ID 144

AI-Driven Innovations in Cost Estimation and Quality Control

in Sri Lankan Construction Projects
LADCN Wijesinghe#, SD Jayasooriya, and MKM Nilupulee

Faculty of Built Environment and Spatial Sciences (FBESS), General Sir John Kotelawala Defence
University, SriLanka
#[email protected]

Abstract— The integration of artificial intelligence complying with industry regulations. By addressing
(AI) in construction project management has these aspects, construction firms can harness AI's full
significantly enhanced cost estimation and quality potential, leading to more successful and sustainable
control processes. This paper, employing a mixed- project outcomes.
method research design, explores the impact of AI-
driven innovations on these critical aspects of the Keywords— artificial intelligence, cost estimation,
construction industry. Through a comprehensive quality control, quality management, construction
analysis involving surveys and regression analysis, industry, data security
this study investigates the adoption rate, benefits,
challenges, and overall influence of AI on project I. INTRODUCTION
outcomes. Data were collected using stratified random
sampling techniques to ensure a representative sample The construction industry is undergoing a
of quantity surveyors, civil engineers, project transformative shift, driven by the integration of
managers, and quality control engineers. The findings artificial intelligence (AI) technologies. Traditional
reveal a high adoption rate of AI tools among methods in construction, particularly in cost
respondents, resulting in notable improvements in estimation and quality control, often encounter
accuracy, efficiency, and predictive capabilities. The significant challenges such as human error,
positive correlation between AI adoption and project inefficiencies, and delays (Abualdenien et al., 2021).
outcomes, such as cost savings, quality enhancements, These challenges can lead to increased project costs
project timeliness, client satisfaction, and return on andcompromised quality, necessitating the adoption of
investment (ROI) underscores the transformative more advanced and accurate systems. AI, with its
potential of AI technologies. Despite the challenges capabilities inmachine learning, computer vision, and
like high initial costs, lack of skilled personnel, and data analytics,presents an innovative solution to these
data privacy concerns, the advantages of AI, including enduring problems(Chen and Liu, 2020).
increased accuracy and data-driven decision-making
AI-driven innovations in cost estimation have shown
outweigh the drawbacks. The study also highlights
promising results in enhancing accuracy and
AI's versatility, extending its applications to project
efficiency. Machine learning algorithms can analyze
scheduling, risk management, and procurement,
vast amounts of historical data to predict project costs
thereby enhancing communication and collaboration
with greater precision than traditional methods
among stakeholders. The paper concludes with
(Boussabaine, 2013). By automating the quantity
recommendations for the construction companies to
takeoff process, AI tools can significantly reduce the
maximize the benefits of AI adoption, including
time and effort required, thereby minimizing human
investing in AI tools, training personnel, ensuring data
errors and inconsistencies (Son, Kim and Kim, 2019).
security, developing integration strategies, fostering a
Moreover, AI's predictive analytics can help identify
culture of innovation, conducting pilot projects, and
potential cost overruns early in the project lifecycle,
159
allowing for proactive management and decision- B. AI-Driven Cost Estimation
making (Zhang et al., 2020). Traditional cost estimation methods often face
challenges such as human error and inefficiencies. AI-
In parallel, AI is revolutionizing quality control driven cost estimation models, leveraging machine
processes in construction projects. Computer vision learning algorithms, can analyze vast datasets to
technology, for instance, can be employed to automate predict project costs with high accuracy
defect detection andensure that construction standards (Boussabaine, 2013). These models reduce the
are consistently met (Chi et al., 2021). Real-time reliance on manual processes, thereby minimizing
monitoring systems powered by AI can continuously errors and increasing efficiency (Son, Kim, and Kim,
assess construction progress and quality, providing 2019). Studies have shown that AI can significantly
instant feedback to project managers (Wang et al., improve cost estimation accuracy by learning from
2020). This not only enhances the accuracy of historical project data and identifying patterns
inspections but also improves the overall safety and (Russell and Norvig, 2021).
qualityof construction projects.
The integration of AI in these domains offers a dual AI-driven cost estimation tools use historical data and
advantage: it not only streamlines cost estimation machine learning techniques to provide more accurate
processes but also ensures rigorous quality control, andreliable cost forecasts. These tools can adjust for
leading to more successful project outcomes. variables such as material price fluctuations, labor
However, the adoption of AI in the construction costs, and project-specific factors, resulting in more
industry is not without challenges. Issues such as high precise budgeting and financial planning (Zhang et
implementation costs, data privacy concerns, and the al., 2020). By automating routine tasks, these tools
need for skilled personnel to manage AI systems must free up human estimators to focus on strategic
be addressed (Marques et al., 2020). Despite these decision-making and risk management.
challenges, the potential benefits of AI-driven
innovations in cost estimation and quality control are C. AI in Quantity Takeoff
substantial and warrant further exploration and Quantity takeoff is a crucial step in the cost estimation
investment. process, where the quantities of materials needed for
a project are determined. AI-powered tools can
II. LITERATURE REVIEW automate the quantity takeoff process, extracting
measurements directly from BIM models and 2D
A. AI in Construction drawings. This automation not only speeds up the
The construction industry is increasingly adopting AI process but also enhances accuracy by reducing
to enhance various processes, with a significant focus human error (Zhang et al., 2020). As Son, Kim, and
on cost estimation and quality control. AI Kim (2019) note, AI can streamline the takeoff process,
technologies, such as machine learning and computer allowing quantity surveyors to focus on more
vision, are instrumental in transforming these strategic tasks. AI-enhanced quantity takeoff systems
traditional practices. According to Abualdenien et al. utilize advanced image recognition and machine
(2021), AI applications in construction can lead to learning algorithms to interpret complex architectural
more accurate cost estimations and improved quality drawings and BIM data. These systems can quickly
control, ultimately resulting in better project identify and quantify materials, ensuring precise
outcomes. The integration of AI in construction calculations and reducing discrepancies (Khan and
includes applications such as automated design, Gotman, 2021). The integration ofAI in this area also
project planning, and predictive maintenance. For supports real-time updates and adjustments,
instance, AI-driven project management tools can accommodating design changes and ensuring
optimize resource allocation, schedule tasks, and continuous accuracy throughout the project lifecycle.
predict potential delays, thereby enhancing overall
project efficiency (Chen and Liu, 2020). Moreover, AI D. Predictive Analytics in Construction
is used in smart construction sites where sensors and Predictive analytics, a subset of AI, involves using
IoT devices collect real-time data, providing insights statistical techniques to analyze current and historical
that drive decision-making and improve safety data to make predictions about future events. In the
standards (Wang et al., 2020). context of construction, predictive analytics can be
used to forecast cost overruns, project delays, and
potential quality issues (Shmueli and Koppius, 2019).
By providing early warnings, predictive analytics
enables project managers to take proactive measures,
160
thus reducing risks and improving project outcomes that embraces technological advancements are crucial
(Chen and Liu, 2020). steps. Additionally, developing industry-specific AI
solutions tailored to the unique needs and constraints
Predictive analytics tools leverage vast amounts of of the construction sector can facilitate smoother
projectdata to identify patterns and correlations that transitions and broader acceptance (Russell and
might not be immediately apparent. By applying Norvig, 2021).
machine learning algorithms, these tools can forecast
potential issues such as equipment failures, supply III.RESEARCH METHODOLOGY
chain disruptions, and labor shortages, allowing
project managers to develop contingency plans and This chapter outlines the methodological approach
optimize resource allocation (Marques et al., 2020). undertaken to investigate the role of AI-driven
This proactive approach enhances project resilience innovations in cost estimation and quality control in
and ensures timely completion. Sri Lankan large scale construction projects. The
study aims to explore how AI can enhance the
E. AI-Enhanced Quality Control efficiency and accuracy of these processes,
Quality control is essential to ensure that construction contributing to better project outcomes. The
projects meet the specified standards and methodology is designed to ensure a comprehensive
requirements. AI technologies, particularly computer and rigorous analysis, incorporating both qualitative
vision, can enhance quality control processes by andquantitative data collection and analysis methods.
automating defect detection and real-time monitoring.
Computer vision systems can analyze images and A. Data Collection Methods
videos from construction sites to identify defects and The study employs a mixed-methods approach to
deviations from the design (Khan and Gotman, 2021). gather a broad range of data, combining qualitative
This automation improves inspection accuracy and insights with quantitative analysis to provide a holistic
efficiency, leading to higher quality construction understanding ofthe impact of AI in construction.
projects (Sundar, Prabhu, and Rahman, 2022). AI- i. Literature Review: A systematic literature review
enhanced quality control systems use computer vision was conducted to identify existing research, theories,
and deep learning algorithms to inspect construction and findings related to AI applications in cost
sites continuously. These systems can detect estimation and quality control. Sources included
structural anomalies, material defects, and adherence academic journals, industry reports, and conference
to safety protocols, providing instant feedback and papers from databases such as IEEE Xplore,
enabling immediate corrective actions (Chi et al., ScienceDirect, and Google Scholar.
2021). The ability to monitor quality in real-time
ensures that construction standards are consistently ii. Surveys: A structured questionnaire was
met, reducing rework and enhancing project integrity. designed and distributed to a sample of Sri Lankan
construction industryprofessionals who are engage in
F. Challenges and Barriers to AI Adoption large scale projects, including quantity surveyors,
Despite the potential benefits, the adoption of AI in civil engineers, project managers, and quality control
the construction industry faces several challenges. engineers. The survey aimed to collect data on the
High implementation costs, data privacy concerns, current use of AI technologies, perceived benefits,
and the need for skilled personnel to manage AI challenges, and future potential in the industry.
systems are significantbarriers (Marques et al., 2020).
Additionally, there is oftenresistance to change within iii. Interviews: Semi-structured interviews were
the industry, which can hinder the adoption of new conducted with key stakeholders in the Sri Lankan
technologies. Addressing these challenges requires a construction sector. These interviews provided in-
strategic approach, including investment in training depth insights into the practical applications of AI,
and education, development of industry-specific AI real-world experiences, and expert opinions on the
tools, and fostering a culture of innovation future of AI in cost estimation and quality control.
(Abualdenien et al., 2021). B. Data Analysis Methods
The data collected from surveys and interviews were
Overcoming the barriers to AI adoption involves analyzed using a combination of qualitative and
addressing the high initial costs through strategic quantitative techniques:
investments and demonstrating the long-term benefits
of AI integration. Training and upskilling the i. Descriptive Statistics: Survey data were analyzed
workforce to handle AI tools and fostering a culture using descriptive statistics to summarize the current
161
use andperceived benefits of AI in construction. This percentage of AI adoption is as follows:
included frequency distributions and means.
ii. Thematic Analysis: Qualitative data from
interviews and open-ended survey responses were
analyzed using thematic analysis. Key themes and
patterns were identified to understand the experiences
and perspectives of industry professionals regarding
AI adoption.
iii. Comparative Analysis: The responses from
surveys and interviews were compared to identify
commonalities and differences in AI implementation
and its impact on cost estimation and quality control. Figure 2: AI Adoption in Construction

iv. Regression Analysis: To assess the B. Thematic Analysis:

relationship between AI adoption and project From the qualitative data collected through
outcomes, regression analysis was performed. This interviews,several key themes emerged regarding the
quantitative method helped to determine the extent to benefits of AI inconstruction:
which AI technologies influence cost savings and
quality improvements.

IV. RESEARCH FINDINGS

The survey and interviews conducted targeted

professionals within the Sri Lankan large scale
construction industry, including quantity surveyors,
civil engineers, project managers, and quality control
engineers. A total of 65 respondents participated in
the survey, providing a diverse range of insights into
the application of AI in cost estimation and quality
control.
Figure 3: Perceived Benefits AI Adoption in Construction

Respondents emphasized that AI significantly

enhances cost estimation accuracy by 80% and boosts
efficiency by 75%. Key themes identified include
predictive capabilities,data-driven decisions, resource
optimization, and enhanced collaboration. They
highlighted AI's capacity to swiftly and precisely
analyze vast datasets, particularly beneficial in
complex projects with stringent timelines.

Figure 1: Respondents’ Details The interviews also revealed several challenges faced
by the construction industry in adopting AI
A. Descriptive Statistics: technologies:
The survey results revealed varying levels of AI
adoption across different functions within The adoption of AI tools and technologies faces
construction projects. 70% of respondents reported significantbarriers in many companies, primarily due
using AI tools for cost estimation. 50% of respondents to the high implementation costs. Additionally, there
indicated the use of AI for quality control processes is a shortage of professionals skilled in operating and
and 30% of respondents mentioned the use of AI in maintaining AI systems, noted by 55% of respondents.
project scheduling and risk management. The mean Concerns about data privacy and security pose further
162
obstacles, cited by 50%. Integrating AI with existing outcomes such as cost savings and quality
systems and workflows is often complex and time- improvements. The analysis indicated a positive
consuming. Resistance from employeesand managers correlation between the extent of AI adoption and
to adopting new technologies and changing improved project outcomes.
established processes is also a common challenge.
Moreover, ensuring AI solutions comply with
industryregulations and standards presents additional
complexities.

Figure 5: Relation between AI adoption and Project

Outcomes

Figure 4: Challenges in AI Adoption in Construction Projects that implemented AI technologies reported

anaverage cost saving of 15%. AI adoption led to a
The adoption of AI tools and technologies faces 20% reduction in defects and rework, and contributed
significantbarriers in many companies, primarily due to a 10% improvement in project timelines.
to the high implementation costs. Additionally, there Additionally, projects with AI saw a 25% increase in
is a shortage of professionals skilled in operating and client satisfaction. Companies experienced a 30%
maintaining AI systems, noted by 55% of respondents. ROI on AI investments within the first year.
Concerns about data privacy and security pose further
obstacles, cited by 50%. Integrating AI with existing
systems and workflows is often complex and time- V. CONCLUSION AND RECOMMENDATIONS
consuming. Resistance from employeesand managers
to adopting new technologies and changing A. Conclusion
established processes is also a common challenge. The research findings underscore the transformative
Moreover, ensuring AI solutions comply with potential of AI-driven innovations in cost estimation
industryregulations and standards presents additional and quality control within the Sri Lankan large scale
complexities. construction projects. The adoption of AI technologies
has demonstrated significant improvements in
C. Comparative Analysis accuracy, efficiency, and predictive capabilities,
A comparative analysis of survey and interview contributing to enhanced project outcomes. A
responses revealed some commonalities and substantial proportion of respondents, particularly
differences in the perceptions of AI adoption: quantity surveyors and projectmanagers, have adopted
AI tools for cost estimation and quality control,
Commonalities: Both survey and interview highlighting the industry's readiness to embrace
respondents agreed that AI significantly improves advanced technologies.
accuracy andefficiency in cost estimation and quality
control. The integration of AI has led to notable improvements
in cost savings, quality enhancements, project
Differences: Survey respondents were more timeliness, client satisfaction, and return on
optimistic about the potential of AI in quality control, investment (ROI). The regression analysis established
whereas interviewees expressed more concerns about a positive correlation between AI adoption and these
the challengesand barriers to adoption. project outcomes, affirming the value of AI in
D. Regression Analysis construction project management. Despite the
Regression analysis was conducted to assess the challenges associated with AI adoption, such as high
relationship between AI adoption and project initial costs, lack of skilled personnel, and data privacy
concerns, the perceived benefits such as increased
163
accuracy, efficiency, and data-driven decision-making identify areas for improvement and optimization.
far outweigh the drawbacks. Thematic analysis of
qualitative data supports this conclusion, as By implementing these recommendations,
respondents consistently emphasized the significant construction companies can effectively harness the
advantages AI offers. power of AI to enhance cost estimation and quality
control, ultimately leading to more successful and
AI is not limited to cost estimation and quality control sustainable project outcomes.
but extends to other critical functions like project
scheduling, risk management, and procurement. This REFERENCES
versatility indicates that AI has broad applicability
across various aspects of construction projects. Abualdenien, J., Hosseini, M. R., Banihashemi, S., and
Furthermore, AI technologies facilitate better Golizadeh,
communication and collaboration among project
stakeholders, contributing to more cohesive and H. (2021). 'Artificial intelligence applications in
efficient project execution. construction: A systematic review and future
directions', Automation in Construction, 123.
B. Recommendations
Based on these findings, several recommendations are Boussabaine, H. A. (2013). Cost Planning of PFI and
proposed to maximize the benefits of AI adoption in PPP Building Projects. 1st ed. Routledge.
construction projects and address the associated Chen, J., and Liu, R. (2020). 'Application of AI in the
challenges. Construction companies should prioritize Construction Industry: Current Trends and Future
investing in AI tools and technologies. Although the Prospects', Journal of Construction Engineering and
initial costs may be high, the long-term benefits in Management.
terms of cost savings, quality improvements, and ROI
justify the investment. To address the shortage of Chi, H. L., Kang, S. C., Wang, X., and Lin, J. R. (2021).
skilled personnel, companies should invest in training 'Research trends and opportunities of augmented
and development programs. This includes upskilling reality applications in architecture, engineering, and
current employees and hiring new talent with expertise construction', Automation in Construction.
in AI and data analytics.
Jordan, M. I., and Mitchell, T. M. (2015). 'Machine
Implementing robust data security measures to protect learning: Trends, perspectives, and prospects', Science,
sensitive information is crucial. This includes 349(6245), pp. 255- 260.
adopting industry best practices for data encryption,
access control, and compliance with relevant Khan, S., and Gotman, J. (2021). 'Automating defect
regulations. Developing strategies for seamlessly detection in construction using computer vision', IEEE
integrating AI technologies with existing systems and Transactions on Automation Science and Engineering,
workflows is also essential. This may involve 18(3), pp. 827-839.
collaborating with technology providers and
Marques, G., Ramos, R. F., Pitarma, R., and
consultants to ensure smooth implementation and
Anastasios, A. (2020). 'Artificial intelligence in the
minimize disruption. Addressing resistance to change
construction industry: a review', Construction
by fostering a culture of innovation and continuous
Management and Economics, 38(10), pp. 839-854.
improvement is key. This can be achieved through
effective communication, leadership support, and Russell, S. J., and Norvig, P. (2021). Artificial
involving employees in the AIadoption process. Intelligence: A Modern Approach. 4th ed. Pearson.
Conducting pilot projects and documenting case Shmueli, G., and Koppius, O. R. (2019). 'Predictive
studies to demonstrate the practical benefits of AI in analytics in information systems research', MIS
construction canhelp build confidence and support for Quarterly, 35(3), pp. 553-572.
wider AI adoption within the industry. Ensuring that AI
solutions comply with industry regulations and Smith, P. (2019). 'Cost management in construction
standards is also vital. Engaging with regulatory projects: An overview of the profession and its
bodies to stay informed about evolving requirements practices', International Journal of Project
and best practices will help in this regard. Establishing Management, 37(2), pp. 273-284.
mechanisms for continuous monitoring and evaluation
of AI technologies is important for regularly assessing Son, H., Kim, C., and Kim, C. (2019). 'Early detection
their performance and impact on project outcomes to of construction cost overruns using machine learning',
Journal of Construction Engineering and
164
Management, 145(3). Dispute resolution.
Sundar, K., Prabhu, P., and Rahman, M. (2022).
'Quality control in construction: The role of AI and
machine learning', Construction Technology Today,
42(5), pp. 67-74.
Wang, Q., Yuan, W., Xiang, T., and Li, Q. (2020).
'Real-time quality monitoring and control for
construction projects using artificial intelligence',
Journal of Computing in Civil Engineering, 34(6).
Zhang, J., Wu, Y., Liu, G., and Zhang, J. (2020).
'Machine learning-based approach for cost estimation
in construction projects using historical data', Journal
of Civil Engineering and Management, 26(8), pp. 759-
769.
AUTHOR BIOGRAPHIES

Mr. LADCN Wijesinghe, is a

graduate of the Department of
Quantity Surveying the Faculty
Built of Environment & Spatial
Sciences, General Sir John
Kotelawala Defence University
(KDU). His research interests
include Sustainable Construction, Cost Management,
Construction Quality Management and Building
information modeling.

Dr. Sanjaya Jayasooriya, is a

Senior Lecturer attached to the
Department of Industrial Quality
Management of the Faculty of
Built Environment & Spatial
Sciences. He holds a doctoral
degree in Management the
University from of Jayawardenapura. He has
authored a number of research articles in his field of
study and his research interests include Quality
Management, Intellectual Capital, Human Capital,
Financial and Management Accounting, Finance
management, HRM, Taxation and Business Law.

Miss MKM Nilupulee, is an

undergraduate of General Sir
John Kotelawala Defense
University, following a BSc
(Hons) Quantity Surveying
degree. Her research interests
include BIM, Sustainability, and

165