Capturing Organisational Knowledge, Bypassing Heterogeneity
The Lingering Challenge of Lost Knowledge in a Project-Based World
In the Architecture, Engineering, and Construction (AEC) sector, our work is defined by projects. Each successful delivery, be it a building, a bridge, or complex infrastructure, represents a massive undertaking of coordination, design, and execution. Yet, how much of the invaluable knowledge generated during these projects truly persists? How often do lessons learned evaporate once the team disbands, or critical design rationale remain buried in disconnected documents? We face a persistent challenge: data heterogeneity. Information crucial for future efficiency, risk mitigation, and innovation often sits fragmented across diverse systems, proprietary formats, and siloed disciplines. This isn’t just an inconvenience; it’s a significant strategic handicap, hindering our ability to learn, adapt, and consistently deliver excellence.
Beyond Data Storage: Introducing the Power of Ontologies
Addressing this requires moving beyond simple data storage towards intelligent knowledge management. This is where embedding organisational data as an ontology presents a transformative opportunity. An ontology, in essence, acts as a formal, explicit specification of a shared conceptualisation; a rigorously defined semantic framework. Think of it less as a database and more as an intelligent map of your organisation’s knowledge domain. It defines not just the data points (like ‘concrete specification’, ‘HVAC unit model’, ‘project milestone date’) but also their properties (compressive strength, energy rating, planned vs. actual date) and, crucially, the rich relationships between them (‘HVAC unit requires specific maintenance schedule’, ‘milestone date depends on task completion’, ‘concrete specification complies with standard X’). By structuring heterogeneous data using these rigorous formal semantics, we enable enhanced integration, clearer communication across disciplines, and powerful automated reasoning capabilities that were previously unattainable (Izhar & Shoid, 2020; Wang, 2021).
From Fragmentation to Cohesion: The Strategic Value Proposition
The strategic value for AEC organisations lies in turning fragmented data into a cohesive, consistent, and queryable knowledge base. An ontology allows us to consolidate disparate data sources: design files, construction logs, maintenance records, sensor data, into a unified framework (Izhar & Shoid, 2020). This structured approach explicitly formalises the intended meaning of concepts through axioms and relationships, creating a reliable backbone for integrating new data streams and processing complex queries that span multiple domains. For instance, understanding the impact of a specific material substitution might require linking design specifications, procurement records, installation logs, and potentially even long-term performance data, a task made feasible and efficient by an underlying ontological structure. This foundation is not static; it’s flexible and extensible, crucial for incorporating emerging data from IoT devices or evolving operational processes.
Building the Knowledge Foundation: A Blend of Expertise and Technology
Developing such an ontology is a strategic undertaking, typically involving both top-down and bottom-up approaches. It commences with defining the core concepts, classes, properties, and relationships that accurately mirror the operational and strategic realities of the organisation (Izhar & Shoid, 2020; Jia et al., 2024). This requires deep domain expertise to identify critical entities and their interdependencies, constructing a formal vocabulary that transcends ad-hoc data representations. This expert-driven modeling is often complemented by leveraging technology, employing data mining, natural language processing (NLP), and machine learning techniques to help extract, classify, and structure information from existing legacy databases, unstructured text documents, and other digital artifacts (Ghazal et al., 2021; Previtali et al., 2020). The synergy between advanced computational tools and invaluable domain expert input is key to reconciling semantic discrepancies and building a harmonised, comprehensive model that truly supports large-scale integration and reflects existing practices (Ghazal et al., 2021).
Enhancing AEC Workflows: Interoperability and Intelligence
Within the AEC sector specifically, integrating ontologies with established standards like Industry Foundation Classes (IFC) significantly boosts data sharing and interoperability across the entire project lifecycle (Jia et al., 2024). Engineers, architects, and construction managers can leverage these shared semantic frameworks to model complex building elements, intricate workflows, and detailed maintenance histories far more effectively, enhancing communication and shared understanding among diverse stakeholder groups (Jia et al., 2024; Farghaly et al., 2022). Embedding IFC semantics within an ontology enables a systematic mapping of geometric data, spatial relationships, material properties, and construction processes, directly mitigating the persistent problem of data silos (Jia et al., 2024). This integration empowers Building Information Modelling (BIM) platforms, transforming them from primarily geometric repositories into rich, semantic knowledge hubs that improve design fidelity, enable more sophisticated analysis, and drive operational efficiency (Jia et al., 2024; Maliha et al., 2020). The ability to perform context-aware queries across these semantically enriched datasets unlocks deeper insights and supports more complex decision-support systems, such as assessing building energy performance or simulating construction sequencing (Kaur et al., 2022; Hu et al., 2021).
Powering Next-Generation Capabilities: Reasoning, Digital Twins, and AI
The true power of well-embedded ontologies extends beyond integration and querying into the realm of automated reasoning and predictive capabilities. The explicit, formal semantics allow systems to infer new knowledge, facts not explicitly stated in the raw data but logically derivable from the defined relationships and rules (Hu et al., 2021). In high-stakes AEC environments, this inferencing capability can proactively identify potential risks, flag compliance issues, or optimise processes based on codified best practices or regulations (Xie et al., 2022; Farghaly et al., 2022; Kücükavci et al., 2024). This semantic foundation is also critical for realising the full potential of Digital Twins. By providing the framework to integrate real-time sensor data and operational logs with the static design information (BIM), ontologies enable the creation of dynamic, high-fidelity virtual replicas of physical assets. These ontology-based Digital Twins support real-time monitoring, predictive maintenance, performance optimisation, and sophisticated scenario planning throughout the asset lifecycle (Xie et al., 2022; Gardner, 2022). Furthermore, the structured, context-rich data curated within an ontology serves as high-quality fuel for downstream machine learning algorithms, improving the accuracy of predictive analytics for forecasting costs, schedules, or potential failures (Kaur et al., 2022; Hu et al., 2021).
Navigating the Implementation Path: Strategic Considerations and Challenges
Embarking on an ontology initiative is not merely a technical project; it’s a significant strategic commitment demanding careful planning and organisational alignment. Success requires more than just technology; it necessitates strong executive support and buy-in from managerial layers to ensure the ontology genuinely reflects and supports corporate goals and core business processes (El-Diraby, 2023; Izhar & Shoid, 2020).
Organisations must anticipate and address several inherent challenges. Designing a comprehensive yet practical ontology requires balancing expressive power with maintainability and computational efficiency, demanding specialised expertise and continuous validation by domain specialists (Sadeghineko & Kumar, 2021; Ghalandar & Lindkvist, 2023).
Integrating with heterogeneous legacy systems often involves complex data cleansing, semantic mapping, and potentially developing bespoke software (El-Diraby, 2023; Ahmed et al., 2022). Furthermore, managing the organisational change aspect is often the most critical factor. This involves investing in training, addressing the learning curve, overcoming potential resistance to new data governance paradigms, and fostering a culture of data quality stewardship (El-Diraby, 2023; Gardner, 2022). Finally, scalability must be considered from the outset, as large, complex ontological graphs can impose performance overhead on reasoning and query engines, potentially requiring investments in optimised algorithms and high-performance computing (HPC) infrastructure.
The Imperative for Change: Building a Knowledge-Driven Future
While the challenges are real, the strategic imperative for AEC organisations to adopt more sophisticated knowledge management strategies like ontological embedding is undeniable. The limitations discussed, computational complexity, integration hurdles, organisational inertia, and the need for ongoing maintenance, are increasingly being mitigated by technological advancements in graph databases, AI-driven semantic tools, and cloud computing (Ghalandar & Lindkvist, 2023).
The cumulative benefits: dramatically improved data integration and interoperability, enhanced decision support through powerful querying and reasoning, robust traceability and compliance, the creation of truly intelligent Digital Twins, and a foundation for advanced AI applications, offer a compelling return on investment (Izhar & Shoid, 2020; Wang, 2021; Xie et al., 2022).
In an industry grappling with complexity, tight margins, and accelerating digital transformation, the ability to effectively capture, connect, and leverage organisational knowledge is no longer a ‘nice-to-have’; it’s rapidly becoming a prerequisite for sustained competitive advantage. Investing in an ontology-based approach is investing in a future where data works smarter, collaboration is seamless, and innovation is fueled by insight.
Get Houssame E. Hsain’s stories in your inbox
Join Medium for free to get updates from this writer.
References:
Ahmed, Y., Shehzad, H., Khurshid, M., Abbas, H., Abdalla, S., & Alrefai, N. (2022). Examining the effect of interoperability factors on building information modelling (BIM) adoption in Malaysia. Construction Innovation, 24(2), 606–642. https://doi.org/10.1108/ci-12-2021-0245
El-Diraby, T. (2023). How typical is your project? the need for a no-model approach for information management in AEC. Journal of Information Technology in Construction, 28, 19–38. https://doi.org/10.36680/j.itcon.2023.002
Farghaly, K., Soman, R., Collinge, W., Mosleh, M., Manu, P., & Cheung, C. (2022). Construction safety ontology development and alignment with industry foundation classes (IFC). Journal of Information Technology in Construction, 27, 94–108. https://doi.org/10.36680/j.itcon.2022.005
Gardner, N. (2022). Digital transformation and organisational learning: situated perspectives on becoming digital in architectural design practice. Frontiers in Built Environment, 8. https://doi.org/10.3389/fbuil.2022.905455
Ghalandar, T. and Lindkvist, C. (2023). Digital transitions for future building scenarios. IOP Conference Series Earth and Environmental Science, 1176(1), 012007. https://doi.org/10.1088/1755-1315/1176/1/012007
Ghazal, R., Malik, A., Raza, B., Qadeer, N., Qamar, M., & Bhatia, S. (2021). Agent-based semantic role mining for intelligent access control in multi-domain collaborative applications of smart cities. Sensors, 21(13), 4253. https://doi.org/10.3390/s21134253
Hu, S., Wang, J., Hoare, C., Li, Y., Pauwels, P., & O’Donnell, J. (2021). Building energy performance assessment using linked data and cross-domain semantic reasoning. Automation in Construction, 124, 103580. https://doi.org/10.1016/j.autcon.2021.103580
Izhar, T. and Shoid, M. (2020). Ontology application for the organizational goals. International Journal of Human Resource Studies, 10(4), 48. https://doi.org/10.5296/ijhrs.v10i4.17839
Jia, J., Ma, H., & Zhang, Z. (2024). Integration of industry foundation classes and ontology: data, applications, modes, challenges, and opportunities. Buildings, 14(4), 911. https://doi.org/10.3390/buildings14040911
Kaur, P., Nand, P., Naseer, S., Gardezi, A., Alassery, F., Hamam, H., … & Shafiq, M. (2022). Ontology-based semantic search framework for disparate datasets. Intelligent Automation & Soft Computing, 32(3), 1717–1728. https://doi.org/10.32604/iasc.2022.023063
Kücükavci, A., Seidenschnur, M., Pauwels, P., Rasmussen, M., & Hviid, C. (2024). Efficient management and compliance check of HVAC information in the building design phase using semantic web technologies. Semantic Web, 15(5), 1959–1989. https://doi.org/10.3233/sw-243595
Maliha, M., Tayeh, B., & Aisheh, Y. (2020). Building information modeling (BIM) in enhancing the applying of knowledge areas in the architecture, engineering and construction (AEC) industry. The Open Civil Engineering Journal, 14(1), 388–401. https://doi.org/10.2174/1874149502014010388
Previtali, M., Brumana, R., Stanga, C., & Banfi, F. (2020). An ontology-based representation of vaulted system for HBIM. Applied Sciences, 10(4), 1377. https://doi.org/10.3390/app10041377
Sadeghineko, F. and Kumar, B. (2021). Application of semantic web ontologies for the improvement of information exchange in existing buildings. Construction Innovation, 22(3), 444–464. https://doi.org/10.1108/ci-03-2021-0058
Wang, M. (2021). Ontology-based modelling of lifecycle underground utility information to support operation and maintenance. Automation in Construction, 132, 103933. https://doi.org/10.1016/j.autcon.2021.103933
Xie, X., Moretti, N., Merino, J., Chang, J., Pauwels, P., & Parlikad, A. (2022). Enabling building digital twin: ontology-based information management framework for multi-source data integration. IOP Conference Series Earth and Environmental Science, 1101(9), 092010. https://doi.org/10.1088/1755-1315/1101/9/092010
