Arriva’s data journey: building an enterprise Data Hub with AWS

This is a guest post by Neil Brown, Chief Technology Officer for Arriva Group.

Arriva, one of the leading providers of passenger transport in Europe, has long struggled to measure its business performance due to a lack of consistent and identifiable ways of sourcing data. Their businesses, including those in the UK, Denmark and Sweden, were lacking actionable intelligence and insights.

Arriva needed a step-change in their approach to data so that data was at the heart of their decision making. That meant unifying data, having a common data strategy, innovating and modernising data infrastructure and putting data into the hands of the decision makers.

Amazon Web Service, Inc. (AWS) Travel and Hospitality sat down with Neil Brown, Chief Technology Officer for Arriva Group for an insightful conversation about Arriva’s data journey.

What prohibitors did you have to overcome to kick-start the journey?

Four years ago, working with data was difficult at Arriva. We were lacking fit-for-purpose tools to manage the scale and the variety of data and the many purposes for which we wanted to use it.

To manage a bus and train company we needed: point-to-point data integration systems, performance management, governance, real-time location, visualisation and insight generation to facilitate decision-making, to name but a few.

For example, in the area of tenders and contracts, it was difficult for decision makers to see accurate up-to-date information in a single, easy-to-use place. This made it challenging for leaders to track the status of franchise tenders. There were also missed opportunities to optimize resources across different tenders and contracts due to a lack of consolidated data.

How did you plan your data transformation?

The CTO and central team created a blueprint for a new, modern data platform, then worked with AWS to hone the design before validating this with Gartner. The platform was a means of achieving seamless integration, creating a single set of data and driving consistency and compliance across business units.

At all stages the focus was on short, medium, and long-term return on investment (ROI). Our journey was need- and value-driven; the more comprehensive end-to-end solutions available were too expensive.

We made the decision to work with AWS Professional Services (and partners) to build a solution based on the AWS open source data.all framework. We call the solution our Data Hub.

What value did this deliver for your business and what changed compared to status quo?

The Data Hub’s ability to augment our base platform and allow big data ingestion and visualisation, helped us contextualise the ‘so-what.’ Access to that actionable intelligence was a game-changer for the business.

The tenders and contracts use case was one of the first ones implemented using the Data Hub. A key feature is that this has enabled fine-grained control over who can see data about tenders at different stages of their progress. Contracts are now managed in a consistent manner. Ways of responding to RFPs and contract bids are standardized—allowing the company to optimise the process.

Moving forward, how do you see this capability maturing and evolving in your business?

Initially the Data Hub datasets were based on financial or enterprise resource planning (ERP) data sources. We then moved through our understanding of commercial data and HR data. A future area of interest for us is to add sensor-related (Internet of Things [IoT]) data-feeds, engineering data, and external data sources obtained using third party data connectors.

Currently, Data Hub provides access to data lake and data warehouse types of data sets. Our vision is for it to be a centralised place for users to discover and request access to data provided in the widest possible variety of formats.

For example, we will add real-time streaming data and machine learning (ML)-powered data products to Data Hub. In the near future, a data user will be able to use Data Hub to search for a term such as “delays”. They would then be given a list of available options such as a data lake dataset of historical delays, a real-time stream of current delays, and an ML-powered data product API that predicts likelihood of future delays, and more.

Conclusion

It is definitely an exciting future for Arriva, with data playing a more central role than it has ever before. AWS Travel and Hospitality thanks Neil Brown for highlighting how data centralization and integration has been a part of Arriva’s data journey.

To learn more about how you can modernize your systems and solutions to better collect and analyse data, visit the AWS Solutions Library for Travel and Hospitality for Travel and Hospitality Data Mesh guidance or read Design a data mesh architecture using AWS Lake Formation and AWS Glue.

About Arriva
Arriva is one of the leading providers of passenger transport in Europe, employing over 39,000 people and delivering 1.2 billion passenger journeys a year across the Czech Republic, Croatia, Denmark, Hungary, Italy, the Netherlands, Poland, Serbia, Slovakia, Slovenia, Spain and the UK. Arriva’s services include local buses (scheduled and demand-responsive); inter-urban commuter coaches; local, regional and national train services; trams and light rail; waterbuses, and non-emergency patient transport.