Mohammad H. Mehraban.

PhD Researcher

School of Construction, Property and Surveying

London South Bank University

I develop AI-integrated digital twins that let buildings report on their own health — predicting damp, mould and thermal discomfort before they harm the people inside.

View publications Get in touch

Google Scholar ORCID ResearchGate LSBU PURE Scopus LinkedIn

About

Healthier buildings, taught by data

I am a PhD researcher at the School of Construction, Property and Surveying, London South Bank University, where my doctoral project develops AI-integrated digital twins for the early detection and prediction of damp and mould in UK social housing, in collaboration with Wates Group. The aim is practical: give housing providers a way to see moisture risk building up in a home weeks before it becomes visible on a wall.

Alongside the PhD, I work as a Research Associate at Swansea University on the EPSRC SWITCH programme, applying explainable AI and transfer learning to thermal comfort prediction in net-zero buildings, so that comfort models trained where data is rich can serve buildings and communities where data is scarce.

My research sits at the meeting point of construction engineering and machine learning: BIM and digital twins, building performance simulation, and the automation of workflows that the industry still does by hand.

Education

PhD, Construction Management and EconomicsLondon South Bank University
MSc Construction Management, DistinctionKingston University London
BSc Civil EngineeringShahid Beheshti University

Research interests

AI / ML Applications in the Built Environment Building Information Modeling (BIM) Digital Twin and Performance Monitoring Construction Automation Energy Efficiency in the Built Environment Building Energy Simulation Human-Building Interaction Indoor Environmental Quality Personal Thermal Comfort

Research

Current projects

London South Bank University - Wates Group

Physics-informed digital twins for damp & mould prediction in social housing

A physics-informed digital twin that fuses parametric BIM, in-home IoT sensing and calibrated building energy simulation to track temperature and moisture conditions across the building envelope, then applies machine learning to forecast mould risk from hours to days ahead. Developed with Wates Group to enable healthier homes and preventive, data-led maintenance across the UK social housing stock.

Physics-informed AI-integrated digital twin framework for real-time assessment and mould growth prediction in buildings

Physics-informed, AI-integrated digital twin framework for real-time mould growth prediction — view full size.

Swansea University - EPSRC SWITCH to Net Zero Buildings

AI-integrated building management for personalised thermal comfort

Working towards an AI-integrated building management system that predicts each occupant's thermal comfort and recommends personalised control actions in net-zero buildings across the Swansea Bay region and South Wales. The project combines field data collection in occupied buildings with explainable AI and transfer learning, alongside industry partners including Energy Carbon.

London South Bank University - International Collaboration

Multi-objective optimisation of envelope retrofits for thermal resilience

A BIM-based simulation and AI framework for optimising building envelope retrofits so that homes remain habitable during extreme power outages and heat waves. More than 16,000 EnergyPlus simulations spanning four housing typologies and 14 ASHRAE climate zones train an XGBoost model to predict thermal resilience indicators such as passive survivability and time to thermal failure, and NSGA-II multi-objective optimisation then identifies Pareto-optimal retrofit strategies that balance survivability, comfort and post-outage recovery energy.