# Identifying £2.2M in Revenue Leakage for Global Airport Services

> A global airport services operator · Industrial & Aviation · QuantSpark Labs

A global airport services operator needed to identify uncaptured services and revenue leakage. QuantSpark developed an Anomaly Detection Model, identifying £2-2.2m in potential annual revenue.

## At a glance

- **£2.2M** Potential Revenue Identified

## What was the problem?

A global airport services operator had developed a mobile application to capture services, which had already led to 28,000 additional services being captured globally compared to their previous paper-based process. Despite this improvement, the client faced the challenge of determining if there were still more services not being captured and, consequently, if revenue was being missed due to these unrecorded services.

## What did QuantSpark do?

QuantSpark developed a sophisticated Anomaly Detection Model to identify potential revenue leakage from missing services. The solution focused on UK&I data from 2025 FSC data and involved several key components:

*   **Workshops and Stakeholder Engagement**: We conducted a series of workshops with key stakeholders to gather essential input, build awareness and advocacy for the solution, and ensure active engagement throughout the journey.
*   **Impact Scope Filter – Focused Data**: The initial step involved narrowing the scope of all UK 2025 flights to focus on three specific customers and three stations, ensuring a targeted and manageable dataset for analysis.
*   **Machine Learning (ML) Layer**: Supervised ML models were employed to identify complex patterns beyond simple manual rules. This layer enabled the detection of potential missing services across eight different service types, with thresholds carefully optimised to maximise the capture of true positives.
*   **Rules Layer**: Agreed business rules and contractual exceptions were applied to minimise false positives. This involved validating service type, aircraft type, timing constraints, and operational feasibility.
*   **Output and Prioritisation**: The model assigned priority and confidence bands based on both the machine learning outputs and rule-based logic. These results were then presented in intuitive dashboards to support actionable insights and facilitate retro-billing processes.

## What changed?

The Anomaly Detection Model successfully identified significant potential revenue leakage for the global airport services operator:

*   **Potential Revenue Identified**: The model identified an estimated **£2 – 2.2 million** of potential annual revenue in UK&I that was previously being missed.
*   **Missing Services**: Approximately **5%** of estimated flights were found to have missing services.
*   **Operational Insight**: The analysis was based on 68,000 annual flights serviced, providing a comprehensive view of potential uncaptured services.

---

Canonical page: https://quantspark.ai/case-studies/identifying-2-2m-in-revenue-leakage-for-global-airport-services
More about QuantSpark: https://quantspark.ai/llms.txt