Using IoT data to produce mixed-fidelity building models

Posted on

< Project Overview >

There are several significant barriers to implementing IoT solutions to engineering problems, for example understanding the density and positioning of IoT sensors in real world settings, and sharing data between different systems and platforms for effective performance analysis. Many building projects make use of IoT during the design and build phase, but fail to map the in-service conditions of building after years of occupancy.

This project aims to overcome barriers related to sparse sensing IoT and to establish a business case for future work, developing a case study of computationally efficient building model based on a mixed-fidelity multicomponent system. The project will aim to coordinate systems to enable effective knowledge exchange on modelling experience at University of Sheffield and project partners, including University of Cambridge and HMU. By demonstrating the value of a small number of sensors in strategic locations in buildings, the project will improve the mapping of building conditions during occupancy, for example modelling airflow and heat.

This project will also develop a built environment monitoring tool based on mixed-fidelity multicomponent building models. This will take boundary data from data-driven abstract models where the space of interest exchanges air or heat with adjacent spaces, air ducts, etc.


  • A case study report describing the integration of the modelling systems, and the results on the Diamond Building space.
  • An academic paper derived from the case study report.
  • A knowledge exchange workshop discussing IoT buildings data and its applications for improving building modelling. This will be facilitated by the project partners but including other interested parties in building services and building analytics from the South Yorkshire region.

Project lead:

Dr Ramsay Taylor – University of Sheffield