This thesis addresses the difficulty of moving from as-built Building Information Modelling (BIM) to an operational Digital Twin when model and operational evidence are incomplete. It develops a methodology for mechanical, electrical and plumbing (MEP) systems with three entry conditions. Scenario A begins with BIM/IFC but limited or missing operational information; Scenario B begins with sensors, a building management system (BMS) or operational records but no suitable BIM/Asset Information Model; and Scenario C begins with both sources, which are not integrated. Only Scenario A is examined empirically. Scenarios B and C remain methodological transfer pathways.
The research follows a design-science approach. The methodological artefact is a task-led decision method that determines which information can be reused from an available BIM, which information must be verified or enriched, and which evidence must be accepted before model records are connected to operational observations or FM tasks. Entry scenarios select the starting path; stages organise the work; evidence gates control progression; and four records preserve gaps, requirements, identity mappings and acceptance decisions.
Scenario A is demonstrated through a reproducible structural audit of nine available IFC exports. The files contain model identifiers, type relationships, representations and storey containment that can be reused as source evidence. They also contain unresolved name/type conflicts and provide insufficient evidence for accepted physical asset identity, explicit MEP port topology, base quantities, operational observations or FM outcomes. The principal contribution is the decision method: it separates reusable BIM information from information requiring verification or enrichment and specifies the evidence needed before operational use. The case validates the model-readiness part of the method; it does not constitute a deployed Digital Twin.
|
