CIFRE-PhD - Flight & Integration Tests – Machine learning for time series analysis (m/f)

Description de l'offre

CIFRE-PhD - Flight & Integration Tests – Machine learning for time series analysis (m/f)

Airbus Toulouse

Airbus is a leading aircraft manufacturer with the most modern and comprehensive family of airliners on the market, ranging in capacity from 100 to more than 500 seats. Airbus champions innovative technologies and offers some of the world's most fuel efficient and quiet aircraft. Airbus has sold over 13.800 aircraft to more than 360 customers worldwide. Airbus has achieved more than 8,000 deliveries since the first Airbus aircraft entered into service. Headquartered in Toulouse, France.

Airbus is a global leader in aeronautics, space and related services. In 2016, it generated revenues of € 67 billion and employed a workforce of around 134,000. Airbus offers the most comprehensive range of passenger airliners from 100 to more than 600 seats. Airbus is also a European leader providing tanker, combat, transport and mission aircraft, as well as Europe's number one space enterprise and the world's second largest space business. In helicopters, Airbus provides the most efficient civil and military rotorcraft solutions worldwide.

Our people work with passion and determination to make the world a more connected, safer and smarter place. Taking pride in our work, we draw on each other's expertise and experience to achieve excellence. Our diversity and teamwork culture propel us to accomplish the extraordinary - on the ground, in the sky and in space.

Description of the job

A vacancy for a CIFRE-PhD - Flight & Integration Tests – Machine learning for time series analysis (m/f) has arisen within Airbus in Toulouse. You will join the EVIDA Department.

Aircrafts generated data are more and more accessible, from product development to airline in service operations.
Current data analysis approaches are increasingly generating workload on our operations while addressing a few parts of all the value we could get or expect.
We consider new data science techniques capabilities as the enabler to fully leverage our data value for our products maturity and operations efficiency.
In such a context, within the Flight and Integration Tests Centre, we are recruiting a CIFRE-PhD to address the usage of machine learning techniques for time series analysis.
The aim of this thesis is to enhance the use of the machine and deep learning algorithms on aircraft sensors and systems time series, taking both benefits from the sensors and parameters diversity on one hand, and from the physical relationship that may exists between them on the other hand.
We are currently processing these time series by addressing problems independently, meaning to reduce the sensors data up to features we inject in machine learning algorithms built from our knowledge of the physical phenomena. Our goal is to build a model-based system that will automatically sum-up the time series by maintaining the maximum information and adapting the features extraction to the time series itself.
You will be requested to go beyond the simple use of statistical and mathematical methods. Thus, you shall create a new way of modelling time series to highlight the systems' normal or abnormal functioning.

Profil recherché

Tasks & accountabilities

Your main tasks and responsibilities will include:

·  Building automatic models based on statistical and mathematical aspects of the data (univariate and multivariate time series) thus :
·  Identifying and classifying aircraft related physical phenomena based on their normality of abnormality criteria
·  Generating prediction models to help the Tests Specialist identify the likelihood of tests conditions
·  Doing the state of the art of the existing techniques used to process sensors and systems data (special focus on time series)
·  Developing mathematical approaches characterizing physical phenomena or aircraft functions
·  Identifying weirdness in time series or singularity of tests conditions
·  Developing a mock-up to run created algorithms in a big data environment (with scalability

Required skills

·  Educated to a 5 years' Engineering or pre-doctorate degree level (or equivalent) in Applied mathematics /statistics
·  Highly developed mathematical and statistical skills are a prerequisite
·  Comprehensive understanding of machine learning techniques and algorithms
·  Python Development skills are a must
·  You should be able to develop your own creativity and team spirit
·  Great communication skills and ability to work in transnational environments
·  Agile & open-minded about innovative & emerging technologies
·  Rather pragmatic, you can also work autonomously
·  Negotiation Level in English is required & advanced level in French is desirable