EURO 2025 Leeds
Abstract Submission

2257. Navigating Turbulence: Near Real Time Rescheduling of Crew Rosters in Short-Haul Aviation

Invited abstract in session WB-12: Workforce scheduling, stream Scheduling and Project Management.

Wednesday, 10:30-12:00
Room: Clarendon SR 1.02

Authors (first author is the speaker)

1. Louisa Sober
Datasparq
2. Ruth Walton
Data Science, Datasparq
3. Michael Sutherland
Datasparq

Abstract

In the aviation industry, any disruptions to crew rosters can cause a cascade of knock-on effects which impact both crew and scheduled flights. In order to navigate the many regulations and operational constraints, highly trained crewing teams are required to rapidly reconfigure the shifts to minimise disruptions. During busy periods the volume of disruptions can overwhelm crewing teams, potentially resulting in regulatory violations, further delays, cancellations and even fines.

In early 2024 we were tasked with developing and deploying a crew rescheduling tool for a short-haul airline to support the crewing team during the upcoming busy summer period. Within a rapid 5 month timeline, we progressed through a full project pipeline from initial scoping to a fully deployed user interface that updated in near real time and which has been adopted into daily operations.

This talk will look into the metaheuristic approach that formed the core of our solution which facilitated rapid and intelligent roster reconfiguration. We will also discuss the inclusion of an explainability layer which provided the rationale behind these recommendations and enhanced user understanding and adoption.

In addition we will address the essential supporting elements that ensured the adoption of the solution. This includes strong stakeholder engagement, implementation strategies and a cross-functional development team that allowed us to fully leverage cloud computation.

Keywords

Status: accepted

Back to the list of papers