Unlocking Energy and Carbon savings
in buildings

Revolutionising energy and carbon savings in buildings via our patented Digital Twin!

We facilitate energy and carbon reduction for the 90% of buildings that do not have a
Building Management System (BMS).

For those buildings that have a BMS, we virtualise it via our Digital Twin and achieve up to 40% savings.

Our products

optimise ai process - Optimise AI

Non-domestic buildings without a BMS

Saving energy and carbon for the 90% of commercial buildings don’t have a BMS.

Non-domestic buildings with an inefficient BMS

Virtualising the BMS via our Digital Twin and supporting the Facilities Managers to reduce energy and carbon.

Non-domestic buildings with a state-of-the-art BMS

Our Al actuates (operates) the building in sync with the Facilities Manager

See optimise AI in action!

Award-winning savings! We have achieved energy savings of up to 40%

Patented technology

Apart from an immersive 3D viewer and data visualisations, Optimise AI features a patented technology for saving energy and reducing carbon footprint leveraging the power of ontologies and Semantic Web of Things

Optimise-AI is backed by:

Carbon 13 Logo
Barclays EagleLabs Logo

Our clients

We have been working with a number of key actors the industry who elected to leverage on the Optimise AI potential to reduce their costs, while also proving its efficacy in a number of research projects

cardiff uni logo - Optimise AI
cardiff council logo - Optimise AI
atkins company logo 1 - Optimise AI
catapult 2 - Optimise AI
client blaenau gwent - Optimise AI
clients network rail - Optimise AI
logo fidia - Optimise AI
logo self - Optimise AI
logo knoholem - Optimise AI
logo emte - Optimise AI

This is how we save you
energy and carbon:

Data acquisition and twin creation

Optimise AI collects data directly from the meter / BMS and the building owner or the Facilities Manager.

It then creates an optimised Digital Twin using the data from:

  • BIM (Building Information Model)
  • BMS (Building Management System)

Simulation

Optimise AI creates bespoke energy simulation models to forecast energy usage and carbon emissions, on various levels of granularity: buildings, blocks, floors, and spaces.

Optimisation

Optimise AI uses the simulation data and fine-tunable Machine Learning models to:

  • set energy and carbon saving scenarios
  • run optimisations
  • provide control instructions to the Facilities Manager.

Actuation

The building owner or Facilities Manager operates the building via the Digital Twin to generate new savings forecast.

Optimise AI can reduce valuable time and resources spent using the automation manager.

Case studies

Energy savings
CO2 reduction

The Ebbw Vale pilot (Wales) is located on the floor of a valley to the south of the town of Ebbw Vale and covers an area of approximately 78 hectares and was formrly occupied by a steelworks, which closed in 1982.

Demolition and remediation was subsequently undertaken to create sires for residential, commercial, education and leisure developments.

Research papers

Li, Y., Rezgui, Y. and Kubicki, S. 2020. An intelligent semantic system for real-time demand response management of a thermal grid. Sustainable Cities and Society 52, article number: 101857, DOI: 10.1016/j.scs.2019.101857

Jayan, Mr Bejay, et al. 2016. An analytical optimization model for holistic multiobjective district energy management—A case study approach. International Journal of Modeling and Optimization 6.3, 156-165, DOI: 10.7763/IJMO.2016.V6.521

Energy savings
Thermal savings
CO2 reduction

The FIDIA pilot in Rome (Italy) has two swimming pools, one volleyball indoor court, one gym and two outdoor multi-purpose courts. The local generation of energy is provided by gas boilers and a Biomass Plant.

The main automation based intervention performed in FIDIA entailed the pump control bleeding of the AHU and the fan control AHU of the pool, as well as the handling ventilation, hot water valves and dampers.

Before SPORTE2 intervention, the pumps in FIDIA worked 24 hours a day which results in huge energy consumption. The pumps control FIDIA rule based scenario aims to use pumps only when it is needed. A schedule has been thus introduced first to ensure the pumps (PUMP1, PUMP2, PUMP3) availability during various courses which take place in the area.

The best practise values initially defined through literature and normative in rules of FIDIA pump control scenario, were adjusted and tuned, after their implementation in the given facility, once a decent set of monitored results will be available to study the current behaviour of the system.

Research papers

Yuce, B., Li, H., Rezgui, Y., Petri, I., Jayan, B. and Yang, C. An Indoor Swimming Pool Case Study: Utilizing Artificial Neural Network Prediction to Achieve Better Energy Saving and Comfort Level, Energy and Buildings, DOI: 10.1016/j.enbuild.2014.04.052, 2014Petri, I., Li, H., Rezgui, Y., Chunfeng, Y., Yuce, B. and

Petri, I., Li, H., Rezgui, Y., Chunfeng, Y., Yuce, B. and Jayan, B. A HPC based cloud model for real time energy optimization, Enterprise Information Systems, DOI: 10.1080/17517575.2014.919053, 2014, 2012 (Impact Factor: 9.2)

Energy savings
Thermal savings
CO2 reduction

The EMTE Etxebarri Bilbao (Spain) pilot has two swimming pools and a multisport indoor court. The local generation of energy is provided by solar photovoltaic and solar thermal panels.

The EMTE pilot currently adopts a fixed schedule executed by a local BMS for renovation. The aim of EMTE water heating scenario was to maximise the usage of installed solar thermal energy to heat water used in the adult swimming pool.

The optimisation rules in the EMTE scenario:

  • When the water temperature in the solar tank is favourable, the water will be used to supply the swimming pool and ACS (hot sanitary water tank) when the temperature there drops below the required set point.
  • The 5% renovation of water (about 20 m3 of water in EMTE adult swimming pool) can be achieved every 24 hours. In order to meet this requirement, the water flow going into the swimming pool needs to be checked (for renovation) daily between 8 AM to 10 PM.

Facilities covered

  • Adult Pool (indoor) – size: 25m x 12,5m, depth: 1,65 to 2,10, capacity: 578 m3
  • Children Pool (indoor) – size: 12,5m x 6 m, depth: 0,98 to 1,1m, capacity: 75 m3
  • Multiport Court (Indoor, for Basketball, Football, Rhythm Gimnastic, Handball, etc.)
  • Stands with more than 1.400 seats
  • Weight room (14 x 14,85 m)
  • Multipurpose room (5 rooms)
  • Indoor Cycling room (8,60 x6,70 m)
  • Yoga room
  • Paddle courts (2 indoor)
  • Tennis courts with bleachers (2 courts)
  • Football court with bleachers (45,50 x35 m)
  • Indoor rock climbing
  • Sauna-Solarium

Research papers

Petri, I., Li, H., Rezgui, Y., Chunfeng, Y., Yuce, B. and Jayan, B. A HPC based cloud model for real time energy optimization, Enterprise Information Systems, DOI: 10.1080/17517575.2014.919053, 2014 (Impact Factor: 9.2)

Petri, I., Li, H., Rezgui, Y., Chunfeng, Y., Yuce, B. and Jayan, B. A Modular optimization model for reducing energy consumption in large scale building facilities, Renewable & Sustainable Energy Reviews, 10.1016/j.rser.2014.07.044, 2014 (Impact Factor: 6.577)

Yang, C., Li, H., Rezgui, Y., Petri, I., Yuce, B., Chen, B. and Jayan, B. High Throughput Computing based Distributed Genetic Algorithm for Building Energy Consumption Optimization, Energy and Buildings, DOI: 10.1016/j.enbuild.2014.02.053, 2014 (Impact Factor: 3.254)

Yuce, B., Li, H., Rezgui, Y., Petri, I., Jayan, B. and Yang, C. An Indoor Swimming Pool Case Study: Utilizing Artificial Neural Network Prediction to Achieve Better Energy Saving and Comfort Level, Energy and Buildings, DOI: 10.1016/j.enbuild.2014.04.052, 2014 (Impact Factor: 3.254)

Electrical energy savings
Electrical CO2 reduction
Thermal energy savings
Thermal CO2 reduction

The SELF Santa Maria de Lamas pilot (Portugal)  has two swimming pools and two gyms. The pilot has three systems for the generation of hot water: solar system, CHP, boilers.

In SELF pilot, the swimming pool area where the sensor network optimization has been implemented into, has an external wall that is completely glazed with shading while the opposite side includes bleachers. The air heating and ventilation system is composed by two Air Handling Units providing the air supply and return through a distribution system. The module optimises supplied thermal energy and fan electrical consumption to achieve the optimal indoor thermal conditions minimising the energy used for the air treatment.

Facilities covered

  • Olympic Pool (indoor) size: 50m x 25m, depth: 1,20 to 1,80, Capacity: 1650m3
  • Learning Pool (indoor) size: 20m x 6m, depth: 1,1m, Capacity: 132 m3
  • 2 Gyms (indoor) with one provided of electric equipment, such as electric bicycles
  • Water Heating for pools and showers
  • Air Heating for large environments
  • Electric Energy for indoor/outdoor lighting and electric equipment

Research papers

Petri, I., Li, H., Rezgui, Y., Chunfeng, Y., Yuce, B. and Jayan, B. A HPC based cloud model for real time energy optimization, Enterprise Information Systems, DOI: 10.1080/17517575.2014.919053, 2014, 2012 (Impact Factor: 9.2)

Petri, I., Li, H., Rezgui, Y., Chunfeng, Y., Yuce, B. and Jayan, B. A Modular optimization model for reducing energy consumption in large scale building facilities, Renewable & Sustainable Energy Reviews, DOI: 10.1016/j.rser.2014.07.044, 2014 (Impact Factor: 6.577)

Yang, C., Li, H., Rezgui, Y., Petri, I., Yuce, B., Chen, B. and Jayan, B. High Throughput Computing based Distributed Genetic Algorithm for Building Energy Consumption Optimization, Energy and Buildings, DOI: 10.1016/j.enbuild.2014.02.053, 2014, (Impact Factor: 3.254)

Yuce, B., Li, H., Rezgui, Y., Petri, I., Jayan, B. and Yang, C. An Indoor Swimming Pool Case Study: Utilizing Artificial Neural Network Prediction to Achieve Better Energy Saving and Comfort Level, Energy and Buildings, DOI: 10.1016/j.enbuild.2014.04.052, 2014Petri, I., Li, H., Rezgui, Y., Chunfeng, Y., Yuce, B. and

Energy savings

The Care Home Forum Building pilot in Eersel (Netherland) improves energy efficiency of public buildings by up to 30% by offering a system that monitors energy consuming devices and informs negotiable energy optimization plans, orchestrated by the building energy manager, taking into account a wide range of objectives, including occupants’ comfort.

This objective is achieved through progressive cycles of integration of computer models, including the BIM (Building Information Model) and energy simulation models, used to predict and manage energy efficient behaviour.

This approach aims to overcome the high variety, and resulting extant incompatibility, of the different models currently used in the construction and building management industry, while responding to on-going dynamic changes in usage and configuration of individual buildings.

Research papers

Howell, S. K.et al. 2019. User centered neuro-fuzzy energy management through semantic-based optimization. IEEE Transactions on Cybernetics 49(9), pp. 3278-3292, DOI: 10.1109/TCYB.2018.2839700

Yuce, B. and Rezgui, Y. 2017. An ANN-GA semantic rule-based system to reduce the gap between predicted and actual energy consumption in buildings. IEEE Transactions on Automation Science and Engineering 14(3), pp. 1351-1363, DOI: 10.1109/TASE.2015.2490141

Yuce, B. and Rezgui, Y., 2015. An ANN-GA semantic rule-based system to reduce the gap between predicted and actual energy consumption in buildings. IEEE Transactions on Automation Science and Engineering, 14(3), pp.1351-1363: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7317804 (Impact factor: 2.696)

Our team

Nick Tune

CEO, Co-founder,
Lead – Strategy, Sales & Operations

  • 20+ years, Contech & AECO Digitalisation​
  • Established two AECO tech businesses
  • Global Technical Engineering Director – ATKINS
  • Commissioner, National Infrastructure Commission Wales​
Prof. Yacine Rezgui

CSO, Co-Founder,
Lead – 
Product Innovation

  • Professor in Urban Intelligence, Cardiff University
  • Smant Cities Lead Assessor for the European Commission
Ioan Petri

CAIO. Lead – Artificial Intelligence

  • Associate Professor, Cardiff University
Tom Beach

CTO (YEAR 1), Product Development Lead

  • Reader, Cardiff University
Andrei Hodorog

UX&UI designer

  • Research Associate, Cardiff University
Please enable JavaScript in your browser to complete this form.

Contact us

Would you like some more details, a demonstration, or simply meet us for a chat?