Bringing Humanity Back to Air Travels with the help of Big Data

 
The natural choice for us is to leverage cloud data stores that are optimized for elasticity and eliminate the computational constraints of legacy data stores and brick and mortar data centers. We’re leveraging databases that allow simultaneous online transactions and analytical capabilities—thus, simplifying our architecture and environments and enabling big data capabilities with a quicker time to market. In the world of data, designing schemas are a significant component of the development process which adds a considerable amount of overhead potentially causing severe roadblocks. JetBlue’s approach is to adopt schema-less databases coupled with the proper data governance and management. The cloud also provides us with the tools and technologies to build quicker data pipelines, with more robust and flexible analytical environments and offers predictive capabilities.

Simplicity and nimbleness drive the implementation of our data platforms. One of the biggest tasks we encountered while designing our data platform was to incorporate modern and open source software, a thinner stack, PaaS solutions and a single source of truth for both transactional analytical databases. Our platform is flexible enough to handle both structured and unstructured data in the form of pilot logs, customer feedback, customer support notes, and others. Though a lot of the transactional systems do not need the data to be tightly interlinked, it is evident that the data together can yield significantly more benefits than just individually. Coupled with a stable disaster recovery setup, it reduces our costs and allows for a well monitored and highly available platform.

At JetBlue, we’re aiming to eliminate the traditional IT-centered “data control” approach that creates roadblocks to data access for our users and renders them dependent on IT. This new approach seeks to eliminate the “data pools” that previously existed throughout most enterprises. We achieve this by standing up a controlled self-service platform so that data science groups across the organization can directly access the data which is controlled by layers of access control and use the tool of their choice for advanced analytics.

To speed up delivery times, we’re also changing the way we work. Like all businesses, our rapidly changing world is challenging IT to deliver quicker while minimizing our costs. We’re using agile software development techniques, continuous integration, and continuous deployment to speed our data delivery processes. Fundamental to our success is the proper adoption of agile methods from the onset and involves creating epics and prioritizing based on business value. Adopting agile methodologies in the data space has allowed our teams to focus on delivering business value faster and output that drives a better crewmember and customer experience. It also helps us mitigate the risk of building too much complexity into our solutions. In addition, we validate our code releases frequently using soft release processes to gauge early user feedback.

We know that our future competitiveness and ability to stay current, relevant and innovative depends on technology and specifically big data. At JetBlue, we want to leverage the speed and agility which is enabled by big data platforms in the cloud. That’s how IT at JetBlue is doing its part in bringing humanity back to air travel.