Page 185
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
A Systematic Review of Current Risk Assessment Practices in
Construction Projects
Maureen Ambrose
1
, Nur Fadilah Darmansah
2
, Azilah Baddiri
2
1
Postgraduate Student, Universiti Malaysia Sabah, 88400, Kota Kinabalu Sabah Malaysia
2
Universiti Malaysia Sabah, 88400, Kota Kinabalu Malaysia
DOI:
https://doi.org/10.51583/IJLTEMAS.2026.150300017
Received: 13 March 2026; Accepted: 18 March 2026; Published: 02 April 2026
ABSTRACT
Purpose: This systematic review examines current risk assessment practices in construction projects, analyzing
methodological approaches, technological integration, and thematic priorities in literature published between
2020 and 2026. The study aims to identify prevailing risk assessment practices and research gaps while providing
a comparative analysis of methodological evolution.
Methodology: Following PRISMA guidelines, a systematic literature review was conducted across Scopus, Web
of Science, and Dimensions databases. The search targeted peer-reviewed articles on construction project risk
assessment published from January 2020 to February 2026. From 847 initial records, 187 articles met inclusion
criteria and were analyzed using bibliometric and content analysis methods with inter-reviewer agreement
validation.
Findings: Analysis reveals quantitative methods dominate (47%), followed by qualitative (32%) and mixed-
methods (21%). Current practices comprise traditional techniques (31%), advanced methods including AI and
machine learning (37%), hybrid approaches (18%), and real-time assessment (14%). Advanced methods
demonstrate 15–25% higher prediction accuracy but face implementation barriers including data requirements
and interpretability challenges. Real-time assessment, despite 35–50% accident reduction potential in trials,
remains limited due to cost and infrastructure constraints. Significantly, 65% of studies exclude post-
construction variables, and developing economies remain underrepresented (24%).
Originality: This review provides the first comprehensive synthesis of risk assessment practices spanning 2020-
2026, quantifying methodological trends and identifying that 65% of studies lack integration of post-construction
variables. The findings establish a baseline for understanding the evolution toward technology-enabled,
proactive risk assessment frameworks in construction projects.
Keywords: Risk assessment, risk estimation, hazard identification, construction projects, systematic review,
quantitative methods, artificial intelligence, sustainable risk management
INTRODUCTION
The construction company operates in conditions marked by widespread uncertainty, where proficient risk
management is essential for determining project success or failure. Risk assessment, fundamental to risk
management, is the methodical identification, analysis, and evaluation of potential hazards that could affect
project objectives (Almashhour et al., 2025). Despite extensive research and practice over several decades,
building projects still encounter substantial cost overruns, schedule delays, and safety mishaps, indicating
enduring discrepancies between theoretical frameworks and practical execution.
The scale of these challenges is significant. Research demonstrates that 55% of construction projects and 65%
of public infrastructure projects in developing economies experience cost overruns (Yussif et al., 2025).
Conventional construction generates approximately 35% of global waste, whereas sustainable techniques
demonstrate potential to reduce waste by 50% during construction phases (Yussif et al., 2025). These statistics
Page 186
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
underscore the critical need for robust risk assessment approaches adapted to contemporary construction
environments.
Recent years have seen significant advancements in risk assessment methodologies. The amalgamation of
artificial intelligence (AI), big data analytics, and real-time monitoring technologies offers improved predictive
capabilities and proactive risk management (Kumi et al., 2024). The rise of green building construction projects
concurrently presents new risk factors that necessitate specialised assessment frameworks (Yussif et al., 2025).
The COVID-19 pandemic intensified interest in robust and flexible risk management systems, leading scientists
to reframe conventional methods via the perspectives of sustainability and resilience (Almashhour et al., 2025).
Research Gap and Novelty
Despite increasing academic interest, the current literature lacks a systematic synthesis that compares the
methodological evolution of traditional, advanced, hybrid, and real-time approaches. Previous reviews
investigate individual strategies (Kumi et al., 2024; Almashhour et al., 2025) yet do not offer a comparative
examination of their respective strengths, limitations, and contextual relevance. Moreover, no thorough
assessment examines the trade-offs among accuracy, implementation cost, and scalability, which are essential
factors for industry adoption. This review specifically addresses three significant gaps: (1) a systematic
comparison of four methodological categories utilising quantifiable performance metrics, (2) a trade-off analysis
evaluating implementation feasibility across diverse project contexts, and (3) the identification of technology
integration pathways and barriers to adoption.
Current Risk Assessment Practices
Traditional Risk Assessment Methods
Traditional risk assessment approaches remain prevalent owing to their established framework, regulatory
endorsement, and reduced implementation expenses. Fault Tree Analysis (FTA) delineates the root causes of
accidents by depicting the interconnections between equipment failures and dangers (Kabir et al., 2020).
Researchers have used Fault Tree Analysis to examine tower crane falls and scaffolding failures (Wang et al.,
2022; Aljassmi et al., 2013). Failure Mode and Effects Analysis (FMEA) predicts likely failures prior to their
occurrence. Hassan et al. (2022) utilised FMEA to discern risks in high-rise buildings that traditional checklists
failed to recognise. Recent iterations integrate FMEA with fuzzy logic to more effectively address uncertainty
(Chen & Lee, 2023), enabling a nuanced analysis of ambiguously defined or non-quantifiable risks.
The Analytic Hierarchy Process (AHP) aids decision-makers in assessing various risks via significance
weighting. Darko et al. (2020) discovered that 42% of infrastructure projects utilise AHP for risk assessment.
Researchers have employed AHP in public-private partnerships and sustainable construction projects (Li et al.,
2018; Nguyen & Macchion, 2023). Risk matrices, fundamental diagrams depicting probability and outcomes,
continue to be extensively utilised despite their constraints. Bao et al. (2023) improved traditional matrices by
using fuzzy bounds to more accurately depict variable site conditions. Checklists and risk registers provide
systematic risk documentation; Wuni and Shen (2022) created specialised checklists for modular building
projects.Traditional methods provide numerous benefits, including governmental approval, fewer training
demands, and accessibility for small- to medium-sized firms. Nonetheless, they exhibit considerable limitations:
they see risks as static in spite of dynamic building settings, rely on subjective judgement prone to cognitive
biases, and fail to reflect intricate risk interrelationships adequately (Aljassmi et al., 2013). These inadequacies
illustrate academics' growing amalgamation of traditional approaches with modern technologies to improve
decision-making and reduce evaluation biases.
Advanced Risk Assessment Methods
Advanced techniques employing artificial intelligence and machine learning provide improved accuracy and the
ability to handle large data sets. Machine learning comprises 45% of advanced methodological research,
forecasting risks with 15–25% greater accuracy than traditional methods (Kumi et al., 2024). Deep learning
evaluates construction site images for safety infractions with an accuracy of 87% (Akinosho et al., 2020). Natural
Page 187
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
language processing examines accident reports to discern emerging risk patterns (Resende et al., 2024).
Predictive models anticipate project delays four weeks ahead utilising real-time data (Zhong et al., 2022).
Bayesian networks are present in 23% of advanced research publications. These models demonstrate causal links
between dangers and revise predictions as new information becomes available, which is crucial for effectively
managing risks in projects like tunnel construction and pipeline developments. Zhang et al. (2021) employed
Bayesian networks to assess tunnel construction risks in the context of geological conditions and monitoring
data. Kabir et al. (2022) examined 43 unique risks associated with pipeline developments, including
environmental impacts, regulatory challenges, and technological failures that can affect project outcomes. Fuzzy
logic tackles confusing terminology in risk assessment, demonstrating notable efficacy in multinational projects
marked by increased uncertainty (Celik et al., 2023; Moheimani et al., 2024), particularly in clarifying risk
factors and improving decision-making processes in complex environments. Big data analytics analyses previous
project data to identify predictors of cost overruns (Adebayo et al., 2025). Although sophisticated approaches
exhibit enhanced accuracy and predictive capacities, they encounter considerable implementation obstacles.
These procedures necessitate significant amounts of high-quality data, which is sometimes inaccessible at
standard construction sites. Numerous AI technologies operate as "black boxes", concealing their decision-
making mechanisms and undermining industry confidence in their results. Furthermore, execution involves
considerable setup expenses and necessitates specialised competence (Albasyouni et al., 2025), hence limiting
their application to large enterprises and research projects rather than conventional construction practices.
Hybrid Risk Assessment Models
Hybrid models integrate many methodologies to optimise the advantages of each. The strategy, usually referred
to as the integration of classical and fuzzy approaches, addresses uncertainty issues. Fuzzy fault tree analysis
integrates the explicit framework of fault tree analysis with the capacity of fuzzy logic to manage ambiguous
probability (Liu et al., 2021). Susanto et al. (2026) discovered that integrating risk assessment with the work
breakdown structure identified 40% additional dangers in foundation work, highlighting the importance of a
comprehensive approach to risk management in construction projects. Fuzzy FMEA, incorporating insights from
several experts, yielded superior risk assessments for high-rise buildings (Wang et al., 2022). The integration of
simulation and analysis yields superior outcomes compared to each method in isolation. Monte Carlo simulation
quantifies uncertainty, whereas system dynamics illustrates the interrelationship of hazards over time.
Nasirzadeh et al. (2023) found that traditional methods missed these interactions, which led to a 34%
underestimation of delay risks. Integrating artificial intelligence with specialised expertise transcends the
limitations of solely data-driven approaches. Choi et al. (2023) integrated case-based reasoning with neural
networks to achieve a delay prediction accuracy of 84%, surpassing the performance of each method
independently. Medaa et al. (2025) indicated that 78% of building collapse investigations currently employ a
combination of methodologies. Incorporating risk assessment into BIM transforms building models into
instruments for risk management. BIM systems can autonomously identify vulnerabilities in building designs,
reducing identification time by 60% (Zou et al., 2020). Contemporary technologies integrate BIM with real-time
sensors to identify evolving risk trends (Wang et al., 2024). Hybrid models provide a more comprehensive
assessment of risk; nevertheless, they are intricate to establish and require expertise in many methodologies,
such as statistical analysis, machine learning, and domain-specific knowledge in architecture and engineering.
Real-Time Risk Assessment Methods
Real-time risk assessment represents an emerging evaluation paradigm for construction projects. Given
constantly evolving site conditions, monitoring risks in real-time is prudent, though obstacles including high
costs and inconsistent site connectivity impede progress. Sensor systems constitute 55% of real-time research.
Wearable devices monitor worker location and physiological indicators. Proximity alerts when workers approach
machinery reduce incidents by 47% (Awolusi et al., 2020). Positioning systems detect hazardous worker-
equipment proximity with 91% accuracy (Li et al., 2021). Environmental sensors monitor dust, noise, and
temperature, reducing exposure incidents by 35% (Kang et al., 2021). Smart helmets identify worker fatigue
before accidents occur (Hwang et al., 2023). Computer vision comprises 32% of real-time research. Camera
systems with intelligent algorithms assess personal protective equipment compliance with 82% accuracy (Fang
et al., 2020). Systems identify hazardous ladder use and excavation risks (Han & Lee, 2022). Integrating camera
Page 188
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
views with architectural models detects personnel in restricted zones with 88% precision (Chen, H., 2024). IoT
platforms integrate these components, with edge computing enabling on-site data processing for immediate alerts
(Ojghaz et al., 2024). Digital twins provide virtual site replicas that continuously update risk assessments
(Almatared et al., 2023). Preliminary trials indicate real-time systems could reduce accidents by 35–50% (Pereira
et al., 2026). However, sensors malfunction in abrasive, dusty environments; workers express privacy concerns;
and managers require training to manage continuous alerts. As costs decrease and evidence accumulates, real-
time assessment is expected to proliferate.
METHODOLOGY
Review Protocol
This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses
(PRISMA) guidelines to ensure methodological rigor and transparency. The PRISMA framework provided
structured protocols for literature search, screening, eligibility assessment, and data synthesis.
Search Strategy
Comprehensive literature searches were conducted in three academic databases: Scopus, Web of Science, and
Dimensions. These databases were selected for their extensive coverage of construction engineering, project
management, and risk assessment literature. The search strategy combined keywords related to construction
projects, risk assessment methodologies, and publication timeframe
Inclusion and Exclusion Criteria
Studies were eligible for inclusion if they met the following criteria: (a) specifically addressed risk assessment
in construction project contexts; (b) presented original empirical research or systematic reviews with clear
methodological descriptions; (c) were published in English; and (d) appeared in peer-reviewed journals or
conference proceedings. Conversely, studies were excluded if they: (a) discussed risk management without a
specific focus on assessment phases; (b) explored contexts beyond the construction sector; (c) comprised
editorials, commentaries, or book reviews devoid of empirical or analytical content; or (d) had inaccessible full
texts.
Screening and Selection Process
The screening process involved four stages. Database searches initially yielded 847 potentially relevant records.
After removing 234 duplicates, 613 records proceeded to title and abstract screening, which excluded 356
records that did not meet inclusion criteria.
The remaining 257 articles underwent full-text eligibility assessment, with two independent reviewers
conducting screening and achieving 89% inter-reviewer agreement. The 11% of disagreements were resolved
through discussion with a third reviewer. To minimize selection bias, search results were systematically
documented, and exclusion reasons were recorded for all full-text rejections. Detailed full-text evaluation
resulted in exclusion of 70 articles due to insufficient methodological focus or peripheral relevance to risk
assessment, yielding a final sample of 187 studies meeting all inclusion criteria.
Data Extraction and Analysis
A standardized data extraction form captured: publication details (authors, year, journal), research context
(project type, geographic focus), risk assessment methodology (qualitative, quantitative, mixed), specific
techniques employed (e.g., Fault Tree Analysis, Monte Carlo simulation, machine learning), technology
integration, and key findings.
Extracted data were analyzed using descriptive statistics to quantify methodological distributions and thematic
patterns. Bibliometric analysis using VOSviewer software identified co-occurrence networks of keywords and
research themes (Ullah et al., 2025).
Page 189
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
RESULTS AND DISCUSSION
Publication Trends and Geographic Distribution
Analysis of the 187 included studies revealed growing scholarly attention to construction risk assessment
between 2020 and 2026. Publication volume increased steadily from 22 articles in 2020 to 41 articles in 2025,
with 15 papers already indexed by early 2026, suggesting continued growth. This upward trajectory reflects
increased recognition of risk assessment's critical role and the emergence of innovative methodological
approaches. Geographically, research originated from 38 countries, with highest contributions from China
(31%), the United States (18%), the United Kingdom (12%), and Australia (8%). Developing economies
contributed 24% of studies. However, Yussif et al. (2025) observed that collaboration between developed and
developing nations remains limited, with only 15% of articles featuring cross-national research teams. This gap
constrains knowledge transfer and method adaptation across different regulatory, economic, and cultural
contexts.
Distribution of Research Methodologies related Risk Assessments
Quantitative methods dominated the literature, employed in 47% (n=88) of reviewed studies. These approaches
encompassed statistical analysis, mathematical modeling, simulation techniques, and AI applications.
Qualitative methods appeared in 32% (n=60) of studies, including case studies, expert interviews, and document
analysis. Mixed-methods approaches constituted 21% (n=39) of the sample. The predominance of quantitative
methods reflects growing emphasis on objective, data-driven assessment. Kumi et al. (2024) noted that statistical
methods appear in 38% of quantitative studies examining risk factor relationships. Mathematical modeling
techniques feature in 27% of quantitative research, while simulation methods are employed in 22%. However,
qualitative and mixed methods approaches maintain significance for exploring contextual factors and stakeholder
perspectives that quantitative techniques may overlook.
Current Risk Assessment Practices in Construction Project
Analysis of 187 studies on current risk assessment methodologies in construction projects reveals the distribution
of methodological approaches. Table 1, comparative analysis of risk assessment Methodologies. Figure 1
categorizes the reviewed techniques into four principal groups: traditional, advanced, hybrid, and real-time risk
assessment methods. The results demonstrate that advanced risk assessment approaches are the most commonly
utilized, followed by traditional, hybrid, and real-time methods, in that order.
Table 1. Comparative Analysis of Risk Assessment Methodologies
Method
Category
Key
Technique
s
Advantages
Limitations
Adoptio
n (%)
Maturity
Level
Scalabilit
y
Tradition
al
FTA,
FMEA,
AHP, Risk
Matrices,
Checklists
Regulatory
acceptance;
Low cost;
Minimal
training;
Accessible to
SMEs
Static
representation;
Subjective bias;
Poor handling of
interdependenci
es
31%
Mature
Limited
Advanced
(AI/ML)
Neural
Networks,
Bayesian
Networks,
NLP, Big
Data
15–25%
higher
accuracy;
Pattern
recognition;
Predictive
capability
Data hunger;
Black-box
opacity; High
setup costs;
Expertise
required
37%
Emerging-
Mature
High
Page 190
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
Hybrid
Fuzzy-
FTA,
Fuzzy-
FMEA,
BIM
Integration
, System
Dynamics
Comprehensiv
e coverage;
Synergistic
advantages;
40% better
hazard
detection
Complexity;
Multiple
expertise
needed;
Integration
challenges
18%
Developin
g
Medium
Real-time
IoT
Sensors,
Computer
Vision,
Wearables,
Digital
Twins
35–50%
accident
reduction;
Immediate
alerts;
Continuous
monitoring
Infrastructure
requirements;
Privacy
concerns; Data
reliability issues
14%
Emerging
Medium
Figure 1. Percentage of current risk assessment methods
Advanced Risk Assessment Methods
Advanced techniques were present in 37% (n=69) of the analysed papers, surpassing traditional methods and
signifying a shift towards technology-driven evaluations. Machine learning for risk prediction was the primary
category, accounting for 45% of advanced method investigations. Neural networks attained 82% accuracy in
predicting safety occurrences when trained on comprehensive datasets (Kumi et al., 2024). Such algorithms
could revolutionise safety management from reactive investigation to proactive intervention. Bayesian network
models (23% of advanced research) provide distinct advantages by integrating probabilistic reasoning with
expert knowledge and refining predictions when new information becomes available (Kabir et al., 2022; Zhang
et al., 2021). These characteristics are especially beneficial when historical data is scarce and expert knowledge
is plentiful.
Page 191
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
The transition to advanced methodologies is driven by three converging factors: exponential growth in
computational capabilities facilitating complex analysis, enhanced accessibility of digital project data from BIM
and IoT sources, and industry demand for more precise predictions in the face of increasing project complexity.
The 37% acceptance rate in literature considerably exaggerates actual industry implementation, given the
majority of research is experimental rather than operational. Despite evident potential, the industry's
preparedness for AI implementation remains inadequate—Poh et al. (2024) found that merely 12% of
construction businesses have the requisite data infrastructure and analytical competencies. The disparity between
academic advancement and industry application, with 35% of studies proposing AI models yet just 18%
incorporating real-world validation, constitutes a significant obstacle necessitating immediate focus from both
Traditional Risk Assessment Methods
Conventional methodologies appeared in 31% (n=59) of investigations. These strategies persist due to validated
efficacy across decades of application (Liu et al., 2024), accessibility requiring neither specialized software nor
technical expertise—particularly advantageous for small and medium enterprises (Fernández-Muñiz et al.,
2024). Regulatory endorsement strengthens their position, as safety standards and insurance mandates explicitly
recognize these procedures.
However, conventional approaches encounter well-documented constraints: static representations fail to capture
evolving site dynamics (Liu et al., 2024); dependence on subjective expert evaluations includes cognitive biases
potentially underestimating certain risk categories (Smith et al., 2022); and limited capacity to manage
interdependencies may overlook cascade effects where minor hazards combine for significant repercussions
(Kabir et al., 2020).
For standard evaluations of ordinary projects with proficient teams, customary procedures suffice. For complex
or high-stakes projects, they should function as foundations upon which more advanced analyses are constructed
researchers and practitioners.
Hybrid Risk Assessment Methods
Hybrid models were present in 18% (n=33) of studies, reflecting an increasing acknowledgement that no singular
approach sufficiently encompasses all risk factors. Susanto et al. (2026) demonstrated that the amalgamation of
work breakdown structure with risk assessment frameworks improved hazard detection efficacy by 40%
compared to isolated approaches, whereas Medaa et al. (2025) reported that 78% of structural collapse studies
currently employ integrated computational, qualitative, and data-driven methodologies.
Hybrid approaches leverage synergistic advantages by combining the structured frameworks and stakeholder
engagement of traditional methods (Darko et al., 2020), the predictive capabilities and extensive dataset analysis
of advanced methodologies (Kumi et al., 2024), and the capacity of qualitative approaches to elucidate contextual
subtleties and organisational dynamics (Almashhour et al., 2025).
Successful integration necessitates careful consideration of reconciliation methods to prevent inconsistent
outcomes (Kiani et al., 2025). This optimal comprehensiveness incurs heightened complexity, necessitating
expertise in several disciplines and extended implementation timescales. Susanto et al. (2026) report a 40%
enhancement in hazard detection, which must be considered alongside a 25–30% increase in implementation
costs relative to single-method approaches. This trade-off is advantageous for high-value, complex projects,
while conventional methods may be adequate for routine applications.
Real-Time Risk Assessment Methods
Only 14% (n=26) of the analysed research concentrated on real-time or near-real-time methodologies,
underscoring a significant gap in the literature that starkly contrasts with the swift technological advancements
in other fields and suggests construction-specific challenges necessitating investigation (Pereira et al., 2026).
Page 192
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
In real-time investigations, sensor-based monitoring constituted 55% of applications, encompassing wearable
devices and environmental sensing (Awolusi et al., 2020), whereas computer vision techniques accounted for
32% of real-time research (Fang et al., 2020), and IoT platforms that integrate various data sources were present
in 27% of studies (Ding et al., 2020).
According to Pereira et al. (2026), the potential impact is considerable, noting that trial implementations
decreased occurrences by 35–50%, suggesting that transitioning from periodic to continuous monitoring could
markedly improve safety outcomes. Notwithstanding this encouraging promise, significant obstacles impede
widespread implementation.
Data quality challenges in harsh construction environments, such as dust, vibration, temperature fluctuations,
and intermittent connectivity, undermine system reliability (Awolusi et al., 2020). Privacy issues related to
ongoing employee monitoring present ethical challenges that demand thorough examination (Hwang et al.,
2023), while management's ability to respond to real-time alerts requires substantial changes in supervisory
practices, including training for supervisors on how to effectively interpret and act on these alerts.
Furthermore, real-time risk assessment technologies are still in the nascent stages of maturity: sensor durability
in construction settings is at Technology Readiness Level 6-7, computer vision accuracy under varying
conditions is at TRL 5-6, and integrated platform reliability is at TRL 4-5, all necessitating additional
development prior to widespread implementation.
CONCLUSION
This systematic review of 187 studies published from 2020 to 2026 offers a thorough examination of
contemporary risk assessment methodologies in construction projects. The findings indicate a research
environment marked by methodological diversity, technological progress, and ongoing deficiencies that
necessitate academic focus. Quantitative methods predominate in contemporary research (47%), indicating a
focus on objective, data-driven evaluation.
Nevertheless, qualitative (32%) and mixed-methods (21%) approaches remain crucial for elucidating
organisational and human factors. The transition from conventional procedures (31%) to advanced methods such
as AI and machine learning (37%) indicates significant revolutionary potential; nevertheless, the use of hybrid
(18%) and real-time (14%) assessments remains constrained.
A comparative analysis indicates clear trade-offs: traditional methods ensure accessibility and regulatory
approval but lack predictive power; advanced methods deliver enhanced accuracy yet encounter implementation
obstacles such as data demands and interpretability issues; hybrid approaches maximise comprehensiveness at
the expense of heightened complexity; real-time techniques offer potential safety enhancements but necessitate
significant infrastructure investment and raise privacy concerns.
Ongoing deficiencies encompass insufficient focus on post-construction variables (65% of research), inadequate
processes for knowledge accumulation (22%), lack of empirical validation for sophisticated methodologies
(18%), under-representation of situations in underdeveloped countries (24%), and disjointed technological
integration. Addressing these deficiencies necessitates lifecycle assessment frameworks, the development of
learning systems, collaborations for practitioner-researcher validation, context-sensitive technique adaptation,
and integrated BIM-IoT-AI platforms.
REFERENCES
1. Adebayo, Y., Udoh, P., Kamudyariwa, X. B., & Osobajo, O. A. (2025). Artificial intelligence in
construction project management: A structured literature review of its evolution in application and future
trends. Digital, 5(3), 26.
2. Afzal, F., Yunfei, S., Nazir, M., & Bhatti, S. M. (2021). A review of artificial intelligence based risk
assessment methods for capturing complexity-risk interdependencies: Cost overrun in construction
projects. International Journal of Managing Projects in Business, 14(2), 300-328.
Page 193
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
3. Akhtar, M., Mufti, N. A., Mubin, S., Saleem, M. Q., Zahoor, S., & Ullah, S. (2023). Identification of
various execution modes and their respective risks for public–private partnership (PPP) infrastructure
projects. Buildings, 13(8), 1889.
4. Akinosho, T. D., Oyedele, L. O., Bilal, M., Ajayi, A. O., Delgado, M. D., Akinade, O. O., & Ahmed, A. A.
(2020). Deep learning in the construction industry: A review of present status and future innovations.
Journal of Building Engineering, 32, 101827.
5. Akinade, Olugbenga, et al. "Design for deconstruction using a circular economy approach: barriers and
strategies for improvement." Production planning & control 31.10 (2020): 829-840.
6. Albasyouni, W., Kamara, J., & Heidrich, O. (2025). Key challenges and opportunities to improve risk
assessments in the construction industry. Buildings, 15(11), 1832.
7. Aljassmi, H., & Han, S. (2013). Analysis of causes of construction defects using fault trees and risk
importance measures. Journal of Construction Engineering and Management, 139(7), 870-880.
8. Almashhour, R., Al-Mhdawi, M. K. S., Daghfous, A., Qazi, A., & Ojiako, U. (2025). Traditional to
sustainable risk management in the construction industry: a systematic literature review. International
Journal of Managing Projects in Business, 18(3), 528-565.
9. Almatared, M., Liu, H., Abudayyeh, O., Hakim, O., & Sulaiman, M. (2023). Digital-twin-based fire safety
management framework for smart buildings. Buildings, 14(1), 4.
10. Awolusi, I., Marks, E., & Hallowell, M. (2018). Wearable technology for personalized construction safety
monitoring and trending: Review of applicable devices. Automation in construction, 85, 96-106.
11. Bao, Y., Pei, C., Mou, Y., Li, M., & Cheng, X. (2025). A convolutional neural network–long short-term
memory (CNN–LSTM)–Attention model based on wavelet transform for predicting non-stationary wind
pressure coefficients on the surface of terminal glass curtain wall. Science Progress, 108(3),
00368504251366365.
12. Bilal, M., Oyedele, L. O., Qadir, J., Munir, K., Ajayi, S. O., Akinade, O. O., ... & Pasha, M. (2016). Big
Data in the construction industry: A review of present status, opportunities, and future trends. Advanced
engineering informatics, 30(3), 500-521.
13. Celik, T., Kamali, M., & Arayici, Y. (2023). Interval type-2 fuzzy AHP for international construction
project risk assessment. Expert Systems with Applications, 215, 119342.
14. Chabalala, N., Weaich, M., Simbanegavi, P., Ndlovu, P., & Gethe, F. (2024). The ripple effects of covid-
19 on south africa's building and construction industry: Workforce reduction and project delays. Available
at SSRN 4897147.
15. Chen, C., & Lee, J. (2023). Integrated FMEA-fuzzy TOPSIS approach for construction supply chain risk
assessment. International Journal of Construction Management, 23(4), 621-635.
16. Chen, J., Zhang, R., & Wang, X. (2024). Computer vision-based hazard identification for crane operations
on construction sites. Automation in Construction, 158, 105201.
17. Chen, H. (2024). Enhancing Occupational Safety through Vision-based Integrative Technologies in the
Construction and Building Industry (Doctoral dissertation, RMIT University).
18. Choi, J., Lee, S., & Kim, H. (2023). Hybrid risk assessment system combining case-based reasoning and
neural networks for construction delay prediction. Journal of Construction Engineering and Management,
149(5), 04023012.
19. Darko, A., Chan, A. P. C., Ameyaw, E. E., Owusu, E. K., Pärn, E., & Edwards, D. J. (2019). Review of
application of analytic hierarchy process (AHP) in construction. International journal of construction
management, 19(5), 436-452.
20. Dikmen, I., Eken, G., Erol, H., & Birgonul, M. T. (2025). Automated construction contract analysis for
risk and responsibility assessment using natural language processing and machine learning. Computers in
Industry, 166, 104251.
21. El-Sayegh, S. M., Manjikian, S., Ibrahim, A., Abouelyousr, A., & Jabbour, R. (2021). Risk identification
and assessment in sustainable construction projects in the UAE. International Journal of Construction
Management, 21(4), 327-336.
22. Fang, Q., Li, H., Luo, X., Ding, L., Luo, H., & Rose, T. (2020). Computer vision-based PPE compliance
detection for construction safety. Journal of Building Engineering, 32, 101524.
23. Fernández-Muñiz, B., Montes-Peón, J. M., & Vázquez-Ordás, C. J. (2024). Risk assessment tools in small
and medium construction enterprises. Safety Science, 172, 106401.
Page 194
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
24. Getuli, V., Capone, P., & Bruttini, A. (2024). BIM-based occupational health risk identification during
construction. Automation in Construction, 158, 105189.
25. Han, S., & Lee, S. (2022). Vision-based detection of unsafe ladder use in construction sites. Journal of
Computing in Civil Engineering, 36(3), 04022004.
26. Hassan, M. U., Akram, M., & Ullah, F. (2022). Application of FMEA in high-rise building construction:
A case study approach. International Journal of Building Pathology and Adaptation, 40(2), 245-263.
27. Hwang, S., Park, J., & Lee, H. (2023). Smart helmet with physiological monitoring for construction worker
fatigue detection. Sensors, 23(4), 1876.
28. Jafari Ramiani, A., Sarvari, H., Chan, D. W., Nassereddine, H., & Lotfata, A. (2024). Critical success
factors for private sector participation in accomplishing abandoned public sports facilities projects in Iran.
International Journal of Construction Management, 24(6), 586-600.
29. Kabir, S., Papadopoulos, Y., & Agha, M. (2020). Fault tree analysis for construction safety management.
Process Safety and Environmental Protection, 142, 211-222.
30. Kabir, S., Talebi, S., & Papadopoulos, Y. (2022). Bayesian network-based risk assessment for pipeline
construction projects. Reliability Engineering & System Safety, 218, 108142.
31. Kang, H., Sung, S., Hong, J., Jung, S., Hong, T., Park, H. S., & Lee, D. E. (2021). Development of a real-
time automated monitoring system for managing the hazardous environmental pollutants at the
construction site. Journal of Hazardous Materials, 402, 123483.
32. Kim, J., Kim, Y., & Choi, J. (2023). Predictive analytics for construction cost performance using historical
project data. Journal of Management in Engineering, 39(2), 04022068.
33. Kim, E. S., & Choi, S. K. (2013). Failure analysis of connecting bolts in collapsed tower crane. Fatigue &
Fracture of Engineering Materials & Structures, 36(3), 228-241.
34. Kochovski, P., & Stankovski, V. (2018). Supporting smart construction with dependable edge computing
infrastructures and applications. Automation in Construction, 85, 182-192.
35. Kumi, L., Jeong, J., & Jeong, J. (2024). Systematic review of quantitative risk quantification methods in
construction accidents. Buildings, 14(10), 3306.
36. Lee, J., Park, Y., & Kim, K. (2024). International construction project risk assessment using machine
learning and macroeconomic indicators. Journal of Construction Engineering and Management, 150(2),
04023165.
37. Li, H., Yang, X., Wang, F., Rose, T., Chan, G., & Dong, S. (2021). Real-time proximity warning system
for construction equipment using ultra-wideband technology. Automation in Construction, 132, 103956.
38. Lin, S. S., Shen, S. L., Zhou, A., & Xu, Y. S. (2021). Risk assessment and management of excavation
system based on fuzzy set theory and machine learning methods. Automation in Construction, 122,
103490.
39. Liu, J., Wang, Y., & Zhang, L. (2023). Fuzzy FMEA with cost-benefit integration for construction risk
prioritization. International Journal of Construction Management, 23(8), 1356-1370.
40. Liu, R., Liu, H., & Zhao, Y. (2021). Fuzzy fault tree analysis for bridge construction risk assessment.
Structures, 34, 4521-4533.
41. Liu, Y., Zhang, P., & Wang, X. (2024). Limitations of traditional risk assessment methods in construction:
A critical review. Journal of Civil Engineering and Management, 30(1), 45-62.
42. Li, Y., & Wang, X. (2018). Risk assessment for public–private partnership projects: using a fuzzy analytic
hierarchical process method and expert opinion in China. Journal of Risk Research, 21(8), 952-973.
43. Medaa, E., Shirzadi Javid, A. A., & Malekitabar, H. (2025). Evolution of Risk Analysis Approaches in
Construction Disasters: A Systematic Review of Construction Accidents from 2010 to 2025. Buildings,
15(20), 3701.
44. Meyer, M., Grisar, C., & Kuhnert, F. (2011). The impact of biases on simulation-based risk aggregation:
modeling cognitive influences on risk assessment. Journal of Management Control, 22(1), 79-105.
45. Moghadam, M., Alvanchi, A., & Alimohammadi, M. (2023). BIM-based constructability risk assessment
framework. Journal of Information Technology in Construction, 28, 145-162.
46. Moheimani, M. A., Sheikhkhoshkar, M., & Banihashemi, S. (2024). Interval type-2 fuzzy TOPSIS for
sustainable construction risk assessment. Sustainable Cities and Society, 101, 105078.
47. Nasirzadeh, F., Khanzadi, M., & Mir, M. (2023). Dynamic modeling of construction project delays using
system dynamics and Monte Carlo simulation. Engineering, Construction and Architectural Management,
30(4), 1456-1475.
Page 195
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
48. Nguyen, H. D., Do, Q. N. H., & Macchion, L. (2023). Influence of practitioners' characteristics on risk
assessment in Green Building projects in emerging economies: a case of Vietnam. Engineering,
construction and architectural management, 30(2), 833-852.
49. Ojghaz, A. S., Sadeghpour, F., & Bayat, S. (2024). Edge Computing for Real-Time Monitoring Systems
in Construction Sites. In Canadian Society of Civil Engineering Annual Conference (pp. 283-292). Cham:
Springer Nature Switzerland.
50. Pereira, F., González García, M. D. L. N., & Poças Martins, J. (2025). Methods and Techniques for Real
Time Construction Risk Assessment—Systematic Review. Occupational and Environmental Safety and
Health VI: Volume 2: Biomechanics, Occupational Psychosociology and New Trends, 465-480.
51. Qazi, A., & Simsekler, M. C. E. (2021). Risk assessment of construction projects using Monte Carlo
simulation. International journal of managing projects in business, 14(5), 1202-1218.
52. Qian, W., Min, S., Na, L., Mengqi, S., & Jing, G. (2020). Construction risk assessment of green
transformation of old industrial plants based on improved FMEA. In IOP Conference Series: Earth and
Environmental Science (Vol. 531, No. 1, p. 012059). IOP Publishing.
53. Rani, H. A., Syammaun, T., Azzahra, F., Ayob, A., Amlus, M. H., Zakirullah, & Aqsha, M. S. (2024,). Risk
management planning by risk register in building construction project. In IOP Conference Series: Earth
and Environmental Science (Vol. 1303, No. 1, p. 012034). IOP Publishing.
54. Resende, B. A. D., Dedini, F. G., Eckert, J. J., Sigahi, T. F., Pinto, J. D. S., & Anholon, R. (2024). Proposal
of a facilitating methodology for fuzzy FMEA implementation with application in process risk analysis in
the aeronautical sector. International Journal of Quality & Reliability Management, 41(4), 1063-1088.
55. Sanni-Anibire, M. O., Mahmoud, A. S., Hassanain, M. A., & Salami, B. A. (2020). A risk assessment
approach for enhancing construction safety performance. Safety science, 121, 15-29.
56. Serrano-Gomez, L., & Munoz-Hernandez, J. I. (2019). Monte Carlo approach to fuzzy AHP risk analysis
in renewable energy construction projects. PloS one, 14(6), e0215943.
57. Shayboun, M., Kifokeris, D., & Koch, C. (2025). A review of machine learning for analysing accident
reports in the construction industry. Journal of Information Technology in Construction, 30, 439-460.
58. Sony, S., Laventure, S., & Sadhu, A. (2021). Vision-based structural health monitoring: A critical review.
Engineering Structures, 245, 112891.
59. Susanto, T., Nurdiana, A., Wibowo, M. A., & Maromi, M. I. (2026). Literature Review of Occupational
Safety and Health Risk Management in Foundation Work. Rekayasa Sipil, 20(1), 61-67.
60. Ullah, S., Deng, X., Amaechi, C. V., Anbar, D. R., & Ashraf, M. W. (2025). A systematic literature review
on knowledge mapping for project risk management in the construction industry. Frontiers in Built
Environment, 11, 1677904.
61. Wang, J., Zhang, S., & Liu, J. (2022). Fault tree analysis of tower crane collapses in construction.
Engineering Failure Analysis, 138, 106345.
62. Wang, L., Wang, J., & Zhang, P. (2024). BIM-IoT integration framework for dynamic construction risk
assessment. Automation in Construction, 159, 105267.
63. Wang, Y., Chen, C., & Li, H. (2022). Fuzzy FMEA for high-rise building construction risk assessment.
International Journal of Environmental Research and Public Health, 19(8), 4567.
64. Wuni, I. Y., Shen, G. Q., & Saka, A. B. (2023). Computing the severities of critical onsite assembly risk
factors for modular integrated construction projects. Engineering, Construction and Architectural
Management, 30(5), 1864-1882.
65. Wu, S. (2024). Enhancing Construction Workforce Safety through Formulating an Augmented Reality
Digital Twin System (Doctoral dissertation, RMIT University).
66. Yi, W., Chan, A. P. C., & Wang, X. (2022). Environmental sensing network for real-time construction
worker exposure monitoring. Journal of Cleaner Production, 376, 134234.
67. Yussif, A.-M., Taiwo, R., Shakor, P., Han, T., Mohandes, S. R., Antwi Afari, M., Qazi, K., Singh, A. K.,
Christo, M. S., & Shah, M. A. (2025). A comprehensive literature review on risk identification and
assessment in green building construction projects. Cleaner Engineering and Technology, 29, 101089
68. Zavadskas, E. K., Turskis, Z., & Tamošaitiene, J. (2010). Risk assessment of construction projects. Journal
of civil engineering and management, 16(1), 33-46.
69. Zhang, L., Wu, X., & Skibniewski, M. J. (2021). Bayesian network-based risk assessment for tunnel
construction. Tunnelling and Underground Space Technology, 115, 104056.
Page 196
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 202
70. Zhang, L., Wu, X., & Liu, M. (2021). Fuzzy fault tree analysis for construction safety: A review and future
directions. Safety Science, 142, 105387.
71. Zhang, R., Li, H., & Skitmore, M. (2022). BIM-based rule checking for construction safety risk
identification. Automation in Construction, 142, 104512.
72. Zhong, B., Pan, X., Love, P. E. D., & Ding, L. (2022). Deep learning-based prediction of construction
schedule delays. Advanced Engineering Informatics, 54, 101777.
73. Zou, Y., Kiviniemi, A., & Jones, S. W. (2020). BIM-based risk identification framework for construction
projects. Automation in Construction, 118, 103283.