Abstract

Artificial Intelligence (AI) is increasingly embedded in civil engineering work, yet empirical evidence on how practicing engineers actually use and experience AI remains limited, particularly in emergingeconomy settings [1]. This study examined AI utilization in civil engineering practice in Quirino Province, Philippines, and developed a comprehensive professional organizational support program based on the findings. A descriptivecomparative crosssectional survey was conducted among 150 practicing civil engineers using a structured 24item questionnaire covering four dimensions: technological, organizational, knowledge and skills, and ethical, cultural, and regulatory aspects. Descriptive statistics, independentsamples ttests, and oneway analysis of variance with post hoc testing were used to analyze the data. The overall mean score for AIrelated conditions was 2.89 on a fourpoint scale, indicating moderately favorable conditions. Knowledge and skills obtained the highest mean (3.03), followed by technological (2.96) and organizational aspects (2.88), while ethical, cultural, and regulatory aspects obtained the lowest mean (2.69), although they remained in the positive range. Significant differences were observed across demographic and sectoral groups, with younger engineers, those with 10 years or less of experience, and privatesector practitioners reporting more favorable AIrelated conditions than older, more experienced, and governmentsector counterparts; female respondents also perceived more favorable organizational support than male respondents. The findings indicate an emerging but not yet mature level of AI readiness in civil engineering practice and support the need for multidimensional interventions. A comprehensive professional organizational support program is proposed to strengthen AI adoption through capacitybuilding, stronger institutional backing, improved digital infrastructure, and clearer ethical and regulatory guidance, which may be useful for similar developingcountry contexts [2,3].

1. Introduction

Artificial intelligence has become an increasingly important enabling technology in civil engineering because it can process large volumes of data, recognize patterns, support prediction, and automate complex tasks that would otherwise demand substantial time and human effort ⁴. As civil engineering responds to rapid urbanization, aging infrastructure, and rising expectations for sustainability and efficiency, AI is being considered for a broad range of applications including structural health monitoring, transportation systems, geotechnical analysis, construction planning, cost estimation, construction monitoring, building management, and infrastructure maintenance ¹.

Recent review studies show that AI in the construction and civil infrastructure sectors is no longer confined to laboratory or highly specialized tools; it increasingly supports mainstream project functions such as cost estimation, schedule prediction, safety monitoring, risk assessment, and predictive maintenance ^{5, 6}. AIbased computer vision, machine learning, and reinforcement learning methods can detect unsafe conditions, forecast project delays, optimize resource allocation, and identify defects at early stages, thereby augmenting engineers’ ability to manage complex project environments ^{7, 8}.

Despite these opportunities, AI utilization in civil engineering is not yet widespread or uniform ⁹. Implementation is slowed by poor data quality, fragmented information systems, lack of interoperability with existing software, high startup costs, and varying levels of organizational and workforce readiness ¹⁰. Concerns about explainability, trust, job displacement, and accountability especially for “blackbox” AI models further complicate adoption in safetycritical civil engineering decisions ^{11, 12}.

Workforce capability and organizational conditions emerge as central determinants of successful AI utilization. Engineers may be aware of AI conceptually but lack the practical skills needed to apply it in design, analysis, and project management, while organizations may have limited strategies, budgets, or support mechanisms for integrating AI into routine workflows ^{7, 13}. Ethical, cultural, and regulatory issues such as data privacy, professional liability, and clarity of rules also influence how comfortable practitioners feel using AI in their work ^{12, 14}.

In the Philippine context, and particularly in provincial settings like Quirino, empirical evidence on AI utilization in civil engineering practice remains limited ^{2, 3} Existing studies often focus on technical feasibility or isolated use cases rather than on the broader conditions that shape AI adoption across practitioners and organizations ^{9, 15} Examining the current state of AI utilization among civil engineers in such contexts can generate evidence needed to design relevant capacitybuilding programs and policies, especially where digital transformation is uneven ^{1, 16}.

This study therefore investigates the utilization of AI in civil engineering practice in Quirino Province across four dimensions technological, organizational, knowledge and skills, and ethical, cultural, and regulatory aspects and analyzes differences by sex, age, years of work experience, and work affiliation. The results serve as basis for a proposed comprehensive professional organizational support program intended to strengthen AI utilization in civil engineering

The study aimed to evaluate the utilization of artificial intelligence in civil engineering practice as basis for a comprehensive professional organizational support program. Specifically, it sought to:

1. Describe the demographic profile of civil engineer respondents in terms of age, sex, years of work experience, and work affiliation.

2. Assess the extent of AI utilization in civil engineering in terms of technological, organizational, knowledge and skills, and ethical, cultural, and regulatory aspects.

3. Determine whether significant differences exist in AI utilization when respondents are grouped according to their demographic profile.

4. Propose a comprehensive professional organizational support program to strengthen AI utilization in civil engineering practice.

The study is informed by several complementary theories. Decision Theory explains how professionals choose among alternatives under risk and uncertainty, emphasizing tradeoffs among cost, time, quality, and safety in project environments ^{17, 18}. In such settings, AI serves as a decision aid that processes large volumes of project data to reduce uncertainty and improve judgment ^{7, 19}. This links Turner & Cochrane’s work with decisions in projects where goals and methods are not fully defined, which is exactly where AI‑supported decision tools become relevant ²⁰.

Systems Theory views civil engineering projects as complex systems composed of interdependent components including people, materials, equipment, procedures, and information flows where changes in one part can affect others ¹⁹. AIbased monitoring, optimization, and simulation tools support this systems view by integrating data from multiple subsystems to reveal how local decisions affect overall project performance ¹⁰.

AIenabled Decision Support Systems (DSS) link AI to project management by describing computerbased systems that help professionals analyze data, explore alternatives, and make betterinformed ²¹. When AI methods such as machine learning and predictive analytics are embedded in DSS, they enhance capabilities for risk assessment, cost forecasting, progress monitoring, and resource allocation in construction projects ^{16, 21}.

The Technology Acceptance Model (TAM) posits that perceived usefulness and perceived ease of use shape individual acceptance of new technologies ^{22, 23}. For civil engineers, perceived usefulness relates to whether AI helps deliver projects more efficiently, safely, and at higher quality, while perceived ease of use reflects the compatibility and learnability of AI tools ¹³. Diffusion of Innovations theory explains how new technologies spread across social systems over time, highlighting adopter categories and attributes such as relative advantage and compatibility ²⁴.

Together, these theories support the study’s focus on technological, organizational, knowledge and skills, and ethical, cultural, and regulatory aspects of AI utilization and guide the design of the proposed support program ¹².

2. Methods

A descriptivecomparative crosssectional survey design was used. The descriptive component sought to determine the extent of AI utilization across four dimensions, whereas the comparative component examined differences when respondents were grouped by sex, age, work affiliation, and years of work experience. The crosssectional approach allowed data to be collected at a single point in time without manipulating variables, appropriate given the focus on existing perceptions and conditions rather than causal relationships ²⁵.

The study was conducted in Quirino Province, Philippines, among practicing civil engineers who are members of the Philippine Institute of Civil Engineers (PICE), Quirino Chapter. Participants included professionals from both government agencies and private construction and consulting firms involved in planning, design, construction, and infrastructure management ^{12, 13}.

Sample size was determined using Raosoft and Slovin’s formula, with a 5% margin of error, 95% confidence level, and 50% response distribution, yielding a minimum of approximately 141 respondents. The final sample consisted of 150 civil engineers, exceeding the minimum requirement and enhancing statistical power. A stratified random sampling method was employed to ensure representation across age groups, work affiliation (government vs. private), and years of work experience.

Data were collected using a selfdeveloped structured questionnaire comprising two parts. Part I gathered demographic data (age, sex, years of experience, work affiliation), and Part II consisted of 24 statements measuring four dimensions: technological aspects (7 items), organizational aspects (7 items), knowledge and skills aspects (5 items), and ethical, cultural, and regulatory aspects (5 items). Items were rated on a four-point Likert scale from 1 (Strongly Disagree) to 4 (Strongly Agree), and weighted means were interpreted using defined cut-offs.

The instrument underwent expert validation to assess content relevance and clarity, yielding content validity indices between 0.70 and 0.85, and a pilot test established acceptable internal consistency reliability. These procedures aligned with recommended practices for survey development in engineering management research ^{9 26}.

After pilot testing and refinement, the questionnaire was distributed electronically through professional networks and organizational channels, accompanied by an informed consent statement. Completed responses were screened, coded, and entered into statistical software for analysis.

Frequency and percentage were used to describe the demographic profile of respondents. Weighted means were computed to assess the extent of AI utilization across the four dimensions. Independentsamples ttests were used to compare AI utilization by sex and work affiliation, and oneway analysis of variance with post hoc tests was used to examine differences by age and years of experience. Statistical significance was evaluated at the 0.05 level.

The study followed ethical standards for survey research. Participation was voluntary, respondents were informed of the study’s purpose and their rights, and confidentiality and anonymity were assured. No personally identifiable information was reported, and the data were used solely for academic purposes ²⁷.

3. Results

Table 1 shows that half of the respondents (50.00%) were 30 years old and below, while 28.67% were aged 31–40 years, indicating that the sample is composed largely of younger professionals. Most respondents (64.00%) had 10 years or less of work experience, with only 2.00% reporting more than 40 years in service. Male respondents comprised 76.00% of the sample and female respondents 24.00%. Slightly more than half (54.00%) worked in government, and 46.00% worked in the private sector.

Table 2–Table 5 present the itemlevel means for each dimension, and Table 6 summarizes the dimensionlevel means.

The grand mean of 2.89 indicates that respondents perceive AIrelated conditions as moderately favorable overall. Knowledge and skills emerged as the most favorable dimension, followed by technological and organizational aspects, while ethical, cultural, and regulatory aspects were relatively weaker though still positive.

Independentsamples ttests were conducted to examine differences in AI utilization by sex and work affiliation.

Sexbased differences were generally not significant, except for the organizational aspect, where female respondents reported significantly higher mean scores than male respondents (p = .023). This indicates more favorable perceptions of organizational support for AI among women in the sample.

Privatesector respondents reported significantly higher mean scores across all four dimensions than government respondents, indicating more favorable AIrelated conditions in private organizations.

Oneway ANOVA was used to examine differences in AI utilization across groups defined by years of work experience.

Post hoc tests (Table 9) showed that respondents with 1–10 years of experience consistently reported significantly higher mean scores than those with longer experience, especially those with 31–40 years and 41 years and above.

Only significant comparisons are shown.

Oneway ANOVA was also used to examine differences in AI utilization across age groups

Tukey post hoc tests (Table 11) indicated that the youngest age groups (21–30 and 31–40 years) had significantly higher mean scores than older age groups across all dimensions.

Only significant comparisons are shown.

4. Discussion

The results indicate that civil engineers in Quirino Province operate in an environment that is receptive to AI but still at an emerging stage of readiness. The overall mean of 2.89 suggests that AIrelated technological, organizational, knowledgebased, and ethicalregulatory conditions are present but not yet strong. This pattern is consistent with global reviews describing construction as moving toward AI integration while still facing structural, organizational, and cultural barriers ¹¹.

Knowledge and skills received the highest dimension mean (3.03), indicating that respondents recognize the value of AI competence and report some level of AIrelated training and awareness. However, the scores remain in the moderate range, implying that deeper, practiceoriented capacitybuilding is needed if AI is to become part of routine civil engineering workflows ^{7, 13}.

Technological aspects also scored favorably (mean 2.96), reflecting perceptions that AI tools are available, reasonably accurate, and usable, and that organizations have some related infrastructure. Nevertheless, the absence of “strongly agree” ratings suggests continuing challenges in interoperability, digital infrastructure, and consistent utilization, which are commonly cited in the literature as barriers to full AI integration ^{10, 28}.

Organizational aspects scored slightly lower (mean 2.88), highlighting a familiar gap between recognizing AI’s benefits and providing robust institutional support. The significant differences favoring private sector respondents across all dimensions underscore the role of sectoral context: private organizations may have more flexible cultures, greater investment capacity, and stronger incentives to experiment with AI than government agencies constrained by bureaucratic procedures and legacy systems ^{2, 3}.

Ethical, cultural, and regulatory aspects were the weakest dimension (mean 2.69), though still positive, pointing to ongoing concerns about data privacy, job security, decision transparency, and regulatory clarity. These findings mirror international discussions emphasizing that AI deployment in safetycritical fields like civil engineering must be accompanied by clear guidelines, accountability frameworks, and communication strategies to build trust among professionals and stakeholders ^{4, 12}.

The demographic and sectoral differences observed in the study are compatible with innovationdiffusion and technologyacceptance theories. Younger and less experienced engineers reported more favorable AIrelated conditions than older, more experienced colleagues, suggesting that newer entrants to the profession may be more open to digital technologies and less tied to traditional methods ²³. The single significant sex difference in organizational aspects (favoring women) warrants cautious interpretation but signals that gender may shape how institutional support for AI is perceived or experienced.

Internationally, the study contributes empirical evidence on AI readiness in a localized civil engineering context, complementing more technically focused research on AI algorithms and applications ¹. By centering on practitioner perceptions and institutional conditions, it underscores that successful AI adoption depends on human and organizational factors as much as on technological potential ^{2, 3}.

5. Proposed Comprehensive Support Framework

Based on the findings, a Comprehensive Professional Organizational Support Program is proposed to strengthen AI utilization in civil engineering practice. The framework consists of four pillars designed to address the main gaps identified in the study.

The first pillar focuses on sustained AI education for practicing civil engineers. This includes foundational AI literacy, applicationfocused training for design, analysis, and project management, exposure to relevant case studies, and structured continuing professional development programs delivered through PICE chapters, employers, and academic partners ^{29, 7}. Because knowledge and skills are already the strongest dimension but still only moderately favorable, capacitybuilding should emphasize handson, projectoriented learning rather than purely conceptual coverage.

The second pillar emphasizes stronger institutional support. Organizations should move from generic support for innovation to concrete AI implementation plans, including clear leadership commitments, dedicated budgets, changemanagement strategies, and designated AI “champions” within project teams ¹³. These interventions are especially important for government agencies, which lag behind private organizations across all dimensions.

The third pillar addresses digital infrastructure. AI adoption requires adequate hardware, software, data storage, connectivity, and interoperability with tools such as CAD, BIM, project management systems, and monitoring platforms ^{5, 28}. Improving data quality, standardization, and system integration is essential for moving AI from experimental or isolated use toward sustainable, organizationwide practice.

The fourth pillar concerns ethical, cultural, and regulatory readiness. Professional organizations, employers, and regulatory agencies should collaborate to develop practical guidelines covering data privacy, transparency, accountability, and acceptable AI use in civil engineering decisions ^{12, 14}. These guidelines should explicitly affirm the continuing role of human professional judgment, address worker concerns about job displacement, and offer mechanisms for monitoring and evaluating AI impacts ³⁰.

Together, these four pillars translate the empirical findings into actionable strategies that can be implemented and refined by stakeholders in Quirino Province and similar settings ^{2, 3}.

6. Conclusion

Civil engineers in Quirino Province perceive AIrelated conditions in their professional environment as moderately favorable overall, with knowledge and skills and technological aspects rated higher than organizational and ethicalregulatory conditions. Significant differences across sex, sector, years of experience, and age show that AI readiness is uneven, with privatesector, younger, and less experienced engineers generally reporting more favorable conditions than governmentsector and more senior counterparts.

These results confirm that AI adoption in civil engineering requires more than technological tools; it depends on aligned efforts in capacitybuilding, organizational systems, infrastructure, and governance. The proposed fourpillar support framework offers a structured response that can guide professional organizations and institutions in advancing toward more strategic, inclusive, and responsible AI integration in civil engineering practice ¹⁶.

7. Limitations and Future Research

The study is limited to PICEaffiliated civil engineers in a single Philippine province and used a crosssectional survey design, which may restrict generalizability and preclude causal inference. Future research could adopt longitudinal designs to track changes in AI readiness over time, employ qualitative methods such as interviews or focus groups to deepen understanding of barriers and enablers, and compare multiple regions or countries to explore contextual differences in AI utilization ^{29, 30}.

ACKNOWLEDGEMENTS

The authors extend their sincere appreciation to Quirino State University and La Salette University Inc. for the institutional support provided throughout the conduct of this study. They gratefully acknowledge the assistance of PICE Quirino chapter for facilitating access to participants and endorsing the data collection activities.

Heartfelt thanks are also due to the civil engineers in Quirino Province who generously shared their time and insights by responding to the survey questionnaire; their contributions made this research possible. The authors likewise recognize the constructive comments of colleagues and peers whose suggestions helped refine the research design, analysis, and presentation of findings.

References