S7 Airlines, a oneworld® global airline alliance member, is the first Russian airline to complete the pilot project to develop predictive maintenance system for the Airbus A319 aircraft fleet. The plan is to connect the majority of S7 Airlines aircraft fleet to the system throughout the year.
The system allows making long-term predictions of the potential failures of each S7 Airlines aircraft based on the analysis of the historical datasets on aircrafts maintenance and individual components’ operation. The goal of system implementation is to reduce the number of flight delays caused by technical issues.
The system estimates the probability of various types of failures in the defined upcoming period. If this probability turns out to exceed the preset level, additional aircraft diagnostics is recommended.
«S7 Airlines, as the industry technology leader, applies the cutting-edge solutions to maximize the efficiency of the company. Innovative machine learning technologies allow bringing the flights safety at a new level. For instance, mathematical models can predict failures with a high degree of accuracy. Up-to-date innovations of aircraft manufacturers enable virtually instant collection of all aircraft systems functioning data right after landing. Coupled with many years of experience of S7 Technics team, this leads to substantial growth in aircraft maintenance efficiency», — states Pavel Voronin, CIO of S7 Group.
S7 Airlines and DATADVANCE, one of the leading players in predictive maintenance, jointly developed the software for data analysis and predictive modeling. S7 Technics team provides consultancy in the domain of aircrafts maintenance. As of today, the dataset consists of the data recorded by aircrafts telemetry systems, of the database belonging to S7 Technics maintenance and repair holding company, and of the meteorological data collected between the years 2012 and 2017.
Companies’ specialists are working on the system further development. The next step is to implement the production system with increased efficiency, enabling the online data export and equipped with a user-friendly interface.