Data Science boosts performance in the aero industry

22nd November 2019

LINCOLN, a consulting firm specializing in business intelligence, big data, and data science, combines technical and functional know-how to respond to the needs of its customers. This ALTEN subsidiary specializes in data, maintaining a sharp focus on data processing and analysis. LINCOLN gathers, transforms, models, and visualizes data so that businesses can extract actionable knowledge. We sat down with LINCOLN’s Innovation Director Dorothée Delaunay, and Development Director Maxime Boursin.

What makes LINCOLN different on the data science market?

We have more than 30 years of experience in data, and we have built a reputation for our know-how across the entire data-processing value chain. Our added value is in our business-oriented approach and in our capacity to deliver technical, methodological, and functional solutions. For us, the end goal of any project is to transform data to move our customer’s business forward. The return on investment has to be measurable.

Our Innovation Division and Consulting Division are there to make sure that we are constantly increasing the value we deliver.
The Innovation Division focuses on topics like anomaly detection and recommendation engine development—both of which are relevant to the aero industry. Our Lab develops AI algorithms that can handle all kinds of data, from images and video to IoT and text. All of the Lab’s R&D projects have operational goals. The Consulting Division draws on its extensive experience to bring end-to-end solutions to help companies organize and extract value from their data. It also helps companies develop data-based strategies to improve their decision making processes.

What does the value you deliver look like in the aero industry specifically? What are the challenges?

First of all, when we talk about the aero industry, we are talking about all segments. This includes aeronautics, of course, but also airports.
Data science can help with at least three issues we feel are strategic to the industry: anomaly detection, mobility within airports, and digitization.
Some of the projects we have completed in the industry give a good picture of what we can do.

Where is the use of data likely to grow in the industry?

Over the past several years, we have been using more and more data that is getting bigger and bigger in volume, from images, audio, and video to IoT and text. The next big challenge is to develop interpretable artificial intelligence algorithms that will be able to extract even more value from all of this data.

“The digital revolution is just beginning. The advent of IoT and text and image processing should allow us to improve maintenance and security. Digital technology will also help streamline costly processes and speed up decision making.”

LINCOLN Use Cases

Predictive Maintenance

LINCOLN developed methodological approaches for the detection of anomalies on avionics components for Airbus and created algorithms based on massive data from thousands of sensors that constantly measured various parameters as time series.

LINCOLN’s innovative solution combined statistical analysis, machine learning , and deep learning to identify anomalies and determine when and where the anomalies occurred.

Data science will be used in addition to aeronautics engineering to bring real benefits in several areas:

Reducing aircraft downtime through more accurate troubleshooting
Limiting “overmaintenance” by more effectively achieving “just-in-time” performance
Lowering costs resulting from equipment breakdowns and shutdowns

Data

Digital transformation will shape the future of many companies, and a data strategy is crucial to any digital transformation project. LINCOLN’s digital transformation services go from process automation and tools through to data monetization. And, to effectively deliver these services, we first have to “work” the data. We need to gather data, store it, process it, and analyze it to develop artificial intelligence or another relevant solution.

One example is the algorithm we developed for Airbus to automatically archive technical documentation. The algorithm processes text and images to categorize hundreds of thousands of documents based on their content, without ever opening the documents.

More generally, we help our customers improve productivity by:

Freeing employees from repetitive tasks through RPA
Creating “bots” that can automate operational tasks
Digitizing reporting for more effective management and simpler processes

Airport Mobility

The capacity to dynamically manage traffic inside airports is a major airport operations and maintenance challenge. The goal is to improve service quality in terms of capacity, the variety of transportation services, on-time performance, comfort, security, and shopping and other services.

Our big data science know-how encompasses modelling and real-time visualization and has enabled us to address train station use cases that are completely relevant to airport situations:

Analysis of traveler traffic to identify bottlenecks
Insights into traveler routes and high-traffic areas for more effective design of spaces
Impact of signage on traveler behavior
Impact of equipment failures (doors, escalators, etc.) and repairs or other work

We were able to use large volumes of information from heterogeneous sources like Wi-Fi, GPS, laser sensors, video feeds, and display and booking data to gain a deeper understanding of the issues.
In addition, real-time data analysis can power a personalized traveler experience, with functions like real-time information sent right to travelers’ phones to help them find their gates and help keep flights running on time.