Optimizing air travel with efficient airport flow
Faster data processing, from 80 hours to 4 hours
Fewer labor costs with automation and centralized simplicity
More accurate flight-level error forecasting, from 30% to 10%
Heathrow Airport aims to be an extraordinary and future-ready travel hub while welcoming over 82 million passengers and handling over 460,000 flights annually. Previously reliant on spreadsheets for passenger predictions, it faced expensive delays, labor demands and accuracy issues. Using the Databricks Data Intelligence Platform, the airport now processes forecasting data in just hours, not weeks. In a year, Heathrow drastically modernized its forecasting model, reduced errors and costs, and gained insights for AI expansion. Today, the airport continues to advance data and analytics to ensure customer satisfaction at every touchpoint.
Legacy tools struggle to keep pace with enterprise scale and speed
As one of the world’s best-connected airports, Heathrow relies on over 500GB of data weekly to make critical real-time decisions. Powering over 1,300 daily flights involves data from passengers, flights, airlines, seasonal performance, and traffic and weather patterns. Unfortunately, legacy data tools scattered and siloed the data, slowing processing and productivity, increasing costs, and compromising accuracy.
Without distributed computational power and a high-performing data platform, Heathrow’s forecasting model required two weeks and two people to manage. Eduardo Teixeira Garrido Junior, Tactical Forecast Manager at Heathrow, emphasized the need for a digital data strategy. “We’re trying to improve data processing enough that we can focus on other things. Previously, we could see gaps in the models because they couldn’t observe much data. We were pulling a lot of data, but we weren’t producing much.” Moving from a simple, code-based spreadsheet to an intelligent forecasting model was necessary, but the airport felt limited by the market.
To realize Heathrow’s future-based vision for excellence, it needed a data intelligence platform that enables various skill sets and scripting languages. To refine its data approach, Heathrow also needed tools for user education, data governance and security, and ML model training. With these built-in support systems, Heathrow could build a foundation for expanding data maturity with faster data, more insightful analyses and fewer margins of error to forecast and influence passenger flow.
Streamlining data processing on the Databricks Data Intelligence Platform
After comparing various platforms, Heathrow Airport centralized its data and analytics on the Databricks Data Intelligence Platform on Azure. The platform’s integrated tools, easy-to-manage foundation, and user experience laid the base for continuing opportunities to innovate and elaborate.
Previously stalled by insufficient and unscalable computing power, Heathrow now has the computer strength required for advanced AI and ML use cases. Starting with its passenger flow forecasting model, Heathrow leveraged the consolidated and business-agnostic platform to serve different business users with new data insights.
Ross Flannigan, Product Manager of Data and Technology at Heathrow Airport, said, “When performing a big job or model training, we can throw a lot at that cloud computation, depending on the job, then wind it down accordingly. These remote computations are important to streamline model training across the airport business functions and teams.” Moreover, the airport can scale quickly, accepting different scripting languages and allowing high-level forecasts to be distributed downstream in easy-to-digest, self-serve analytics.
Not only does this new capability speed time to insights, but it also fosters cross-team collaboration with shareable code and insights via team-based notebooks. Jump-starting internal adoption and use, Heathrow’s users are capitalizing on the self-paced training and certification available through Databricks Academy. The airport is also implementing data governance with Databricks Unity Catalog to secure data and AI asset collaboration within a single permission model. Furthermore, Heathrow integrated Power BI with Databricks to ingest data from the lakehouse into dashboards and visualizations for business reporting — thereby furthering safe and collaborative data use across the airport.
Scaling passenger flow use cases drives customer satisfaction
Since moving to the Databricks Data Intelligence Platform a year ago, Heathrow Airport has been improving customer satisfaction by optimizing passenger flow within varying areas. Heathrow sped forecast insights from two weeks and two people to four hours and one person while decreasing the margin of error from 30% to 10%.
Airport customer satisfaction is directly impacted by passenger wait times. If a passenger is in line for longer than five minutes, Heathrow considers it a breach of customer experience. With the accurate, automated and fast forecasting model in Databricks, the airport can plan predictive maintenance, cleaning, and service interruptions during slow times and days. It prepares teams for peak travel periods so they can proactively respond with efficient passenger flow, thus putting customer satisfaction at the forefront. Flannigan said, “Our strategy with Databricks was to democratize data, bring it into one place and make it easier to access. We’re doing that, but what we took for granted was the power of collaboration in achieving these goals.”
The adoption and excitement around the platform are laying a path to continually achieve the airport’s goal of being extraordinary. Heathrow is recording every trained model on MLflow with plans to expand, looking at Databricks Delta Sharing for data sharing with airlines, ground handlers and other Heathrow companies, and is in the early stages of establishing a center of excellence to support the ongoing strategy.
Eduardo closed by saying, “Databricks enables us to provide a more efficient and accurate forecast than we’ve ever been able to before. Now, passengers and stakeholders get a greater level of service in a more efficient airport.”
