Artificial intelligence now used by Maltem Insight Performance

The Izarbel company specialising in information measurement and surveillance has developed software using artificial intelligence to detect and predict incidents on its customers’ IT networks.

The programme is named MARIA - augmented network management with artificial intelligence (maitrise augmentée des réseaux par l’intelligence artificielle).MARIA was launched in October 2017 by Maltem Insight Performance, and has been on the market since spring, adding to the tools offered by this digital services company which was created in 2008 at Izarbel under the name Alaloop. Maltem Insight Performance has around twenty employees and works with major groups in controlling and supervising their IT systems. The company is itself part of the international Maltem Consulting group of nearly 700 employees.

MARIA was developed over a period of six months at the Izarbel site by a team of three people. It uses data gathered via network equipment used by the company’s customers. Maltem Insight Performance can therefore collect information from over 7,000 sites representing 20,000 pieces of machinery. With results communicated every 5 minutes, the volume of information gathered over a month represents several billion results for customers. For example, MARIA can predict the saturation of IT networks and, in general, automate network diagnosis on a large scale using a real-time data collection system. 

Big data capture

“We capture a lot of data on our customers’ networks and are looking for a way to use them with artificial intelligence to automate problem detection and prediction” explains Pierre Rebours, Manager of the company’s customer service and a data scientist. MARIA can predict network saturation up to eight hours in advance to anticipate and reduce the the damaging effects on company activity, to detect cases which cannot be manually detected and to provide network technicians with quick response recommendations.

“MARIA software is currently in advanced use by only one customer. Its point of difference is associating saturation probability with cost in Euros. MARIA therefore allows network teams to go beyond detecting incidents to qualify the impact on the company’s economic activity. MARIA brings professional objectives to the heart of the decision-making process” says Pierre Rebours. 

For more information: www.insight-performance.com/detection-predictive-d-incidents