Computational Decision Support for Historic Hospital Adaptive Reuse: Development and Internal Verification of HARP v1.1

The adaptive reuse of historic hospitals requires comparing prospective functions against inherited architectural configurations, heritage constraints and contextual conditions, often under incomplete documentation. This study develops and internally verifies HARP v1.1, a hospital-specific rule-based decision-support framework designed to make preliminary reuse assessments transparent and traceable. Drawing on 43 comparative cases, 15 supplementary cases and iterative architectural modelling, the framework connects evidence-supported inputs, derived indicators and conditional rules to eight prospective reuse directions. Rule-normalised compatibility scores are reported separately from Evidence Adequacy. The original 19 software-conformance tests were passed, and deterministic replay reproduced the Pirot reference assessment. Mixed Use ranked first in the reference assessment, while sensitivity analyses and alternative scoring controls identified modelling dependencies that could have affected the rankings. The developmental coding pilot informed protocol refinement; a separate, focused, independent-coder check of the final 35 inputs for Pirot yielded 32/35 exact agreements (91.4%; unweighted Cohen’s κ = 0.859), with three one-level disagreements. The findings demonstrate the implementation consistency and reproducibility of the reference assessment, while the framework’s traceable computational structure supports the inspection of its assumptions and dependencies. Further evaluation across independent assessments and additional hospital cases is needed. HARP v1.1 is proposed as a research prototype for preliminary architectural screening and briefing, rather than as a validated predictor of optimal reuse.