Article in Journal
Automated Surface Crack Identification of Reinforced Concrete Members Using an Improved YOLOv4-Tiny-Based Crack Detection Model
Previous Article in Special Issue
Research on the Loss Rule of the Leakage Problem in Residential Construction Based on Water Spray and Storage Tests
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Buildings
Volume 14
Issue 11
10.3390/buildings14113403
buildings-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Systematic Literature Review on Knowledge-Driven Approaches for Construction Safety Analysis and Accident Prevention

by
Sonali Pandithawatta
1,
Seungjun Ahn
2,*,
Raufdeen Rameezdeen
1,
Christopher W. K. Chow
1 and
Nima Gorjian
1
1
Sustainable Infrastructure and Resource Management, UniSA STEM, University of South Australia, Adelaide, SA 5000, Australia
2
Department of Civil and Environmental Engineering, Hongik University, P506, 94 Wausanro, Mapo-gu, Seoul 04066, Republic of Korea
*
Author to whom correspondence should be addressed.
Buildings 2024, 14(11), 3403; https://doi.org/10.3390/buildings14113403
Submission received: 5 August 2024 / Revised: 7 October 2024 / Accepted: 16 October 2024 / Published: 26 October 2024
(This article belongs to the Special Issue Application of Innovative Technologies in Construction Project Management)
Browse Figures
Versions Notes

Abstract

:
Due to its inherent complexities in the process and the dynamic interactions with external environmental factors, the construction industry is widely considered one of the most hazardous industries worldwide. With advancements in artificial intelligence (AI), construction safety management practices have increasingly used knowledge-driven approaches. Such incorporation of knowledge-based methods has led to significant improvements in various elements of construction safety management systems, including hazard identification and risk assessment, selection of risk mitigation strategies, analysis of accident information, sharing of health and safety knowledge, access to regulations, and identification of applicable safety requirements. Against this background, this paper presents a systematic literature review to provide an overview of the current state of the art in the use of knowledge-driven approaches in construction safety management. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) procedure, this study reviews how the knowledge-driven approach is utilized in the construction safety management field to automate different activities that come under it. Journal papers published from 2000 were considered for this review, and the analysis focused on the contributions of research, the evolution of knowledge-driven approaches, sources of incorporated knowledge, methods of system development, yearly publications, and publication by journals. The results provide a comparison of related studies over two decades and offer insights into trends and gaps in this research field. Notably, the trend analysis shows a dramatic increase in the number, as well as the depth, of research efforts utilizing AI techniques for analyzing unstructured data, such as construction images and texts from construction documents, and drawing data-based decisions for accident prevention.

1. Introduction

The construction industry, characterized by its inherent complexity, is widely recognized as one of the most hazardous sectors that accounts for about 20% of occupational fatalities worldwide. In addition to the social costs imposed on the community, construction accidents incur a significant economic burden on contractors [1]. Several studies have identified that safety issues are correlated with hazardous working conditions and the lack of supervision and thus emphasize the essential role of construction management in ensuring safety and preventing accidents [2]. As the active identification of potential workplace hazards is the core process of construction safety management [3], construction managers’ knowledge and decisions play a crucial role in the construction industry in ensuring a safe work environment for all stakeholders.
Construction safety management is a knowledge-intensive activity [4]. The process of hazard identification in the construction industry, for example, is subject to a larger number of variables and unknowns compared to other industries [5], and safety management personnel rely on various health and safety information and knowledge to identify hazards and develop corresponding mitigation strategies [6]. Typically, construction safety-related information and knowledge are gathered from a range of sources and stored in unstructured formats. These sources include expert experience, safety regulations, construction drawings and organization plans, incident databases, and other project documents [6,7]. The project management team is required to stay informed about this content but faces practical difficulties in referring to these documents due to tight project schedules and work pressure [1]. Furthermore, the management and reuse of this knowledge have not been sufficiently formalized, hampering stakeholders from effectively implementing safety management processes in the construction industry [4]. Thus, many researchers have tried to address this problem by undertaking knowledge representation and reasoning techniques, i.e., knowledge-driven approaches.
The knowledge-driven approach, a branch of artificial intelligence (AI), has found application in addressing complex problems and providing decision-making support for real-world scenarios [8]. Knowledge-driven systems can store and manipulate the domain knowledge, including the heuristic knowledge of domain experts, to offer solutions that mimic those of the experts [9,10]. Knowledge-driven systems include ontologies, knowledge graphs, and systems that use knowledge bases such as expert systems. An ontology is a formal representation of knowledge through a set of concepts and relationships within a specific domain [11]. Ontologies facilitate the capturing, reusing, and integrating of domain knowledge [12] and make it accessible to both humans and computer systems [13]. A knowledge graph is a multi-relational graph consisting of entities (nodes) and relations (edges). Each edge is indicated as a triple of the form (head, relation, and tail), representing that two entities are linked by a specific relation [14], e.g., (Welding, Creates, Fumes). A knowledge base is a collection of rules, facts, and assumptions used to store knowledge in a machine-readable format [11,15].
As the construction industry becomes increasingly knowledge-driven and information-intensive [16], research on knowledge management has gained popularity over the last two decades [17]. With advancements in AI and its applications, research on knowledge management has become more inclined to use knowledge-driven approaches to advance construction safety management practices. Several researchers have conducted reviews on related topics, observing the substantial growth of AI applications in the construction industry. For example, Abioye, Oyedele, Akanbi, Ajayi, and Delgado [8] and Pan and Zhang [10] conducted reviews on AI applications in the construction industry; Chen and Bria [18] performed a review on ontology-based safety management in construction; and Zhou, Goh, and Shen [13] reviewed ontology studies supporting the industry’s development. However, none of these studies specifically focused on the scope of knowledge-driven studies in construction safety management. Therefore, it would be worthwhile to survey the state-of-the-art knowledge-driven methods used in the construction safety management field to understand their contributions and to identify trends, gaps, and opportunities in the research field. The aim of this study is, therefore, to provide a comprehensive systematic review of existing knowledge-driven approaches utilized by researchers in the construction safety management field. To guide the systematic review, the following research questions (RQs) were formulated:
  • RQ1: What contributions have existing knowledge-driven research studies made to the construction safety management field?
  • RQ2: How have different knowledge-driven methods been utilized in the construction safety management field?
  • RQ3: What level of automation (e.g., automated, semi-automated, or manual) has been achieved in this research field?
  • RQ4: What types of knowledge sources have been incorporated into the studies?

2. Method

The study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting checklist to ensure replicability, as shown in Figure 1. Specifically, four main criteria were adopted to guide the search of papers: (1) The search was limited to academic databases such as Scopus, Google Scholar, Web of Science (WOS), Emerald, ProQuest, and, in some cases, journal websites to search for papers published in selected journals. (2) Only journal papers were considered for the review, taking into account the rigorous peer-review process undertaken to ensure that the papers meet high-quality standards of validity and reliability. Further, the document type of the journal papers was limited to ‘Articles’ and ‘Reviews,’ and the language was selected as ‘English’ considering the authors’ limited fluency in other languages. (3) Existing reviews over the past two decades have built a profile of relevant publications, which have proven suitable for understanding the developmental changes and advancements within the targeted topic [19,20]. According to Pan and Zhang [10], the growth of data and computational power has steadily increased since 2002, contributing to various aspects of the construction industry. Prior to 2002, the most commonly adopted AI-enabled method was expert systems, and those systems were relatively simple and intuitive in their operation. Taking these facts into consideration, the search period for this review was limited from 2000 to 2023, allowing for an additional 2-year margin. (4) The selection of keywords was performed considering the three main aspects of the review: construction industry, safety, and knowledge-driven approaches. Furthermore, other related reviews were referred to in order to identify specific keywords. Accordingly, three types of keywords were combined using Boolean operations, and the search was performed based on the following rule: (“web base*” OR “knowledge graph” OR ontolog* OR knowledge-driven OR “expert system” OR “knowledge base” OR automat*) AND (safety OR hazards) AND (“construction industry” OR “construction sector” OR “construction setting” OR “construction activit*” OR “construction management” OR “construction project” OR “construction safety” OR “construction site”). In order to enhance the search results, a fuzzy search was executed using the TITLE-ABS-KEY approach, which involved considering the titles, abstracts, and keywords of the documents.
As a result, a total of 685 documents were chosen after removing duplications that were retrieved from different sources. Based on the identified records, a single Excel file was prepared to capture key information such as the title, year of publication, authors’ names and affiliations, abstract, keywords, and journal name, with the aim of facilitating the subsequent analysis. Consequently, the authors conducted a screening process, examining the titles and abstracts of the papers to identify those that were relevant to the study, resulting in a selection of 76 papers. Subsequently, the authors carefully and thoroughly reviewed the full text of these 76 papers to determine their suitability for inclusion in this review. During this examination, the authors specifically focused on identifying studies that demonstrated contributions to construction safety management through the adoption of knowledge-driven approaches, thereby enhancing the ease of execution in various related functions. Additionally, studies that indirectly contributed to such an approach were also selected for the review, even if they did not develop a knowledge-driven system explicitly. This rigorous selection process yielded a final set of 54 papers, comprising 53 articles and one review, which met the criteria for inclusion in the study. The Excel sheet was then updated with the information from this final set of papers, specifically relevant to the research questions. The updated Excel sheet was then transformed into Table 1, which was specifically designed to present the obtained results in an effective and organized manner.

3. Findings

3.1. Evolution of Studies in Knowledge-Driven Safety Management in Construction

To analyze and visualize the interconnectivity of the selected papers and their influence, a citation network visualization tool, Litmaps [71], was utilized. This program facilitated the examination of relationships and patterns among the selected papers through visualization, allowing for a comprehensive understanding of their interconnectedness and impact. Figure 2 represents a map of interconnections among the selected papers based on the number of citations they received from subsequent research studies. The x-axis shows the total citations received by each paper, which is further represented by the diameter of the nodes. The y-axis indicates the publication timeline. Additionally, a color code was adopted to represent the different knowledge-driven approaches utilized by the researchers.
The research on knowledge modeling started as early as the 1980s [72]. However, despite the introduction of different ontology development methods since then, the utilization of ontology within the construction industry-related research progressed at a slower pace compared to other sectors. According to Zhou (2016) [13], ontology-related research in construction only intensified from 2001 onward due to various issues in the construction industry, such as lack of information interoperability, low productivity, lack of automation, and inefficient use of prior knowledge. However, light was shed on this technique again in 2011 by Wang and Boukamp (2011) [64], with a specific focus on improving access to a company’s Job Hazard Analysis (JHA) knowledge. Their research work sparked a boom of ontology-related studies in the field again, constituting 49% of the total knowledge-driven studies, and contributed to enhanced safety performance throughout the different phases of construction safety management.
During the period from 2000 to 2011, researchers predominantly developed web-based systems and expert systems by incorporating knowledge bases into their system structures. Additionally, researchers also created other types of systems (not classified as web-based or expert systems), including a knowledge base [66,70], as well as web-based systems that did not incorporate a knowledge base [68], during this period. In this paper, the knowledge-driven systems that were not categorized as web-based or expert systems but still incorporated a knowledge base are classified under the knowledge base approach to prevent an excessive number of categories.
The knowledge graph is still a novel technique for researchers in construction safety. Since its introduction by Fang et al. [40], several researchers have been exploring its potential to enhance safety management practices. These include automated retrieval of relevant construction safety standards and JHA [24,39]. Additionally, some researchers have contributed by developing knowledge structures to facilitate the construction of domain-specific knowledge graphs for the effective management of construction safety risks [27,28]. Due to its ability to provide more sophisticated reasoning and inference capabilities, as well as greater flexibility and interoperability, knowledge graph approaches have been identified as having a greater potential to contribute to the field of construction safety management.
Figure 2 also represents that the studies are interconnected through citations irrespective of whether the study focuses on developing a knowledge graph, ontology, knowledge base, expert system, or web-based system. The main reason behind such influence networks could be the common principles and concepts that underlie each of these study approaches. These approaches aim to formalize knowledge using knowledge representation and reasoning techniques to assist decision-making processes. Among these links, knowledge-graph studies have a substantial number of connections with ontology studies, indicating a significant influence. This is thought to be because ontologies serve as the foundational layer of knowledge graphs [73]. Knowledge graphs refer to a graph-based method for representing and organizing knowledge, and thus they can greatly benefit from the concept of ontology as it provides a standardized and structured framework for defining concepts and relationships. This is evident as researchers have identified the suitability of ontology development methods in constructing the blueprint of knowledge graphs [24,39,74,75]. Therefore, ontology-related studies play an influential role in knowledge graph-related studies, demonstrating a significant relationship between them.

3.2. Descriptive Analysis of Publications Data

The yearly number of publications provides an effective measure of the level of attention and focus of the knowledge-driven approaches received within the field of construction safety management. As depicted in Figure 3, during the period of 2000–2013, related research progressed at a relatively slow pace. This observation is supported by the fact that a substantial portion, specifically 83.33% of the total research, was conducted after the year 2013. This indicates a notable increase in research activity and interest in knowledge-driven approaches in construction safety management in the years that followed. More specifically, in 2022, a total of 14 studies were published, accounting for 25.93% of the overall research, indicating a substantial increase in related studies in recent years. This trend appears to be continuing, as in the first five months of 2023 alone, six related studies have already been published. This growing interest in related research can be attributed to the persistently high number of accidents in the construction industry [76] as well as the recent developments in AI technologies and the industry’s tendency to incorporate them for enhanced performance [10].
The number of citations and publications by a journal can indicate the impact and influence of those journals on the field and the quality of the journal papers. As depicted in Figure 4, Automation in Construction has the highest number of papers as well as the most citations, making it dominant in the field. The journal Safety Science stands in the second position in terms of both the number of citations and the number of papers. Both Automation in Construction and Safety Science initially published their papers relevant to this area in 2014 and 2015, respectively. Thus, the high number of publications and citations recorded over nearly a decade strongly indicates their influence in the field of construction safety management. The Journal of Construction Engineering and Management and the Journal of Computing in Civil Engineering hold the third and fourth positions, respectively, in terms of the highest number of citations. Collectively, they account for 12.20% of the total citations, with just two related papers published in each journal. Despite the limited number of papers, the substantial citation count signifies the significant impact of these journals on the field. The journal Engineering, Construction, and Architectural Management holds the fifth position in terms of total citations, with only a single paper that has a noteworthy influence in the field. The five most cited journals account for 65.12% of the total citations, indicating a notable influence on the field of construction safety management.

3.3. State of the Art in Knowledge-Driven Approaches to Construction Safety Management

This section analyzes the contributions of previous studies that have utilized different knowledge-driven approaches in various areas of the construction safety management field. The section is divided into five subsections, depending on the main technical aspects, such as knowledge base, knowledge graph, ontology, expert system, and web-based approaches.

3.3.1. Developments in Construction Safety Knowledge Bases

The knowledge base is established through the accumulation of domain expert knowledge, past cases or experiences, and other pertinent sources. This knowledge base forms the basis for implementing an advanced control system, offering practical assistance in resolving identified issues. With its numerous advantages, including enhanced productivity, efficiency, and seamless access to a substantial repository of domain knowledge, the application of this approach in the field of construction safety management presents an ideal solution [8,69].
Rey-Merchán et al. [29] developed an Internet of Things (IoT) system integrated with Java Fuzzy Markup Language (JFML) for a Fuzzy Logic System (FLS). The purpose of this system is to protect workers from falls from height (FFH). The system consists of sensors that collect data related to significant variables in FFH hazards. Based on the information collected by the sensors, working conditions and risk levels can be evaluated. Utilizing expert knowledge on FFH, a knowledge base, and a rule base were established, which form the FLS. JFML receives input data from the sensors and assigns them to the input variables defined in the FLS. Inferences are then made, activating rules based on the input values and defined rules. Consequently, output values are transmitted to actuators to signal the corresponding risk levels. In their study, Guo et al. [51] introduced a Big Data-based platform for observing workers’ behavior, which combined traditional behavior observation with advanced technologies. This integration aimed to overcome the limitations associated with traditional Behavior-Based Safety (BBS) approaches. The platform featured a behavioral risk knowledge base, which served as the foundation for behavior observation, addressing the issue of different observers potentially interpreting the same behavior differently. The knowledge base not only served as a reference for behavior observation but also functioned as a proactive tool for managing safety on-site.
Rozenfeld, Sacks, and Rosenfeld [66] developed a model for Construction Hazard Assessment With Spatial and Temporal Exposure (CHASTE) for predicting construction risk levels to support proactive safety management. The model is dependent on both time and space and facilitates automated calculations to quantify risk levels. CHASTE utilizes a knowledge base comprising construction activities and probabilities of loss-of-control events, combined with a project’s construction plan and a digital building model, to predict risk levels for work teams. In their study, Elbeltagi, Hegazy, Hosny, and Eldosouky [70] proposed a practical model for schedule-dependent site layout planning that combines multiple artificial intelligence tools to generate, optimize, and re-organize construction site layout plans. The proposed model consists of five main components: a flexible representation of the site, facilities, and placement options; a knowledge base for facility identification and area determination; fuzzy-logic assessment to address the vagueness and ambiguity in relationships among the facilities; and a genetic algorithm search for an optimal layout solution; and direct integration with a scheduling tool to identify the site requirements based on the schedule. This proposed approach improves the efficiency of site planning through the integration of all site layout planning tasks into a single environment.
The integration of BIM with knowledge bases enables the utilization and storage of information extracted from past designs and projects [77]. In their study, Haji et al. [21] integrated BIM and a knowledge base of safety leading indicators to develop a safety management framework. This framework facilitates the decision-making process of safety managers and project participants in order to prevent accidents. The knowledge required to develop the knowledge base was extracted from experts’ expertise, documents on safety leading indicators, and best practices. These best practices refer to the safety-leading indicators that have occurred in previous similar projects. Liu et al. [23] established a knowledge base by proposing a structured identification method of safety risks based on the knowledge inherent in metro design specifications, previous journal papers, and expert experience. This safety knowledge base is integrated with BIM software as a plug-in for automatic safety risk inspection, which facilitates safety hazard identification and elimination in the design phase. Xiahou et al. [30] built a safety management knowledge base by identifying logical relationships between accidents and the design process using accident causality and trajectory theories. With the use of an improved FEC (Frequency, Exposure, and Criticality) risk quantification method and rules embedded in the knowledge base, design-oriented subway construction safety risks were quantified. Finally, the researchers proposed a method to transform safety design knowledge and risk quantification into a computer language that Revit can understand. This facilitates the automated review of design schemes and the quantification of safety risks related to design. In their study, Yuan et al. [45] proposed an automated rule-based inspection plug-in by integrating BIM and a knowledge base for Prevention through Design (PtD). The PtD knowledge base was constructed by referencing safety regulations, best practices, safety documents, construction safety risks, and pre-control measures. Additionally, an automatic detection plug-in was developed using Revit API and Visual Studio. The effectiveness of this method was validated through a case study, which demonstrated its capability to efficiently identify construction safety risks and effectively utilize the PtD knowledge base for standardized design.
The Case-Based Reasoning (CBR) technique captures lessons from past problem-solving experiences to find solutions to new problems [78]. CBR models consist of knowledge bases to facilitate the precise and comprehensive representation of past cases, making them advantageous for use in the construction safety management field to enable systematic feedback of past knowledge. Lu et al. [22] developed a CBR platform for the realization of automation in construction safety risk management. An accident attribute system was established within the platform for precise and comprehensive case representation. These cases include information about the problems, solutions, and their contextual details. Thus, these cases represent experiential knowledge derived from previous problem-solving instances or domain expertise. The developed system aims to provide this risk-related knowledge by retrieving similar past cases from the library to support decision-makers in performing more accurate analyses of construction safety risks. The retrieval process adopts the k-Nearest Neighbor (k-NN) algorithm to compare cases in the case library with the new case by computing their similarities. Jiang et al. [41] introduced a decision-making approach for construction safety risk management based on ontology and CBR. They integrated similarity and correlation algorithms to enhance the performance of the CBR algorithm. The approach effectively identifies the case with the highest similarity by calculating case similarity and correlation.
Goh and Chua [67] proposed an approach for construction hazard identification based on Case-Based Reasoning (CBR), which facilitates systematic feedback of past knowledge through incident cases and hazard identification. The approach consists of two main components: a comprehensive knowledge representation scheme and an intelligent retrieval mechanism. The retrieval mechanism utilizes the knowledge stored in semantic networks and the importance weights derived from past cases to measure their similarity to the input case. The system also offers detailed information regarding the reasoning behind the retrieval of a stored case as relevant.

3.3.2. Developments in Knowledge Graph Applications for Construction Safety

Pandithawatta, Ahn, Rameezdeen, Chow, Gorjian, and Kim [24] developed a Job Hazard Analysis Knowledge Graph (JHAKG), which represents both the implicit and explicit knowledge inherent in the Job Hazard Analysis (JHA) process. The incorporation of safety experts’ implicit knowledge into the development process of the JHAKG schema has resulted in a comprehensive knowledge graph with increased reasoning capability. A hypothetical case study has been performed to demonstrate that the JHAKG can answer queries regarding hazards, external conditions, the level of risks, and appropriate control measures to mitigate safety risks. Wang and El-Gohary [25] proposed an information extraction and information modeling method to automatically extract safety requirements based on their syntactic and semantic features. They adopted a fully supervised deep learning-based method which uses a Convolutional Neural Networks (CNN)-based method for information extraction from construction safety requirements. Then, the extracted information was represented in the form of knowledge graph-based queries to facilitate faster and more comprehensive information retrieval. The proposed method was tested by applying it to 20 OSHA sections and calculating relation extraction performance metrics.
Xu, Chang, Xiao, Zhang, Li, and Gu [27] and Xu, Zhang, Gu, Li, and Wang [28] developed a rule-based Chinese-language natural language processing (C-NLP) approach for extracting domain knowledge elements and hierarchical relations. They revealed Chinese linguistic patterns and linguistic features from domain text documents in the context of metro construction safety risk management. The researchers introduced a hybrid model combining a co-word co-occurrence network (CCN) and association rule mining (ARM) to identify connected knowledge elements and expand the domain knowledge base. The resulting knowledge structure enriches safety management knowledge in metro construction, enhances decision-making capabilities, and improves the levels of knowledge-based safety risk management. In their study, Jiang, Gao, Su, and Li [39] developed a knowledge graph aimed at facilitating efficient analysis, querying, and sharing of knowledge related to construction safety standards. Their research involved a thorough investigation of the content within 218 safety standards to construct a robust knowledge graph that comprehensively captures and represents essential information pertaining to construction safety standards. The resulting knowledge graph serves as a valuable resource, empowering stakeholders in the construction industry to access and exchange knowledge on safety standards through the utilization of NLP.
Fang, Ma, Love, Luo, Ding, and Zhou [40] introduced an approach integrating computer vision with ontology technology to develop a knowledge graph that can automatically and precisely identify hazards while complying with safety regulations. Their developed knowledge graph comprised three main elements: (i) an ontological model for hazards, (ii) knowledge extraction, and (iii) knowledge inference for hazard detection. By implementing this approach, site management is equipped with a proactive means to identify, document, and analyze unsafe behaviors, enabling them to take necessary action to minimize and mitigate the risk of falls from heights hazards.

3.3.3. Developments in Construction Safety Management Ontologies

Gao, Ren, and Li [4] developed a domain ontology for construction health and safety management to effectively share health and safety knowledge and assist professionals in making risk mitigation recommendations for decision-making. Pedro et al. [31] built an ontological approach guided by the Linked Open Terms (LOT) methodology to semantically model the construction safety information and formalize knowledge related to accident cases. Guo and Goh [48] developed an ontology to represent the design knowledge of active fall protection systems (AFPS), aiming to facilitate knowledge sharing and reuse. The ontology was constructed using METHONTOLOGY, which is recognized as one of the prominent methods for ontology development [79]. The performance of the ontology was assessed through automated consistency checking, as well as criteria-based and task-based evaluations. In their study, Xing et al. [46] developed a domain ontology to formalize safety risk knowledge related to metro construction to support safety risk identification. The resultant ontology can be used for knowledge sharing and reuse across various parties and computer applications. Furthermore, it can contribute to the development of domain-specific decision support systems. Ensuring consistent descriptions of domain knowledge, it enables the effective and rational acquisition of accurate risk checklists for decision-making purposes. Zhong, Li, Luo, Zhou, Fang, and Xing [3] introduced an ontological approach to represent potential safety risks implied in construction images. With their proposed approach, safety professionals can manually annotate images using sentences or phrases, treating them as query sentences for subsequent processing with NLP. Based on similarity values, similar standard specifications are then recommended. This enables professionals to select the most similar specification as the standard and identify construction safety risks implied in the images.
Wang [55] introduced an approach to semi-automatically identify safety requirements from construction safety standards based on ontological modeling and document modeling techniques. These techniques helped to model the safety-related concepts and requirements in semantically rich, human-readable, and computer-interpretable formats. The approach also includes reasoning mechanisms that leverage the semantically rich nature of a concept ontology to assess the applicability of safety concepts and identify relevant safety requirements. Zhong and Li [57] developed a method that facilitates the representation of construction risk knowledge in a computer-interpretable and semantically inferable way. The resulting risk-oriented ontology model can be utilized as the underlying structure of a knowledge-based risk management system for the construction industry. Lu et al. [58] designed a meta-model for construction safety checking, which consisted of a comprehensive knowledge base organized into five main grouping concepts: Line of work, Task, Precursor, Hazard, and Solution. When the Task and Precursor concepts are entered, the semantic relationships between these concepts automatically extract potential hazards and appropriate solutions. Their approach has effectively facilitated the reuse of safety-checking knowledge.
Le et al. [61] proposed a Social Network System for Sharing Construction Safety and Health Knowledge (SNSS) by integrating semantic wiki web and ontology technologies to enhance communication and representation of construction safety information. The system consists of three main components: Safety Information Module (SIM), Safety Knowledge Module (SKM), and Safety Dissemination Module (SDM). SIM transfers the information to the SKM and facilitates the process of knowledge contribution and refinement by domain experts through ontology tagging and wiki editing, converting safety information into safety knowledge. The SDM module enables users to monitor, manage, and retrieve safety information and knowledge from the system. Gangolells and Casals [62] proposed an approach to implement incorporated environmental and health and safety management systems in construction organizations. They developed a domain ontology to represent the knowledge related to environmental and health and safety operational control on-site, using the seven-step ontology development method. Ontological representation and management of the knowledge related to integrated on-site environmental and health and safety management facilitate the sharing and reuse of knowledge among domain experts and stakeholders engaged in construction sites simultaneously. Wang and Boukamp [64] developed a framework aimed at improving access to a company’s JHA knowledge by eliminating the complexity and time-consuming nature of the traditional JHA process. The proposed framework includes a representation model and a reasoning mechanism. The representation model aims to provide a systematic structure for modeling JHA knowledge in a computer-interpretable format, and the ontological reasoning mechanism is used for identifying applicable safety rules. Consequently, the resulting construction safety domain ontology was capable of checking the applicability of safe approaches among existing JHA documents for a given activity, task, and potential hazard. In their study, a combination of activity, task, and potential hazard defines the ‘condition’ for an unsafe scenario. Their work showcased the effectiveness of using an ontology-based knowledge representation and reasoning technique to provide a shared understanding of the domain of construction safety management and a formal machine-readable model of the domain knowledge.
Farghaly et al. [33] developed a Safety and Health Exchange (SHE) ontology that codifies the construction safety knowledge that can be accessible through online tools that are compatible with the Industry Foundation Classes (IFC) schema. Their research established a foundation for a decision-making tool that utilizes both BIM and ontological approaches. Collinge et al. [34] proposed a digital tool and safety risk library to support designers in work related to health and safety in BIM digital environments. A risk/treatment ontology was developed by closely studying health and safety documents and collaborating with industry experts. This ontology constitutes a fundamental component of the proposed tool. Shen et al. [35] introduced a BIM-based construction process safety risk inspection system through the integration of ontology and NLP to provide real-time safety risk information and prevention measures. The incorporated construction safety management ontology represents the knowledge of construction processes and accidents, precursor information, and corresponding safety solutions. Li et al. [36] introduced the first steps towards a unifying formal domain model of construction safety that consists of a semantically enriched ontology that builds on and integrates with existing construction safety ontologies and BIM. Li et al. [37] presented a semantic web approach aimed at integrating diverse safety risk data and text-based regulatory knowledge within a BIM environment. Their approach included four interconnected ontologies to provide a semantic schema for enhanced automated subway construction safety checking.
Shen et al. [38] developed a safety risk management system for prefabricated building construction, integrating BIM and ontology technology. The results of their study show that ontology technology can be integrated into the Revit software through a plug-in, enabling the sharing, reuse, and accumulation of knowledge on safety risk management. This allows for the detection of problems related to the design and construction of prefabricated buildings that fail to meet specification requirements. Ding et al. [53] introduced an ontology-based framework that automatically derives relationships between construction risk factors, causes, and preventive measures by leveraging the advantages of BIM, ontology, and semantic web technology. They developed a prototype system as a tool to facilitate the management and reuse of construction risk knowledge, with the aim of indirectly improving the construction risk analysis process. The system enables users to achieve the following: (1) identifying different construction processes and the potential risks associated with monitoring objects; (2) analyzing risk factors and pathways; and (3) implementing risk precautions to prevent accidents resulting from recognized risks. Zhang et al. [59] developed a construction safety ontology to formalize the knowledge of construction safety management. This ontology comprises a construction product model, a construction process model, and a construction safety model. It is integrated with the BIM platform to visualize inferred knowledge, including required safety protection systems and protective safety zones.
Li et al. [26] introduced a novel framework that integrates computer vision, ontology, and natural language processing (NLP) to improve construction safety management by enhancing hazard prevention and elimination. The integrated construction safety ontology was manually developed following the seven-step ontology development method proposed by Stanford University [80]. Wu, Zhong, Li, Love, Pan, and Zhao [6] proposed a framework that integrates ontology technology and computer vision to facilitate semantic reasoning for hazards and their associated mitigations, ultimately enhancing safety management. The framework consists of three key modules: the computer vision module, the ontology module, and the knowledge reasoning module. Within this framework, the visual information obtained from images is inserted into the ontology model as instances to facilitate knowledge reasoning. Zhang et al. [42] proposed an automated method to identify risks in the construction process and prevent accidents by combining object detection and ontology technology. They utilized a faster region-convolutional neural network (R-CNN) to extract low-level semantic information from scene elements and captured spatial element relationship attributes from images obtained from a surveillance video. Xiong et al. [44] developed an Automated Hazards Identification System (AHIS) for the construction industry by applying a graph-structured method to depict the on-site construction scene. By utilizing a construction safety ontology, the system can evaluate operation descriptions derived from site videos and compare them to safety guidelines extracted from text documents.
Chen et al. [32] presented an innovative framework based on graphs that combines linguistic and visual information to analyze regulatory rule sentences and images, aiming to identify on-site occupational hazards. The proposed framework included a linguistic information extraction module to transform natural language sentences into processable graph structure representation. This module comprised a novel ontology that can work together with NLP techniques to automatically extract semantic features from regulatory rules. In their study, Chi et al. [60] discovered the possibility of leveraging current construction safety resources to assist JHA, expecting to minimize the required level of human effort. Authors applied ontology-based text classification to match safe approaches identified in existing resources with unsafe scenarios. Various document modification strategies such as ‘Information Retrieval’ were applied to existing resources in order to achieve superior text classification effectiveness. The results of this study provide significant assistance in executing the JHA process by automatically retrieving applicable safety documents to different situations.

3.3.4. Developments in Web-Based Knowledge Base Systems for Construction Safety

Poghosyan et al. [43] introduced a novel web-based design for Occupational Safety and Health (DfOSH) capability maturity model to assess the DfOSH ability of design companies in the built environment. The necessary data to design the web-based system was collected through a literature review, expert focus group discussions, and a Delphi study. The analysis of the data resulted in DfOSH capability attributes, maturity levels, and attribute weights, which are the main features of the web-based system. The resultant system can be used by design companies to assess areas of capability deficiency and strength. In their study, Goh and Guo [47] introduced a web-based system to support the design and selection process of Active Fall Protection Systems (AFPS). The hybrid system incorporates a combination of case-based reasoning (CBR) and rule-based reasoning (RBR) to enhance retrieval performance. The system includes a case library consisting of fifty real work-at-height design cases. They utilized the previously developed AFPS-Ontology to define the case structure and query vocabulary [48]. By integrating both CBR and RBR approaches, the web-based system simulates the knowledge and behavior of expert AFPS designers. Park et al. [56] invented a web-based Construction Safety Management Information System (CSMIS) by investigating risk factors and accidents associated with various work processes for workers and general safety managers. The CSMIS facilitates the efficient identification of high-risk activities on construction sites and corresponding improvement measures. Furthermore, the CSMIS can be employed for safety education and training purposes.
Kamardeen [63] conducted a study aimed at introducing a knowledge-based approach to OHS planning. In this study, he developed a web-based system to address the inherent limitations associated with traditional OHS planning methods. The system comprised three logical layers: the client side, the middle layer, and the server side. On the client side (tier one), the user is connected to the knowledge base located on the server side (tier three) through the intermediary middle layer (tier two). This configuration allows the user to access all the OHS knowledge contents stored within the knowledge base. The web-based system underwent testing and evaluation by various end-user groups in the construction industry, and the results indicate numerous advantages of the proposed system to builders. In their study, Cooke, Lingard, Blismas, and Stranieri [68] developed a web-based decision support tool by capturing and structuring expert reasoning regarding design impacts on occupational health and safety (OHS). They adopted an innovative method to structure the knowledge that is well suitable in the context of uncertainty and discretionary decision-making. The final knowledge representation structure, in the form of an argument tree, was translated into a program to facilitate an interactive automated risk assessment. The development of this tool provided convenient access to expert OHS information and decision support in an area where learning from one’s mistakes is undesirable. Cheung, Cheung, and Suen [69] developed a web-based Construction Safety and Health Monitoring (CSHM) system to create an automated safety and health management tool. CHSM comprises four main components: (i) Web-based Interface, (ii) Knowledge Base, (iii) Output Data, and (iv) Benchmark Group. The knowledge base comprises a concise compilation of expert advice and essential guidelines necessary for safety and health management. Its purpose is to support and complement the automated assessment system by offering practical advice for addressing identified problem areas.

3.3.5. Developments in Expert Systems for Construction Safety

The Expert Systems technique offers a powerful tool for integrating knowledge and implementing autonomous inference mechanisms [81]. They are specifically designed to tackle complex problems through knowledge-based reasoning, emulating the expertise of human professionals [52]. By leveraging this technology, the construction safety management field can reduce reliance on domain experts, thus providing valuable advantages. Amiri et al. [49] introduced a fuzzy probabilistic expert system designed for occupational hazard assessment within the construction industry. The system utilizes a fuzzy probabilistic model that incorporates a rule base generated through fuzzy risk-based statistical analysis and data mining of accident databases. Additionally, expert interviews and an extensive literature review were conducted to enhance the accuracy and comprehensiveness of the rule base. Birgonul et al. [50] developed a rule-based expert system for quantifying fault rates in fall accidents in the Turkish construction industry. The validation of the system confirmed its ability to successfully simulate experts’ judgments in quantifying fault rates. In their study, Adeyemi et al. [54] proposed an expert system that can assess the risk associated with manual lifting in the construction industry and provide first aid advice. This system serves as a quick and efficient decision-making tool for manual material lifting workplaces that comply with the application of the NIOSH lifting equation. Kamardeen [65] developed an expert system to automate the Workers’ Compensation Insurance (WCI) premium-rating model for building projects. Based on the results of a literature review and a questionnaire survey, a novel premium-rating model for WCI was developed. To automate the proposed model, a conceptual model of a fuzzy expert system was developed through a combination of interviews and analysis of historical workers’ compensation claims data. It was then prototyped and subsequently tested and validated using Turing tests.
Zhang, Wu, Ding, Skibniewski, and Lu [52] introduced a novel approach to improve the traditional risk identification process in tunnel construction by integrating BIM and expert systems. This expert system comprises three subsystems: BIM extraction, knowledge base, and risk identification subsystem. The integrated knowledge base consists of a fact base, a rule base, and a case base. A hybrid inference approach combining CBR and rule-based reasoning (RBR) was introduced to improve the system’s reasoning capacity.

4. Discussion and Future Trends

The construction industry has witnessed a growing interest in knowledge-driven approaches that have the potential to automate different areas of construction safety management. This emerging research topic explores the application of advanced techniques to leverage knowledge and optimize safety management practices, aiming to enhance efficiency and effectiveness in the construction industry. This study analyzes various knowledge-driven approaches used by researchers in the field of construction safety management to understand the contributions of their studies to industry practices.
Out of the total articles considered in this study, 13 articles have integrated BIM into their research, indicating its popularity in the field. However, while there are studies available where BIM is integrated with ontologies, knowledge bases, and expert systems, none of the studies have explored the integration of knowledge graphs with BIM. Although there are studies that integrate BIM and knowledge graphs [82], indicating the potential of such integration, these studies do not specifically focus on the field of construction safety management. With the ability to capture rich semantic relationships between entities and enable a more nuanced representation of information, knowledge graphs can facilitate more advanced reasoning and analysis, support complex queries, and enable the discovery of hidden connections within the BIM data, thereby offering opportunities for enhanced information representation.
Despite the advantages that BIM integration could bring to knowledge-driven approaches, there are also limitations that hinder such integrations. Two previous studies conducted by Wang and Boukamp [64] and Pauwels et al. [83] pointed out that the IFC standard, which is commonly used for BIM, does not primarily focus on addressing semantic issues. As a result, the relationships defined in IFC are often too general and lack the specificity required to represent more precise semantic information. This limitation can impact the capability of IFC to accurately represent and convey specific concepts of safety management in a comprehensive manner. Further, the integration of BIM has the potential to diminish the advantages that knowledge-driven approaches can bring. This is because the effectiveness of knowledge-driven approaches heavily relies on the level of BIM adoption and implementation. Furthermore, in developing countries, the adoption of BIM remains relatively low [84,85,86,87]. Given that construction safety is a major concern in these regions [88,89,90,91], integrating knowledge-driven approaches with BIM could potentially restrict their utilization in the countries where they would have the greatest impact and prove most beneficial.
As depicted in Table 1, most of the studies have used manual methods to develop their knowledge graphs, ontologies, knowledge bases, expert systems, and web-based systems. One of the main advantages of using manual methods is that it allows for the integration of experts’ expertise into the development process. Knowledge-driven approaches heavily rely on the specialized knowledge of experts, which is acquired through their training, education, and practical experience. Therefore, integrating this expertise is crucial for creating a comprehensive system with enhanced reasoning capability. However, despite the fact that acquiring experts’ expertise requires specialized methods such as the Delphi technique, very few researchers have adopted such techniques to develop the underlying knowledge structure of their systems [24,43,68].
In addition to the use of manual methods, researchers have also utilized semi-automated and automated methods in their development processes. Among these methods, NLP and machine learning algorithms are the most commonly employed techniques for automatically developing knowledge-driven systems. With the ability to capture relationships between entities from large volumes of text, machine learning techniques can be employed on databases, such as incident databases, to automatically generate knowledge graphs. However, the application of such techniques in construction safety management is limited. One possible reason for this could be the challenge of identifying complex relationships that exist between numerous variables, which is difficult to capture without human intervention.
When it comes to systems that were specifically built for hazard identification and risk assessment, very few studies were able to develop an approach that can be applied to all types of hazards. For example, most CV approaches have a limited ability to identify biological and chemical hazards and perform risk assessments based on contextual characteristics. Furthermore, some knowledge structures of the knowledge-driven systems were specifically built to process information about specific hazards only. Thus, researchers still have the opportunity to explore the creation of a common knowledge structure that can be embedded into a knowledge-driven system to perform universal hazard identification and risk assessment.
Recently, chatbot applications like ChatGPT have garnered significant attention across various fields, suggesting a future where users can ask complex questions and receive natural language responses with minimal delay. The potential for industrial applications of such chatbots is being explored in many sectors, including construction. This creates the possibility of developing a tool using a knowledge-driven approach, where construction planners can ask chatbots safety management questions and receive comprehensive information on hazard and risk management.

5. Conclusions

Construction safety management, being a knowledge-intensive task, can greatly benefit from the utilization of knowledge-driven approaches, enabling automation and streamlining processes. By harnessing these approaches, the industry can leverage the power of structured knowledge representation and reasoning to enhance the performance of various applications, such as information analysis, the development of decision support systems, risk assessment, and knowledge sharing. This utilization enables the enhancement of safety practices and the effective mitigation of risks in construction management. With the advancements in AI, the utilization of these approaches in the construction safety management field has significantly improved in recent years to address the issue of an increased number of accidents in the construction industry. Given that this study aims to conduct a systematic review of knowledge-driven approaches used in the field of construction safety management following the PRISMA protocol. Accordingly, the studies that developed expert systems, web-based systems, and other systems that use ontologies, knowledge graphs, and knowledge bases were reviewed to mainly understand their contributions to the construction safety management field. This study also analyzes the evolution of knowledge-driven approaches utilized by researchers over time and proposes trending approaches. Furthermore, the number of publications per year, number of citations, and number of publications per journal were analyzed to understand the attention and focus given to knowledge-driven approaches within the field of construction safety management as well as the impact and influence of the journals on the field and the quality of their papers.
The study has several limitations, and readers should interpret the findings within the context of these limitations. The search for research papers was conducted using 17 keywords, limiting the extraction of papers to those where the specified keywords were mentioned in their title, abstract, and keywords sections. The search for journal papers was only conducted in limited academic databases and journal websites. This process influenced the extraction of relevant journal papers for this study. However, despite these limitations, a sufficient number of papers were successfully extracted, allowing for a comprehensive understanding of trends and gaps in the field. Even though there are a significant number of studies in the field, there are still numerous areas in the construction safety management field that require improvement. Researchers can utilize knowledge-driven approaches to address these areas and bring the benefits of advancements in AI technologies to the construction industry.

Author Contributions

Conceptualization, S.P.; methodology, S.P.; software, S.P.; formal analysis, S.P. and S.A.; writing—original draft preparation, S.P. and S.A.; writing—review and editing, S.P., S.A., R.R., C.W.K.C. and N.G.; visualization, S.P.; supervision, S.A., R.R., C.W.K.C. and N.G.; funding acquisition, S.A. All authors have read and agreed to the published version of the manuscript.

Funding

A part of the research work presented in this paper and the APC was funded by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. 2022R1F1A1074448).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Kamardeen, I. Web-based Safety Knowledge Management System for Builders: A Conceptual Framework. In Proceedings of the CIB W099 Conference, Melbourne, Australia, 21–23 October 2009. [Google Scholar]
  2. Li, X.; Yi, W.; Chi, H.-L.; Wang, X.; Chan, A.P. A critical review of virtual and augmented reality (VR/AR) applications in construction safety. Autom. Constr. 2018, 86, 150–162. [Google Scholar] [CrossRef]
  3. Zhong, B.; Li, H.; Luo, H.; Zhou, J.; Fang, W.; Xing, X. Ontology-Based Semantic Modeling of Knowledge in Construction: Classification and Identification of Hazards Implied in Images. J. Constr. Eng. Manag. 2020, 146. [Google Scholar] [CrossRef]
  4. Gao, S.; Ren, G.; Li, H. Knowledge Management in Construction Health and Safety Based on Ontology Modeling. Appl. Sci. 2022, 12, 8574. [Google Scholar] [CrossRef]
  5. Mihić, M. Classification of Construction Hazards for a Universal Hazard Identification Methodology. J. Civ. Eng. Manag. 2020, 26, 147–159. [Google Scholar] [CrossRef]
  6. Wu, H.; Zhong, B.; Li, H.; Love, P.; Pan, X.; Zhao, N. Combining computer vision with semantic reasoning for on-site safety management in construction. J. Build. Eng. 2021, 42, 103036. [Google Scholar] [CrossRef]
  7. Ding, L.Y.; Zhou, C. Development of web-based system for safety risk early warning in urban metro construction. Autom. Constr. 2013, 34, 45–55. [Google Scholar] [CrossRef]
  8. Abioye, S.O.; Oyedele, L.O.; Akanbi, L.; Ajayi, A.; Delgado, J.M.D. Artificial intelligence in the construction industry: A review of present status, opportunities and future challenges. J. Build. Eng. 2021, 44, 103299. [Google Scholar] [CrossRef]
  9. Lee, S.S.G.; Ngoi, B.K.A.; Lim, L.E.N.; Tan, P.S. 24—Assembly Systems. In Knowledge-Based Systems; Leondes, C.T., Ed.; Academic Press: Cambridge, MA, USA, 2000; pp. 729–753. [Google Scholar] [CrossRef]
  10. Pan, Y.; Zhang, L. Roles of artificial intelligence in construction engineering and management: A critical review and future trends. Autom. Constr. 2021, 122, 103517. [Google Scholar] [CrossRef]
  11. Subhashini, R.; Akilandeswari, J. A survey on ontology construction methodologies. Int. J. Enterp. Comput. Bus. Syst. 2011, 1, 1–19. [Google Scholar]
  12. El-Diraby, T.; Kashif, K.F. Distributed ontology architecture for knowledge management in highway construction. J. Constr. Eng. Manag. 2005, 131, 591–603. [Google Scholar] [CrossRef]
  13. Zhou, Z.; Goh, Y.M.; Shen, L. Overview and Analysis of Ontology Studies Supporting Development of the Construction Industry. J. Comput. Civ. Eng. 2016, 30, 04016026. [Google Scholar] [CrossRef]
  14. Wang, Q.; Mao, Z.; Wang, B.; Guo, L. Knowledge Graph Embedding: A Survey of Approaches and Applications. IEEE Trans. Knowl. Data Eng. 2017, 29, 2724–2743. [Google Scholar] [CrossRef]
  15. Liu, Z.; Han, X. Deep Learning in Knowledge Graph. In Deep Learning in Natural Language Processing; Deng, L., Liu, Y., Eds.; Springer: Singapore, 2018; pp. 117–145. [Google Scholar] [CrossRef]
  16. Nepal, M.P.; Staub-French, S. Supporting knowledge-intensive construction management tasks in BIM. J. Inf. Technol. Constr. 2016, 21, 13–38. [Google Scholar]
  17. Yu, D.; Yang, J. Knowledge Management Research in the Construction Industry: A Review. J. Knowl. Econ. 2018, 9, 782–803. [Google Scholar] [CrossRef]
  18. Chen, W.T.; Bria, T.A. A Review of Ontology-Based Safety Management in Construction. Sustainability 2023, 15, 413. [Google Scholar] [CrossRef]
  19. Irani, Z.; Kamal, M.M. Intelligent Systems Research in the Construction Industry. Expert Syst. Appl. 2014, 41, 934–950. [Google Scholar] [CrossRef]
  20. Martinez, P.; Al-Hussein, M.; Ahmad, R. A scientometric analysis and critical review of computer vision applications for construction. Autom. Constr. 2019, 107, 102947. [Google Scholar] [CrossRef]
  21. Haji, M.D.; Behnam, B.; Sebt, M.H.; Ardeshir, A.; Katooziani, A. BIM-Based Safety Leading Indicators Measurement Tool for Construction Sites. Int. J. Civ. Eng. 2023, 21, 265–282. [Google Scholar] [CrossRef]
  22. Lu, Y.; Yin, L.; Deng, Y.; Wu, G.; Li, C. Using cased based reasoning for automated safety risk management in construction industry. Saf. Sci. 2023, 163, 106113. [Google Scholar] [CrossRef]
  23. Liu, P.; Shang, Y.; Zhang, L. A Design for Safety (DFS) Framework for Automated Inspection Risks in Metro Stations by Integrating a Knowledge Base and Building Information Modeling. Int. J. Environ. Res. Public Health 2023, 20, 4765. [Google Scholar] [CrossRef]
  24. Pandithawatta, S.; Ahn, S.; Rameezdeen, R.; Chow, C.W.K.; Gorjian, N.; Kim, T.W. Development of a Knowledge Graph for Automatic Job Hazard Analysis: The Schema. Sensors 2023, 23, 3893. [Google Scholar] [CrossRef] [PubMed]
  25. Wang, X.; El-Gohary, N. Deep learning-based relation extraction and knowledge graph-based representation of construction safety requirements. Autom. Constr. 2023, 147, 104696. [Google Scholar] [CrossRef]
  26. Li, Y.; Wei, H.; Han, Z.; Jiang, N.; Wang, W.; Huang, J. Computer Vision-Based Hazard Identification of Construction Site Using Visual Relationship Detection and Ontology. Buildings 2022, 12, 857. [Google Scholar] [CrossRef]
  27. Xu, N.; Chang, H.; Xiao, B.; Zhang, B.; Li, J.; Gu, T. Relation Extraction of Domain Knowledge Entities for Safety Risk Management in Metro Construction Projects. Buildings 2022, 12, 1633. [Google Scholar] [CrossRef]
  28. Xu, N.; Zhang, B.; Gu, T.; Li, J.; Wang, L. Expanding Domain Knowledge Elements for Metro Construction Safety Risk Management Using a Co-Occurrence-Based Pathfinding Approach. Buildings 2022, 12, 1510. [Google Scholar] [CrossRef]
  29. Rey-Merchán, M.C.; López-Arquillos, A.; Soto-Hidalgo, J.M. Prevention of Falls from Heights in Construction Using an IoT System Based on Fuzzy Markup Language and JFML. Appl. Sci. 2022, 12, 6057. [Google Scholar] [CrossRef]
  30. Xiahou, X.; Li, K.; Li, F.; Zhang, Z.; Li, Q.; Gao, Y. Automatic identification and quantification of safety risks embedded in design stage: A bim-enhanced approach. J. Civ. Eng. Manag. 2022, 28, 278–291. [Google Scholar] [CrossRef]
  31. Pedro, A.; Pham-Hang, A.T.; Nguyen, P.T.; Pham, H.C. Data-Driven Construction Safety Information Sharing System Based on Linked Data, Ontologies, and Knowledge Graph Technologies. Int. J. Environ. Res. Public Health 2022, 19, 794. [Google Scholar] [CrossRef]
  32. Chen, S.; Demachi, K.; Dong, F. Graph-based linguistic and visual information integration for on-site occupational hazards identification. Autom. Constr. 2022, 137, 104191. [Google Scholar] [CrossRef]
  33. Farghaly, K.; Soman, R.K.; Collinge, W.; Mosleh, M.H.; Manu, P.; Cheung, C.M. Construction safety ontology development and alignment with industry foundation classes (IFC). J. Inf. Technol. Constr. 2022, 27, 94–108. [Google Scholar] [CrossRef]
  34. Collinge, W.H.; Farghaly, K.; Mosleh, M.H.; Manu, P.; Cheung, C.M.; Osorio-Sandoval, C.A. BIM-based construction safety risk library. Autom. Constr. 2022, 141, 104391. [Google Scholar] [CrossRef]
  35. Shen, Q.; Wu, S.; Deng, Y.; Deng, H.; Cheng, J.C.P. BIM-Based Dynamic Construction Safety Rule Checking Using Ontology and Natural Language Processing. Buildings 2022, 12, 564. [Google Scholar] [CrossRef]
  36. Li, B.; Schultz, C.; Teizer, J.; Golovina, O.; Melzner, J. Towards a unifying domain model of construction safety, health and well-being: SafeConDM. Adv. Eng. Inform. 2022, 51, 101487. [Google Scholar] [CrossRef]
  37. Li, X.; Yang, D.; Yuan, J.; Donkers, A.; Liu, X. BIM-enabled semantic web for automated safety checks in subway construction. Autom. Constr. 2022, 141, 104454. [Google Scholar] [CrossRef]
  38. Shen, Y.; Xu, M.; Lin, Y.; Cui, C.; Shi, X.; Liu, Y. Safety Risk Management of Prefabricated Building Construction Based on Ontology Technology in the BIM Environment. Buildings 2022, 12, 765. [Google Scholar] [CrossRef]
  39. Jiang, Y.; Gao, X.; Su, W.; Li, J. Systematic knowledge management of construction safety standards based on knowledge graphs: A case study in China. Int. J. Environ. Res. Public Health 2021, 18, 10692. [Google Scholar] [CrossRef]
  40. Fang, W.; Ma, L.; Love, P.E.D.; Luo, H.; Ding, L.; Zhou, A. Knowledge graph for identifying hazards on construction sites: Integrating computer vision with ontology. Autom. Constr. 2020, 119, 103310. [Google Scholar] [CrossRef]
  41. Jiang, X.; Wang, S.; Wang, J.; Lyu, S.; Skitmore, M. A decision method for construction safety risk management based on ontology and improved cbr: Example of a subway project. Int. J. Environ. Res. Public Health 2020, 17, 3928. [Google Scholar] [CrossRef]
  42. Zhang, M.; Zhu, M.; Zhao, X. Recognition of High-Risk Scenarios in Building Construction Based on Image Semantics. J. Comput. Civ. Eng. 2020, 34, 04020019. [Google Scholar] [CrossRef]
  43. Poghosyan, A.; Manu, P.; Mahamadu, A.M.; Akinade, O.; Mahdjoubi, L.; Gibb, A.; Behm, M. A web-based design for occupational safety and health capability maturity indicator. Saf. Sci. 2020, 122, 104516. [Google Scholar] [CrossRef]
  44. Xiong, R.; Song, Y.; Li, H.; Wang, Y. Onsite video mining for construction hazards identification with visual relationships. Adv. Eng. Inform. 2019, 42, 100966. [Google Scholar] [CrossRef]
  45. Yuan, J.; Li, X.; Xiahou, X.; Tymvios, N.; Zhou, Z.; Li, Q. Accident prevention through design (PtD): Integration of building information modeling and PtD knowledge base. Autom. Constr. 2019, 102, 86–104. [Google Scholar] [CrossRef]
  46. Xing, X.; Zhong, B.; Luo, H.; Li, H.; Wu, H. Ontology for safety risk identification in metro construction. Comput. Ind. 2019, 109, 14–30. [Google Scholar] [CrossRef]
  47. Goh, Y.M.; Guo, B.H.W. FPSWizard: A web-based CBR-RBR system for supporting the design of active fall protection systems. Autom. Constr. 2018, 85, 40–50. [Google Scholar] [CrossRef]
  48. Guo, B.H.W.; Goh, Y.M. Ontology for design of active fall protection systems. Autom. Constr. 2017, 82, 138–153. [Google Scholar] [CrossRef]
  49. Amiri, M.; Ardeshir, A.; Fazel Zarandi, M.H. Fuzzy probabilistic expert system for occupational hazard assessment in construction. Saf. Sci. 2017, 93, 16–28. [Google Scholar] [CrossRef]
  50. Birgonul, M.T.; Dikmen, I.; Budayan, C.; Demirel, T. An expert system for the quantification of fault rates in construction fall accidents. Int. J. Occup. Saf. Ergon. 2016, 22, 20–31. [Google Scholar] [CrossRef]
  51. Guo, S.Y.; Ding, L.Y.; Luo, H.B.; Jiang, X.Y. A Big-Data-based platform of workers’ behavior: Observations from the field. Accid. Anal. Prev. 2016, 93, 299–309. [Google Scholar] [CrossRef]
  52. Zhang, L.; Wu, X.; Ding, L.; Skibniewski, M.J.; Lu, Y. Bim-Based Risk Identification System in tunnel construction. J. Civ. Eng. Manag. 2016, 22, 529–539. [Google Scholar] [CrossRef]
  53. Ding, L.Y.; Zhong, B.T.; Wu, S.; Luo, H.B. Construction risk knowledge management in BIM using ontology and semantic web technology. Saf. Sci. 2016, 87, 202–213. [Google Scholar] [CrossRef]
  54. Adeyemi, H.O.; Adejuyigbe, S.B.; Ismaila, S.O.; Adekoya, A.F. Low back pain assessment application for construction workers. J. Eng. Des. Technol. 2015, 13, 419–434. [Google Scholar] [CrossRef]
  55. Wang, H.H. Semi-automated identification of construction safety requirements using ontological and document modeling techniques. Can. J. Civ. Eng. 2015, 42, 756–765. [Google Scholar] [CrossRef]
  56. Park, J.; Park, S.; Oh, T. The development of a web-based construction safety management information system to improve risk assessment. KSCE J. Civ. Eng. 2015, 19, 528–537. [Google Scholar] [CrossRef]
  57. Zhong, B.; Li, Y. An Ontological and Semantic Approach for the Construction Risk Inferring and Application. J. Intell. Robot. Syst. Theory Appl. 2015, 79, 449–463. [Google Scholar] [CrossRef]
  58. Lu, Y.; Li, Q.; Zhou, Z.; Deng, Y. Ontology-based knowledge modeling for automated construction safety checking. Saf. Sci. 2015, 79, 11–18. [Google Scholar] [CrossRef]
  59. Zhang, S.; Boukamp, F.; Teizer, J. Ontology-based semantic modeling of construction safety knowledge: Towards automated safety planning for job hazard analysis (JHA). Autom. Constr. 2015, 52, 29–41. [Google Scholar] [CrossRef]
  60. Chi, N.W.; Lin, K.Y.; Hsieh, S.H. Using ontology-based text classification to assist Job Hazard Analysis. Adv. Eng. Inform. 2014, 28, 381–394. [Google Scholar] [CrossRef]
  61. Le, Q.T.; Lee, D.Y.; Park, C.S. A social network system for sharing construction safety and health knowledge. Autom. Constr. 2014, 46, 30–37. [Google Scholar] [CrossRef]
  62. Gangolells, M.; Casals, M. An ontology-based approach for on-site integrated environmental and health and safety management. Rev. Ing. Constr. 2012, 27, 103–127. [Google Scholar] [CrossRef]
  63. Kamardeen, I. E-OHS planning system for builders. Archit. Sci. Rev. 2011, 54, 50–64. [Google Scholar] [CrossRef]
  64. Wang, H.H.; Boukamp, F. Ontology-based representation and reasoning framework for supporting job hazard analysis. J. Comput. Civ. Eng. 2011, 25, 442–456. [Google Scholar] [CrossRef]
  65. Kamardeen, I. An expert system for strategic control of accidents and insurers’ risks in building construction projects. Expert Syst. Appl. 2009, 36, 4021–4034. [Google Scholar] [CrossRef]
  66. Rozenfeld, O.; Sacks, R.; Rosenfeld, Y. ‘CHASTE’: Construction hazard assessment with spatial and temporal exposure. Constr. Manag. Econ. 2009, 27, 625–638. [Google Scholar] [CrossRef]
  67. Goh, Y.M.; Chua, D.K.H. Case-based reasoning for construction hazard identification: Case representation and retrieval. J. Constr. Eng. Manag. 2009, 135, 1181–1189. [Google Scholar] [CrossRef]
  68. Cooke, T.; Lingard, H.; Blismas, N.; Stranieri, A. ToolSHeDTM: The development and evaluation of a decision support tool for health and safety in construction design. Eng. Constr. Arch. Manag. 2008, 15, 336–351. [Google Scholar] [CrossRef]
  69. Cheung, S.O.; Cheung, K.K.W.; Suen, H.C.H. CSHM: Web-based safety and health monitoring system for construction management. J. Saf. Res. 2004, 35, 159–170. [Google Scholar] [CrossRef] [PubMed]
  70. Elbeltagi, E.; Hegazy, T.; Hosny, A.H.; Eldosouky, A. Schedule-dependent evolution of site layout planning. Constr. Manag. Econ. 2001, 19, 689–697. [Google Scholar] [CrossRef]
  71. Litmaps. Available online: https://app.litmaps.com/ (accessed on 29 June 2023).
  72. Yun, W.; Zhang, X.; Li, Z.; Liu, H.; Han, M. Knowledge modeling: A survey of processes and techniques. Int. J. Intell. Syst. 2020, 36, 1686–1720. [Google Scholar] [CrossRef]
  73. AL-Aswadi, F.N.; Chan, H.Y.; Gan, K.H. From Ontology to Knowledge Graph Trend: Ontology as Foundation Layer for Knowledge Graph. In Proceedings of the Knowledge Graphs and Semantic Web Conference, Madrid, Spain, 21–23 November 2002. [Google Scholar]
  74. Qiao, L.; Yang, L.; Hong, D.; Yao, L.; Zhiguang, Q. Knowledge graph construction techniques. J. Comput. Res. Dev. 2016, 53, 582–600. [Google Scholar]
  75. Zhao, Z.; Han, S.K.; Son, I.M. Architecture of Knowledge Graph Construction Techniques. Int. J. Pure Appl. Math. 2018, 118, 1869–1883. [Google Scholar]
  76. Fang, D.; Huang, Y.; Guo, H.; Lim, H.W. LCB Approach for Construction Safety. Saf. Sci. 2020, 128, 104761. [Google Scholar] [CrossRef]
  77. Salehi, S.A.; Yitmen, I. Modeling and analysis of the impact of BIM-based field data capturing technologies on automated construction progress monitoring. Int. J. Civ. Eng. 2018, 16, 1669–1685. [Google Scholar] [CrossRef]
  78. Gupta, U.G. How Case-Based Reasoning Solves New Problems. Interfaces 1994, 24, 110–119. [Google Scholar] [CrossRef]
  79. Fernández-López, M.; Gómez-Pérez, A.; Juristo, N. Methontology: From ontological art towards ontological engineering. In Proceedings of the Fourteenth National Conference on Artificial Intelligence, Providence, RI, USA, 27–31 July 1997. [Google Scholar]
  80. Noy, N.; McGuinness, D. Ontology Development 101: A Guide to Creating Your First Ontology. Knowl. Syst. Lab. 2001, 32, 1–25. [Google Scholar]
  81. Martín, A.; León, C.; Luque, J.; Monedero, I. A framework for development of integrated intelligent knowledge for management of telecommunication networks. Expert Syst. Appl. 2012, 39, 9264–9274. [Google Scholar] [CrossRef]
  82. Peng, J.; Liu, X. Automated code compliance checking research based on BIM and knowledge graph. Sci. Rep. 2023, 13, 7065. [Google Scholar] [CrossRef]
  83. Pauwels, P.; Van Deursen, D.; Verstraeten, R.; De Roo, J.; De Meyer, R. A semantic rule checking environment for building performance checking. Autom. Constr. 2009, 20, 506–518. [Google Scholar] [CrossRef]
  84. Masood, R.; Kharal, M.K.N.; Nasir, A.R. Is BIM Adoption Advantageous for Construction Industry of Pakistan? Procedia Eng. 2014, 77, 229–238. [Google Scholar] [CrossRef]
  85. Memon, A.H.; Rahman, I.A.; Memon, I.; Azman, N.I.A. BIM in Malaysian Construction Industry: Status, Advantages, Barriers and Strategies to Enhance the Implementation Level. Res. J. Appl. Sci. Eng. Technol. 2014, 8, 606–614. [Google Scholar] [CrossRef]
  86. Ahuja, R.; Jain, M.; Sawhney, A.; Arif, M. Adoption of BIM by architectural firms in India: Technology–organization–environment perspective. Arch. Eng. Des. Manag. 2016, 12, 311–330. [Google Scholar] [CrossRef]
  87. Faisal Shehzad, H.M.; Binti Ibrahim, R.; Yusof, A.F.; Mohamed Khaidzir, K.A.; Shawkat, S.; Ahmad, S. Recent developments of BIM adoption based on categorization, identification and factors: A systematic literature review. Int. J. Constr. Manag. 2022, 22, 3001–3013. [Google Scholar] [CrossRef]
  88. Priyadarshani, K.; Karunasena, G.; Jayasuriya, S. Construction safety assessment framework for developing countries: A case study of Sri Lanka. J. Constr. Dev. Ctries. 2013, 18, 33–51. [Google Scholar]
  89. Durdyev, S.; Mohamed, S.; Lay, M.L.; Ismail, S. Key factors affecting construction safety performance in developing countries: Evidence from Cambodia. Constr. Econ. Build. 2017, 17, 48–65. [Google Scholar] [CrossRef]
  90. Boadu, E.F.; Wang, C.C.; Sunindijo, R.Y. Characteristics of the Construction Industry in Developing Countries and Its Implications for Health and Safety: An Exploratory Study in Ghana. Int. J. Environ. Res. Public Health 2020, 17, 4110. [Google Scholar] [CrossRef]
  91. Umeokafor, N.; Okoro, C.; Diugwu, I.; Umar, T. Design for safety in construction in Nigeria: A qualitative inquiry of the critical opportunities. Int. J. Build. Pathol. Adapt. 2023, 41, 476–494. [Google Scholar] [CrossRef]
Buildings 14 03403 g001
Figure 1. PRISMA flow diagram.
Figure 1. PRISMA flow diagram.
Buildings 14 03403 g001
Buildings 14 03403 g002
Figure 2. Citations map of the selected papers. [3,4,6,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70].
Figure 2. Citations map of the selected papers. [3,4,6,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70].
Buildings 14 03403 g002
Buildings 14 03403 g003
Figure 3. Number of publications per year.
Figure 3. Number of publications per year.
Buildings 14 03403 g003
Buildings 14 03403 g004
Figure 4. Number of citations and publications by journal.
Figure 4. Number of citations and publications by journal.
Buildings 14 03403 g004
Table 1. Summary of the selected research studies.
Table 1. Summary of the selected research studies.
AuthorsContribution of ResearchKnowledge-Driven ApplicationSources of KnowledgeDevelopment Process
Haji, Behnam, Sebt, Ardeshir, and Katooziani [21]An add-on to integrate a safety leading indicator knowledge base and BIMA safety leading indicator knowledge baseExperts’ expertise, documents on safety leading indicators, and best practiceManual
Lu, Yin, Deng, Wu, and Li [22]Case-Based Reasoning (CBR) platform for automating construction safety risk management.A knowledge base system for precise and comprehensive accident case representationConstruction accident reportsManual
Liu, Shang, and Zhang [23]Inspection plug-in to BIM softwareSafety design knowledge baseMetro design specifications, journal literature, and expert experienceSemi-automated approach with NLP applications
Pandithawatta, Ahn, Rameezdeen, Chow, Gorjian, and Kim [24]Job Hazard Analysis Knowledge Graph (JHAKG)Knowledge graph to support automated execution of job hazard analysisPrevious JHA documents, expert’s expertise, Australian Code of Practice documentsManual
Wang and El-Gohary [25]Deep learning-based relation extraction method and construction safety requirements knowledge graphA knowledge graph to represent construction safety requirements for compliance checking of site operation with applicable construction safety regulationsOSHA Safety and Health StandardsSemi-automated approach with fully supervised deep learning-based method
Li, Wei, Han, Jiang, Wang, and Huang [26]Computer vision-based hazard identification frameworkConstruction safety ontologySafety handbooks for construction site workers and expert’s expertiseManual
Xu, Chang, Xiao, Zhang, Li, and Gu [27]Domain knowledge elements and hierarchical relations which underpin the knowledge base of metro construction safety risk managementA fine-grained knowledge structure of a knowledge graphText documents of metro construction safety risk managementAutomated approach with C-NLP applications
Xu, Zhang, Gu, Li, and Wang [28]
Gao, Ren, and Li [4]Health and Safety Management-Ontology (HSM-Onto)A domain ontology to structure health and safety knowledge for improved decision makingStandards, technical manuals, occupational Injury and Illness Classification Manuals, and accident reportsManual
Rey-Merchán, López-Arquillos, and Soto-Hidalgo [29]IoT system to protect workers from falls from height hazards.A knowledge base for the representation of expert knowledge on falls from heightsExperts’ expertise, OHS legislation, and construction accident databaseManual
Xiahou, Li, Li, Zhang, Li, and Gao [30]A framework to perform automated examination and visualization of safety risks.A safety management knowledge base.Design regulations, related literature, and best practices.Manual
Pedro, Pham-Hang, Nguyen, and Pham [31]Construction safety information sharing systemAn ontology to represent knowledge pertaining to accident casesAccident case dataManual
Chen, Demachi, and Dong [32]A graph-based framework to process regulatory rules and on-site images for occupational hazard identificationAn ontology capable of integrating with NLP to extract linguistic information, enabling automated processing of regulatory rulesRegulatory rulesManual
Farghaly, Soman, Collinge, Mosleh, Manu, and Cheung [33]An ontology that can be mapped with the IFC schemaSafety and Health Exchange (SHE) ontologyReporting of Injuries, Diseases, and Dangerous Occurrences Regulations (RIDDOR) and press releasesManual
Collinge, Farghaly, Mosleh, Manu, Cheung, and Osorio-Sandoval [34]A digital tool and safety risk library to support designers in BIM digital environmentsRisk/treatment ontologyPrevention through Design online resources, design guidelines, HSE guidance, RIDDOR, press releases, and experts’ expertiseManual
Shen, Wu, Deng, Deng and Cheng [35]A BIM-based construction process safety risk inspection systemConstruction safety management ontologySafety codesManual
Li, Schultz, Teizer, Golovina, and Melzner [36]An effective modeling language that formalizes safety code regulations.An ontology of construction safety that formally captures safety concepts in constructionConstruction safety standards, codes, and research literatureManual
Li, Yang, Yuan, Donkers, and Liu [37]A framework for safety analysis in subway constructionSubway construction safety checking ontologyRegulations and technical manuals, case reports, existing ontologies, and experts’ expertiseManual
Shen, Xu, Lin, Cui, Shi, and Liu [38]A safety risk management system for prefabricated building constructionPrefabricated building construction safety risk ontologyAccident case data, standards, and specificationsSemi-automated approach
Jiang, Gao, Su, and Li [39]An NLP-based question-answering system related to construction safety standardsKnowledge graph of construction safety standard (KGCSS)National, industry, local, and corporate construction safety standardsManual
Wu, Zhong, Li, Love, Pan, and Zhao [6]A conceptual framework integrating computer vision and ontology for construction safety managementOntology to represent construction safety knowledge.Unified Regulation for Construction Quality Acceptance of Construction Engineering (gb50300-2017), Building Engineering Measurement Regulations, Industry Foundation Classes (IFC) standards, Occupational Injury and Illness Classification Manual, and Building Construction Safety Inspection RegulationManual
Fang, Ma, Love, Luo, Ding, and Zhou [40]Knowledge graph framework for hazard identification using computer vision technologyAn ontology helps experts in annotating knowledge and describe the relationships among entities.Engineering documents, historical accident reports, experts’ expertise, and safety codesSemi-automated approach
Jiang, Wang, Wang, Lyu, and Skitmore [41]Construction safety risk management decision-making frameworkSubway construction safety ontologyRelated literature, historical cases, and experts’ expertiseManual
Zhang, Zhu, and Zhao [42]A framework to identify construction risks by using computer vision and ontology technologyOntology to represent construction risk knowledgeSafety-related accident dataManual
Poghosyan, Manu, Mahamadu, Akinade, Mahdjoubi, Gibb, and Behm [43]Design for Occupational Safety and Health Capability Maturity Indicator (DfOSH-CMI) toolA novel web-based design for occupational safety and health (DfOSH) capability maturity modelRelated literature and experts’ expertiseManual
Zhong, Li, Luo, Zhou, Fang, and Xing [3]Proactive approach for construction hazard identification from imagesConstruction hazard ontologyChinese Specification Quality and Safety Inspection Guide of Urban Rail Transit Engineering and Experts’ ExpertiseManual
Xiong, Song, Li, and Wang [44]Automated Hazards Identification System (AHIS)Construction safety ontology to assist the evaluation process of operation descriptions generated from site videos against safety guidelines extracted from the documentsSafety Handbook for Construction Site Workers, The Construction (Design and Management) Regulations 2015 and Recommended Practices for Safety and Health Programs in ConstructionManual
Yuan, Li, Xiahou, Tymvios, Zhou, and Li [45]An automated rule-based inspection plug-inA PtD knowledge base to acquire, store, and make use of the PtD-related knowledge of designersSafety regulations, safety documents, and best practicesManual
Xing, Zhong, Luo, Li, and Wu [46]An ontology to formalize risk knowledge in metro constructionSafety risk identification ontology (SRI-Onto)Standards and technical manuals, case set with related risk research reports, existing research and system platforms, and experts’ expertiseManual
Goh and Guo [47]A decision support system for selecting and designing solutions to work-at-height problems.A web-based system—FPSWizardReal work-at-height scenarios, design standards, and AFPS ontologyManual
Guo and Goh [48]An ontology to provide a formal and shared vocabulary for the domain of AFPS designActive fall protection system ontology (AFPS-Onto)AFPS design standards, AFPS design cases, and experts’ expertiseManual
Amiri, Ardeshir, and Fazel Zarandi [49]A fuzzy probabilistic expert system for occupational hazard assessment in the construction industryA knowledge base of fuzzy rulesAccident databases, experts’ expertise-related literatureSemi-automated approach
Birgonul, Dikmen, Budayan, and Demirel [50]An expert system for the quantification of fault rates in construction fall accidentsA knowledge base of if-then rulesConstruction-related inspection reports, related literature, and experts’ expertiseManual
Guo, Ding, Luo, and Jiang [51]A Big Data-based platform to classify, collect, and store workers’ behavior dataA behavioral risk knowledge baseSafety standards, operating instructions, accident cases, and experts’ expertise.Manual
Zhang, Wu, Ding, Skibniewski, and Lu [52]A BIM-based Risk Identification Expert System (B-RIES)A knowledge base to systematize the fragmented risk identification knowledge in tunnel construction to facilitate knowledge sharing and communicationFact base, rule base, and an accident case baseManual
Ding, Zhong, Wu, and Luo [53]A tool to facilitate the construction risk knowledge managementAn ontology to model and represent construction risk knowledgeDocuments that store construction risk knowledgeManual
Adeyemi, Adejuyigbe, Ismaila, and Adekoya [54]Musculoskeletal disorders—risk evaluation expert system (MSDs-REES)Knowledge base with fuzzy rulesExperts’ expertiseManual
Wang [55]An approach to identify applicable safety requirements from construction safety standardsOntology to model the safety-related concepts and their relationshipsOSHA standardsManual
Park, Park, and Oh [56]Construction Safety Management Information System (CSMIS)A web-based system that facilitates faster and more convenient risk assessmentRisk assessment guidelines, disaster cases, standards for safe work guidelines, safety terms, and regulations for industrial safety and healthManual
Zhong and Li [57]An approach to present construction risk knowledge in a computer-interpretable and semantically inferable wayAn ontology to represent the construction risk knowledgeRelevant building technical codes, construction manuals, best-practice construction rules, experts’ expertise, and relevant study literatureManual
Lu, Li, Zhou, and Deng [58]An ontology-based knowledge model for automated construction safety checkingConstruction Safety Checking Ontology (CSCOntology)Center to Protect Workers’ Rights (CPWR) construction solution database and OSHA regulationsManual
Zhang, Boukamp, and Teizer [59]An approach to organize, store, and reuse construction safety knowledgeConstruction safety ontology to formalize the current construction safety knowledgeOSHA regulations, Occupational Injury and Illness Classification Manual, and Construction Solutions DatabaseManual
Chi, Lin, and Hsieh [60]An approach to leverage available construction safety resources to assist JHA with a minimum level of human effortA construction safety domain ontologyCPWR construction solution database, NIOSH FACE reports, and OSHA standardsSemi-automated approach
Le, Lee, and Park [61]A Social Network System for Sharing Construction Safety and Health Knowledge (SNSS)A safety ontology that offers a classification framework for safety, representing the correlation between safety information and their respective significanceAccident, hazard, and risk records and experts’ expertiseManual
Gangolells and Casals [62]An approach to implement integrated environmental and health and safety management systemsA domain ontology to represent an integrated knowledge model for operational control at construction sitesInternational standards for environmental management systems and occupational health and safety management systemsManual
Kamardeen [63]A web-based system to implement a knowledge-based OHS planning approachOHS knowledge baseCodes of practice, best practice manuals, textbooks, and research publicationsManual
Wang and Boukamp [64]A framework aiming to improve access to a company’s JHA knowledgeConcept ontology to represent JHA conceptsOccupational Injury and Illness Classification Manual, MasterFormat 2004 Edition from Construction Specifications Institute and JHA documentsManual
Kamardeen [65]An automated WCI premium-rating modelA knowledge base of risk rate inferring rules and membership functionsExperts’ expertise, relevant literature, and past workers’ compensation claimsManual
Rozenfeld, Sacks, and Rosenfeld [66]An approach to predict risk levels in construction projects to support proactive safety managementA knowledge base of construction activities and probabilities of loss-of-control eventsExperts’ expertise and Construction Job Safety Analysis (CJSA) databaseManual
Goh and Chua [67]A CBR approach for construction hazard identificationA knowledge base of hazard identification and incident casesPast hazard identification and incident casesManual
Cooke, Lingard, Blismas, and Stranieri [68]A decision support tool for design OHS in the construction industry (ToolSHeD)A web-based system built upon an argument tree that structures the knowledge regarding design impacts upon OHSExperts’ expertise, OHS guidance material, industry standards, and codesManual
Cheung, Cheung, and Suen [69]A web-based Construction Safety and Health Monitoring (CSHM) systemA knowledge base that facilitates online expert advice and instructionsRules, guidelines, best practices, and experts’ expertiseManual
Elbeltagi, Hegazy, Hosny, and Eldosouky [70]A practical model for schedule-dependent construction site layout planningA knowledge base for facility identification and area determinationConstruction safety and health manuals, company handbooks, published dissertations, and technical articlesManual
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Pandithawatta, S.; Ahn, S.; Rameezdeen, R.; Chow, C.W.K.; Gorjian, N. Systematic Literature Review on Knowledge-Driven Approaches for Construction Safety Analysis and Accident Prevention. Buildings 2024, 14, 3403. https://doi.org/10.3390/buildings14113403

AMA Style

Pandithawatta S, Ahn S, Rameezdeen R, Chow CWK, Gorjian N. Systematic Literature Review on Knowledge-Driven Approaches for Construction Safety Analysis and Accident Prevention. Buildings. 2024; 14(11):3403. https://doi.org/10.3390/buildings14113403

Chicago/Turabian Style

Pandithawatta, Sonali, Seungjun Ahn, Raufdeen Rameezdeen, Christopher W. K. Chow, and Nima Gorjian. 2024. "Systematic Literature Review on Knowledge-Driven Approaches for Construction Safety Analysis and Accident Prevention" Buildings 14, no. 11: 3403. https://doi.org/10.3390/buildings14113403

APA Style

Pandithawatta, S., Ahn, S., Rameezdeen, R., Chow, C. W. K., & Gorjian, N. (2024). Systematic Literature Review on Knowledge-Driven Approaches for Construction Safety Analysis and Accident Prevention. Buildings, 14(11), 3403. https://doi.org/10.3390/buildings14113403

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop