Ricardo Eiris

Workforce sustainability remains a significant challenge in the construction industry, as evolving demands, shifting demographics, and persistent labor shortages continue to threaten employee retention. Workforce retention is a critical indicator of workforce sustainability. To strengthen retention, it is essential to understand the lived experiences of professionals and the conditions that support or hinder sustained engagement in the sector. However, little is known regarding factors that affect workforce retention among African American professionals in the construction industry. This study investigates African American construction professionals’ perceptions of race-based employment discrimination, including direct and vicarious experiences across field, office, and hybrid work environments, and explores its perceived relevance to retention intentions. Semi-structured interviews were conducted with thirty-four African American professionals from diverse locations, occupational roles, and experience levels within the U.S. construction industry. A blended thematic analysis identified three themes describing participants’ reported experiences: (1) Workplace Inequities and Employment Disruption, (2) Professional Expertise Devaluation and Unequal Recognition, and (3) Exclusionary Workplace Conditions and Psychosocial Strain. Findings reveal that participants’ accounts included expressed departure intentions and retrospective reports of project departures and employment changes, including waiting for pay before leaving or obtaining another job. Reduced motivation and feeling undervalued were distinguished from explicitly expressed desires or plans to leave. Moreover, participants reported that lack of recognition for their skills and continued discriminatory practices reduced their sense of value, expressing a reduced willingness to remain on a project or with an employer in connection with these experiences. This study adds to the understanding of factors that affect retention intentions by reporting perceived barriers rooted in race-based employment practices. The study contributes interview evidence from African American construction professionals and interprets their reported experiences using the existing workforce sustainability framework, identifying organizational concerns, and potential responses for further evaluation. Furthermore, the study offers practical implications for enhancing workforce retention and workforce sustainability.

Frontline supervisors (FLS), commonly referred to as foremen or forepersons, play a critical role in the construction industry and lean project delivery as they direct the craft workforce and oversee production at the workface. FLS are responsible for ensuring that …

This research presents a framework for real-time task detection and digital twin modeling based on human posture estimation from 360° videos. The system integrates markerless human posture estimation with task classification and digital twin visualization. Posture estimation using MediaPipe provided accurate skeletal tracking, while kinematic feature extraction enabled detailed motion analysis. Gaussian Mixture Models effectively segmented task transitions, distinguishing between different phases of ladder use. Gaussian Splatting helped realistic and adaptive visualizations, for a digital twin that accurately represented human-environment interactions. Using these techniques, the framework achieves a non-intrusive and scalable approach to task detection and digital twin modeling. The system captured human movements from 360° videos and classified them into task-specific segments. The results demonstrated that task detection based on posture estimation could improve workplace safety by identifying inefficient or hazardous postures. The digital twin representation can be analyzed for movement patterns and ergonomic risk assessment.

This paper introduces a novel framework enabling natural language question answering on Piping and Instrumentation Diagrams (P&IDs), addressing a critical gap between engineering design documentation and intuitive information retrieval. Our approach transforms static P&IDs into queryable knowledge bases through a three-stage pipeline. First, we recognize entities in a P&ID image and organize their relationships to form a base entity graph. Second, this entity graph is converted into a Labeled Property Graph (LPG), enriched with semantic attributes for nodes and edges. Third, a Large Language Model (LLM)-based information retrieval system translates a user query into a graph query language (Cypher) and retrieves the answer by executing it on LPG. For our experiments, we augmented a publicly available P&ID image dataset with our novel PIDQA dataset, which comprises 64,000 question–answer pairs spanning four categories: (I) simple counting, (II) spatial counting, (III) spatial connections, and (IV) value-based questions. Our experiments (using gpt-3.5-turbo) demonstrate that grounding the LLM with dynamic few-shot sampling robustly elevates accuracy by 10.6–43.5% over schema contextualization alone, even under high lexical diversity conditions (e.g., paraphrasing, ambiguity). By reducing barriers in retrieving P&ID data, this work advances human–AI collaboration for industrial workflows in design validation and safety audits.

Construction is one of the most dangerous industries for people on jobsites all around the world. In response, safety training has been employed to address safety hazards proactively. However, training that utilizes traditional media (e.g., texts, images, videos) is ineffective due to its passiveness. Virtual reality (VR) –based safety training can address this challenge by offering active, hands-on learning experiences. Yet hazard representation limitations and technical development challenges of computer-generated VR have hindered their ability to become widespread training methods. One emerging alternative to common VR is 360-degree virtual reality (360VR), which relies on 360-degree panoramic captures of reality, significantly increasing realism and reducing VR modelling requirements. This chapter introduces a road map of the technologies that support 360VR reality capturing, visualization, and interaction at the frontier of safety training applications. Three case studies were selected to illustrate how 360VR is applied within the safety training context. The findings of the chapter highlight how 360VR delivers significant sense-of-presence benefits to trainees while producing similar learning outcomes as traditional training. Moreover, the case studies demonstrate several approaches available to provide safety content using 360VR materials.

Trends in the construction domain, educational enrollment, student graduation, student industry employment, and workforce retention demonstrate that minorities—Hispanics, African Americans, Asian Americans, and Native Americans—are often excluded, segregated, or ignored in this domain. A systematic literature review (N = 68) was conducted to investigate the causes and effects of racial disparities produced by systemic racism in the educational and workforce domains of construction. Particularly, this paper focuses on exploring how racial disparities in construction impact minorities, the current state of the racial divide, the practices that perpetuate racial inequities, and the strategies currently used to prevent, in a sustainable manner, such practices. The results of this systematic literature search revealed that exclusionary practices and ideologies cause an underrepresentation of minorities in construction that directly affects employment and industry representation in the United States. Previous studies show that systemic racism in construction has been supported by meritocracy and colorblindness ideologies, creating an unwelcoming environment where racial minorities have difficulties identifying with the construction domain. Furthermore, systemic racism affects students after graduation from construction programs, as industry trends showcase issues with minorities joining or staying in the field. Although racial disparities caused by systemic racism are an existing issue in the education and workforce domains of construction, there are a rising number of publications that strive to understand how to sustainably increase diversity, equity, and the inclusion of racial minorities. An increasing number of available tools, such as anti-bias and awareness training programs, are being used as a sustainable practice in construction education and in the construction industry to mitigate the effects of systemic racism. Ultimately, this paper’s contribution centers on describing the “who”, “how”, and “what” regarding the effect of racial disparities in the construction domain, which reduce the number of minority professionals coming into and staying in the industry.

Building science employs student–professional interactions to increase the learning and engagement of domain topics. Although these interactions are beneficial to students, labor shortages and automation trends will create barriers to expose students to professionals in the future. One solution is to utilize virtual situated learning contexts for students to experience simulated interactions with virtual professionals. Within these simulations, virtual humans are often used as a vehicle to preserve knowledge and deliver it to students. However, the impact of the fidelity of virtual human appearance on providing knowledge to the students within the building science domain remains unexplored. This paper presents a virtual reality (VR) educational system integrated with digital storytellers and grounded in the building science electrical career path context to explore how virtual human appearance fidelity affects student learning and engagement. A between-subject study was conducted with 42 construction management participants to explore the impact of 3 virtual human appearance fidelity conditions (realistic, modeled, and abstracted virtual human) in terms of student learning, social presence, virtual human persona factors, and topic interest. The findings of this study demonstrated that successful educational outcomes could be achieved with low virtual human appearance fidelity due to the lack of significant differences observed across experimental conditions. Furthermore, the study results did not detect significant differences in students' engagement by increasing the virtual human appearance fidelity. However, students across all experimental conditions reported engagement increments in the topic learned. The increments in student engagement suggest that the virtual human's age and gender are important factors to consider when designing educational interventions. These results contribute to the understanding of how to create virtual humans for building science educational interventions.