OccuTwin: Occupancy-Predictive HVAC Optimisation through

Deep Learning and a BIM-Coupled Digital Twin

Anwar Shanwan1,* Mariam Altaema2 Raphaël Orwa Omran3

1 Doctor in Mechanical Engineering, Researcher at University of Orléans, France

2 PhD Candidate, Faculty of Natural Sciences and Engineering, Ankara Yıldırım Beyazıt University, Ankara, Türkiye

3 Mechanical Engineering, University of Rennes, France

Received: January 18, 2025 Revised: February 28, 2026 Accepted: March 30, 2026 ⋆ Corresponding author

ABSTRACT

Buildings account for nearly 40% of global final energy consumption, with heating, ventilation, and air-conditioning

systems responsible for the largest single share of that load. Conventional schedule-based HVAC controllers operate

on fixed occupancy assumptions and are consequently unable to exploit the predictable but irregular occupancy

patterns that characterise modern working environments. This paper proposes OccuTwin, an integrated framework

that couples multi-step occupancy forecasting with a BIM-based digital twin to enable genuinely predictive HVAC

optimisation. Four sequence models—Long Short-Term Memory, Gated Recurrent Unit, XGBoost with lag features,

and a Temporal Fusion Transformer—are trained on the publicly available Mind Your Building occupancy dataset

and the multi-building Building Data Genome Project benchmark to predict room-level occupancy at five-minute

resolution. The best-performing Transformer model achieves 94.8% accuracy and an improved weighted F1-

score on the 30% hold-out set, outperforming the LSTM baseline by 1.4 percentage points. An IFC-coupled

co-simulation environment links real-time occupancy predictions to a virtual HVAC thermal model implemented in

EnergyPlus, enabling zone-level set-point optimisation driven by predicted rather than sensed occupancy. Annual

co-simulation across nine building zones documents 21.2% energy savings over a schedule-based rule controller

while simultaneously improving the fraction of time within ASHRAE thermal comfort bounds from 92.4% to 97.2%.

Ablation experiments identify temporal lag features and SMOTE-based class rebalancing as the two most critical

preprocessing choices, and noise-injection tests confirm that the Transformer retains 95.1% accuracy under 15%

sensor noise—a critical property for edge-deployed IoT environments.

Keywords: Building digital twin HVAC optimisation Occupancy prediction LSTM Temporal Fusion Transformer

BIM IFC co-simulation Smart building Energy efficiency

1. INTRODUCTION

Buildings collectively consume approximately 40% of global

final energy, making the built environment the largest single

sector in national and international energy balances [1].

Within buildings, heating, ventilation, and air-conditioning

systems typically account for 40–60% of total energy use depending

on climate zone and occupancy profile [2]. Despite

decades of investment in building automation, most commercial

HVAC systems continue to operate on fixed weekly schedules

that cannot adapt to the gap between designed and actual

occupancy—the office that empties by noon, the conference

room occupied for one hour instead of four, the open-plan

floor that is half-used every Friday. This schedule-reality mis-