From Fragmented Site Information to Closed-Loop Coordination: Development of an Evidence-Traceable BIM Framework for Super High-Rise Construction in Xi’an, China

Digitalisation has increased the volume of information available on construction sites, yet specialised models, monitoring systems, quality records, equipment platforms, and safety workflows frequently remain disconnected. This study develops an evidence-traceable Building Information Modelling (BIM)-driven smart construction-site management framework for super high-rise projects in Xi’an, China. A design-science-oriented procedure transformed the empirical outputs of 15 semi-structured interviews into a candidate management artefact. The antecedent dataset comprised approximately 87,000 Chinese characters of transcript material, a refined 118-code codebook, 15 role-domain stakeholder needs, and 28 functional requirements (22 Core and 6 Supporting). These inputs were triangulated with international and Chinese information-management standards, recent BIM-IoT and digital-twin literature, and contextual constraints of super high-rise construction. A nine-step development process established the design-input repository, mapped requirements to candidate elements, formed modules, compared layer arrangements, specified BIM-site data associations, designed process and state rules, defined interfaces and permissions, prepared representations, and completed internal checks. The resulting Framework V1.0 contains four technical layers and four configurable management modules connected by a BIM spatial/object index and a unified event-responsibility loop. Structural mapping located all 28 requirements within the framework. Walkthroughs of four representative scenarios produced continuous conceptual paths from data acquisition and spatial association to judgement, task assignment, action, reinspection, closure, and archiving. The contribution is not a new collection of isolated digital functions, but a requirements-driven coordination architecture that distinguishes authoritative data sources from the BIM index and makes responsibility and closure explicit. The reported 28/28 result demonstrates design coverage only; it does not establish software implementation, user acceptance, or field-performance improvement. Expert evaluation and subsequent field testing remain necessary.