METROSCOPE technology was initially designed for Nuclear Reactors, before we started our diversification to Gas Plants and Conventional Utilities.

METROSCOPE is already in operation on several units at EDF since 2017 and is in the process to be implemented on the whole fleet of French Nuclear Plants.
By design, METROSCOPE applies to most types of thermal hydraulic processes.

Diagnosis of a Data Center

The illustration hereunder shows the cooling circuit of a Data Center that can be automatically diagnosed by METROSCOPE.

All the hazards that can affect it may have a direct impact on the different measurements made on the thermal circuit of the Data Center. Most of them are difficult to dissociate both from the cooling demand and from the natural variation of the measurement. It requires expertise to assess the expected performance of the installation.


The list is non-exhaustive and is meant to grow throughout the use of METROSCOPE on your installation. By design, METROSCOPE will easily evolve with your Data Center and integrate new measurements and/or new hazards.


METROSCOPE comes ready to use for the supervision of your Chiller, Cooling Towers, Heat Exchangers, Pumps and Valves.


METROSCOPE is based on a physical model of your Plant, and thus integrates design considerations and physical equations: in a word METROSCOPE has its own expertise. The technology behind METROSCOPE is particularly efficient at integrating expertise in its algorithms.


By comparison with Big Data / Machine Learning / Statistical approaches, METROSCOPE can interpret the measurement with a real understanding of the physics and the uncertainties of the sensors. Unlike Big Data, METROSCOPE relies on Markov Chain, an Artificial Intelligence designed for Small Data problematics, whereas empirical data is not enough to address your stakes.


METROSCOPE learns from the physics to interpret measurement, not the other way around.

Value Added

Energy Performance enhancement

The Power Usage Effectiveness (PUE) is the most popular metric of the Energy Performance of the Data Center. PUE is the ratio of total amount of energy used by a computer Data Center facility to the energy delivered to computing equipment. METROSCOPE will perform an early detection of the hazards and will compute their contribution to the PUE degradation. METROSCOPE SaaS turnkey solution is tailor made for addressing your PUE issues.

Thus, METROSCOPE will help your Data Center reaching the highest possible environmental performance.

Stakes of Maintenance

METROSCOPE provides reliable, automated diagnosis, identifying hazards affecting the industrial process that could result in a high energy consumption, long-term damage to a component, or even the outright shutdown of the installation.

Thanks to METROSCOPE, maintenance team gets a better insight of the behavior of the main components of your process. Many of the hazards detected can be fixed at lower cost, with lesser impact.

Enable a less costly maintenance operation

Have a direct impact on the life expectancy of the components.

Thanks to METROSCOPE you will be able to analyze the whole history of the plant and track down former hazards (as long as you have the data stored). It makes METROSCOPE a powerful capitalization tool! That will in the end help you challenging your maintenance programs and suppliers.


Get rid of the burden of drawing quantities of Excel curves! With METROSCOPE, your experts are automatically alerted on the components that do require their attention. METROSCOPE provides automated results, while displaying every steps of its reasoning, making it accessible to the user.
See METROSCOPE as a sandbox, not as a black box.


Start with one plant, then extend it to your fleet.


We can handle everything on our own or help you building and operating physical Digital Twins. 

In any case, you are the only owner of your data, and you get to keep the digital twin. 

If you can already operate physical digital twins, you might be interested to become our partner. 

We can work either in SaaS mode or “on your premises” in your own proprietary environment.