Exploit the full value of aerospace field data

Do you work with aerospace systems and collect large quantities of data? Do you manage fleets’ reliability and failure reports? Would you like to exploit the full value of your data to optimize your maintenance operations?

With the progressive introduction of a variety of sensors and Health and Usage Monitoring Systems (HUMS) in aeronautic products, data gathering has increased dramatically in the latest years for aircraft manufacturers and operators. However, the value of the data is only materialized when it can be incorporated in the decision-making paths for daily operations.

At DMD Solutions we are striving to fully exploit the potential of aerospace operations data to improve aircraft safety & reliability, systems availability and cost of maintenance programs.

Predictive Maintenance for smarter aerospace operations

Predictive Maintenance aims to apply Artificial Intelligence methods to predict the time of failure for all types of systems, minimizing downtime and maximizing equipment availability and life span. We are implementing a bold approach that would allow for a flexibilization of the standard maintenance schedules applied nowadays in the aerospace industry.

If you are handling datasets from operations in the aerospace industry such as HUMS records, system sensor data or field reports and are able to combine this data with the time of failure for aerospace products you are a good candidate to contribute. We offer to partner up to process this data using Janus to obtain accurate prognosis of system failures.

HUMS data ai
Data collection from HUMS system

What is Janus and how does it work?

DMD Solutions has developed an algorithm named Janus which is able to predict, under certain conditions, the approximate flight hours until the next failure for aircraft and/or aircraft systems.

The algorithm is based on neural networks, as this methodology allows the flexibility to work with a variety of data types with very accurate results. Just like human brains, neural networks can learn from the past, detecting repeated patterns, and adapt to new observations with the help of previous knowledge. Janus was trained with basic information about flights (aircraft type, born date, flight hours, etc.) and the meteorological conditions to which the aircraft was subjected. The initial results obtained were notable considering the first dataset was based only on external conditions in their interaction with aircraft systems.

Neural network AI aerospace
Neural Network

With further algorithm training using system specific data to feed Janus we will reach better results making predictive maintenance for aircraft an applicable reality in the near future.


Who is supporting this project?

Janus has risen interest in several companies and institutions. The project was awarded with the Red.es support grant for Artificial Intelligence projects from the Spanish Ministry with funds from Next Generation EU. The goal of the project is to further develop the Janus prototype to bring it to TRL 7 or 8, for which it is essential to count on an operative environment where to put its result to test.

Predictive Maintenance AI
AI Support from Red.es is funded by NextGenerationEU

We are building up the research team, which includes as of today aircraft manufacturers that are involved in the training and testing of the algorithm, providing field data from sensors and recording devices in aerospace operations. One of them is Dronamics, a UAV manufacturer at the prototyping phase performing flight tests with an innovative unmanned CS-23 type aircraft. We also collect data from several airport operations in Catalonia and Bavaria.

Join the research team!

In addition to our current partners, DMD Solutions is looking for companies to cooperate in this research project in the field of Predictive Maintenance. 

If your company is interested in putting their field data to work to achieve operative results in predictive maintenance, we’d be glad to cooperate, bring our expertise and partake in the results. Paths for collaboration include several possibilities such as cession of data or joint research team. Please contact the project coordinator Lovejinder Singh below to know more.


Aeronautical RAMS Engineer

Lovejinder Singh

Lovejinder Singh has a bachelor's in Air Navigation and a Master's Degree in Aerospace Engineering (Major in Vehicles) from the Polytechnic University of Catalonia. He is currently working in DMD Solutions focused on the sectors of Reliability (FRACAS, RPA, FMEA) and Maintainability (MSG-3, AMM). His thesis verses on Predictive Maintenance Deep Learning methodologies.