Replication package for "Quantifying non-energy co-benefits of building automation upgrades in offices: a simulation-based IAQ assessment under parametric uncertainty"

This dataset is the replication package for the study "Quantifying non-energy co-benefits of building automation upgrades in offices: a simulation-based IAQ assessment under parametric uncertainty", submitted to Developments in the Built Environment. The study evaluates two ventilation control strategies representing different Building Automation and Control Systems (BACS) automation levels in office buildings: a fixed-schedule baseline (representative of EN ISO 52120-1 Class D) and a CO2-based demand-controlled ventilation strategy (representative of EN ISO 52120-1 Class A). The workflow integrates EnergyPlus 23.2 building performance simulation, stochastic occupancy modelling, scripted control logic, and Sobol sensitivity analysis to estimate productivity- and health-related co-benefits under parametric uncertainty. The analysis is conducted across three European climates (Helsinki, Brussels, Rome) for a prototype terraced office. Contents include:- Aggregated key performance indicator outputs per climate (CO2 exposure metrics, fan energy consumption, co-benefit estimates).- Sobol first-order and total-order sensitivity indices for both control scenarios.- Trained surrogate regression models linking input parameters to economic co-benefits, disaggregated by scenario (DCV/Fixed), zone type (Meeting rooms/Private offices), and domain (Productivity/Health).- Post-processing scripts (Python and bash) covering Sobol sample generation, KPI extraction, sensitivity index computation, and co-benefit calculation.- Baseline EnergyPlus 23.2 input file for the terraced office prototype. Raw EnergyPlus simulation outputs are not archived because of storage volume but can be regenerated from the deposited inputs and scripts. Weather data files for the three climates are publicly available from EnergyPlus.net and climate.onebuilding.org. All scripts and the baseline IDF use generic path placeholders (e.g., <PROJECT_DIR>, <SCHEDULES_DIR>) instead of absolute personal paths; see README.md for the full substitution table and reproduction instructions. Code is released under the MIT License; data is released under Creative Commons Attribution 4.0 International (CC-BY-4.0).