Explainable Deep Sequential Network for Flexible Airport Passenger Flow Estimation PhD

Updated: about 2 months ago
Location: Cranfield, ENGLAND

This is an exciting opportunity for a fully funded PhD studentship in the Centre for Autonomous and Cyber-Physical Systems at Cranfield University, to develop explainable deep sequential networks for flexible airport passenger flow estimation. This PhD investigates deep time-series models and their explainability for multi-range passenger flow estimation. This research is sponsored by EPSRC and SAAB UK under the Doctoral Training Partnership Funding 2021/22. The studentship will provide a bursaries contributed by EPSRC and Saab, plus fees* for three years.

Accurately determining Airport passenger Flow (APF) is critical to efficiently manage airport processes and allocate resources and is fundamental to the provision of a seamless passenger journey experience. Precise APF estimation enables efficient resources assignment and planning for airport processes and functions, which will become essential to accommodate the dramatic increase that is expected in international air traffic in the next decade. However, due to the nature of air traffic, in which uncertainty and latency is a common feature, the temporal characteristic of APF has high volatility and irregularity properties, which make the APF very difficult to estimate and forecast. Therefore, an advanced framework for accurate APF estimation that considers the prediction explanation and multiple temporal behaviours is needed to optimise airport processes and resources, which in turn will enable a better integration with air traffic management processes and will result into a better passenger experience.

Currently, the estimation for APF have several limitations. One major limitation is the lack of explainable ability for the APF estimation model. An explainable network will allow the design of a transparency network that helps to clearly identify reasons behind the bottlenecks in airport operations and resources that relate to passenger flow, by providing a traceable inference process that enables effective decision-making to optimise operations. Another limitation is the lack of decoding for the relationship between long-term and short-term APF patterns as the long-term behaviour of APF. Hence, flexible APF estimation considering both long-term and short-term behaviour with sufficient explanation of the network prediction with be studied in this project.

Cranfield is an exclusively postgraduate university that is a global leader for education and transformational research in technology and management. Cranfield ranked 27th in the QS 2022 international university ranking in the theme of Mechanical, Aerospace, and Manufacturing. It is the only University that has its own commercial Airport , controllers, commercial pilots and aircraft . Cranfield Airport was the first to install and operate a Digital Tower in the UK, supplied by SAAB the PhD industrial sponsor.

This PhD will be hosted by the Centre for Autonomous and Cyber-Physical Systems and will be based at DARTeC which is a £67 million new research centre that will enable “Aviation of the Future”. The Centre for Autonomous and Cyber-Physical Systems is one of the world’s largest centres of postgraduate education and research, with over 200 MSc and PhD students. In terms of facility, Cranfield University has a range of specialist research facilities available for different research activities (e.g. MUEAVI-multi-user environment for autonomous vehicle innovation facility). The facility operates as a collaborative and flexible space with specialist equipment available for indoor/outdoor flight tests for UAS systems. Also, the Centre for Autonomous and Cyber-Physical Systems offers the environment for algorithm development and simulation. Also, the centre can offer support, assistance with analysis, and method development for research.

The PhD will demonstrate Explainable Deep Learning (EDL) methods for time-series data modelling with the particular focus on flexible airport passenger flow prediction. The research outcomes will have significant opportunity to accelerate the development of explainable AI and smart airport.

You will have great opportunity to work with world leading academics, scientists, and industrial partner. You will be encouraged and supported in publishing your own work in high quality peer-reviewed journals. Also, you will have opportunities and supports to present your work at relevant flagship international conferences. Working with Saab Technologies UK, the research results will be integrated into the Saab SAFE Airport system for real world testing and implementation.

The experience of being to work on cutting edge explainable AI and time-series modelling and forecasting will be attractive in both academic and industrial world, especially with some of our leading industrial partners. This is a very exciting project for a suitable candidate where you will be exposed to the latest technological developments, learn from the industrial and academic experts working in this area and prepare for an exciting career in either academia or industry.

Entry requirements

Applicants must have a first or second-class UK honours degree or equivalent in a related disciplines such as computer science, mathematics, aerospace engineering, and data science area. This project would suit someone with a strong background in machine learning (both theoretical studies and application), and autonomous systems, and hands-on approach to systems integration and out of the box thinking ability.

The candidate should be self-motivated and have excellent analytical, reporting, and communication skills as well.


This studentship opportunity is open to applicants in the UK, EU and International. There are no restrictions on nationality. The PhD is fully funded for UK applicants and partially funded for EU and International applicants.

Sponsored by EPSRC, Cranfield University, and SAAB UK, this studentship will provide a bursary of up to £18,000 (tax free) plus fees at UKRI level for three years.

Please contact us for further information.

View or Apply

Similar Positions