Launched in February 2013, the “France Très Haut Débit” plan aims to cover the entire country in very high speed by 2022, i.e. to offer high-performance Internet access to all housing, businesses and administrations. In urban areas (large conurbations and departmental capitals), private operators have invested several billion euros in the deployment of their fiber optic networks to the subscriber. Faced with this colossal task, telecom operators call on large service companies to take charge of fiber deployment and connection to customers.
At the end of December 2018, Bouygues Telecom marketed 7.2 million Fiber to the Home (commonly known as FttH) outlets; and the number keeps growing. In order to connect all its customers, the company must perform several thousand operations per day and uses several subcontracting companies for this purpose. Due to the explosion in fibre connection volumes, the quality of interventions is not always consistent and it is sometimes necessary to intervene several times to connect a customer. It is impossible to control all interventions and only 5% to 10% of the installations are controlled.
On the other hand, the explosion in the demand for interventions puts pressure on technicians who, due to lack of time to manage the most difficult situations, may fail to follow certain procedures. As a result, the network is gradually deteriorating due to non-compliant installations.
Faced with fierce competition, Bouygues Telecom wants to differentiate itself by offering a quality of service that is superior to that of its competitors. This quest for quality is at the heart of the company’s transformation process.
From post-control to real-time analysis
Deepomatic has developed a quality control solution for Bouygues Telecom to detect the presence of anomalies in technical intervention photos.
Equipped with a smartphone, the technician mandated to connect a customer to the fiber, takes a picture of the location where the connection between the optical fibers of subscribers and commercial operators is made.
The photos are analyzed in real time in the cloud by Deepomatic-driven image recognition neural networks and report anomalies such as unconnected fiber, bad colors or passage defects in the cable paths. Real-time analysis allows defects to be corrected immediately instead of having to plan a rework later. In case of doubt, the algorithm raises an alert and a technician makes the final decision, thus making it possible to solicit their expertise on complex cases only.
The neural networks trained by Deepomatic successfully recognize more than 75% of defects, a figure that is constantly increasing. Thanks to the Deepomatic platform, production performance improves and algorithms adapt to the appearance of new defects. The project initiated at the level of interventions on the points of mutualization (PM), but the application could cover all the different areas of intervention of the technicians.
- Comprehensive monitoring of FttH fiber optic installations
- Concentration of operational activities on high value-added tasks
- Evolution of control algorithms according to quality requirements