2026 ASEE Annual Conference & Exposition

Artificial intelligence (AI) and data-driven technologies are transforming civil engineering practice by reshaping how infrastructure is planned, designed, and managed (Nyokum & Tamut, 2025). AI tools such as predictive modeling, digital twins, and automated design optimization are integral to engineering workflows (Azanaw, 2024). As these technologies diffuse across industry sectors, they are also changing employers’ expectations for the skills and competencies required of civil engineers (American Society of Civil Engineers, 2024). However, limited empirical research has examined how employers’ expectations for AI-related skills have evolved over time and across geographic regions. Understanding these trends is essential for aligning civil engineering education with the rapidly changing demands of the profession.
The purpose of this study is to investigate longitudinal and regional trends in employers’ expectations for AI-related competencies in civil engineering job postings from 2015 to 2025. The study utilizes job posting data from LinkUp, an employer-sourced database that collects verified postings directly from over 80,000 company websites. This dataset provides a unique and reliable record of real-time labor market signals, including job titles, descriptions, locations, and posting dates. The study will focus on U.S.-based civil engineering positions, including structural, transportation, environmental, geotechnical, and construction occupations. The main research question is: How have employers’ expectations for AI-related skills in civil engineering job postings evolved over time and across geographic regions in the United States from 2015 to 2025? We will use Skill-Biased Technological Change (SBTC) theory (Autor et al., 1998) and Human Capital Theory (HCT) (Becker, 1964) as our theoretical foundation. While SBTC argues that technological progress increases the demand for workers with advanced analytical and computational abilities while HCT provides a lens for interpreting job postings as signals of the knowledge and skills employers value as productive assets.
A longitudinal text-mining and geographic analysis design will be used. Job postings will be filtered for civil engineering-related occupations by using O*NET standard occupation code (SOC). After cleaning and preprocessing, large language model (LLM)-assisted text analysis will extract and classify skill phrases from the job descriptions. Unsupervised and supervised semantic clustering methods was used to group related skills. While unsupervised algorithms identify natural patterns in the data, supervised classification validates these clusters against predefined skill categories. Temporal analysis tracks their frequency and change from 2015 to 2025. Geographic analysis will map the spatial distribution of AI-related skill demand by state and metropolitan area.
Although the study is in progress, the anticipated outcomes include (1) a data-driven map of how AI-related skills in civil engineering have evolved over time, (2) regional comparisons highlighting areas of concentrated AI-related demand, and (3) educational implications for aligning curricula, workforce readiness, and professional development with emerging industry needs. By interpreting employer expectations through the lens of technological and human capital change, this research will contribute to both engineering education and workforce development. The findings are expected to provide actionable insights for curriculum renewal, faculty training, and the design of learning experiences that prepare future civil engineers to thrive in AI-augmented work environments.