I still remember the day I boarded my first flight ever – the idea of crossing such a huge distance in a very short time, the excitement of getting on a plane and getting to look down from the window up in the sky floating over the clouds was heightened by the nervous chill as we went through security and gate checks. In fact, I can safely say that the flight was the best part of my trip.
In today’s world, air travel is not the same anymore, and it is definitely not the best part of a trip. Customers are getting increasingly impatient and intolerant and are expecting airlines to be perfect at executing their operations, so they do not have to deal with anything that is not a security or a regulatory requirement imposed on them. This puts a tremendous pressure on the airline industry and challenges them to absolutely excel at their operations and delight their customers at a level that, in fact, more than makes up for all the inconveniences associated with going through all these security and regulatory processes imposed on air travelers.
The airline industry has been hard hit by the events of the last couple of decades – from 9/11 to the financial meltdown led recession to the recent pandemic. A slight delay in a single flight has a cascading effect on the airline – everything from gate availability, scheduling baggage handling, fuel or catering trucks to the crew itself – easily leads to delays across the board. To minimize the impact of these things airlines could maintain extra equipment or have larger time windows between flights, but this leads to additional operational costs. To save money and be competitive, airlines have to optimize. Imagine maintaining a superior level of operational efficiency and customer satisfaction while not knowing how much to staff for, how many resources to maintain and doing all of this while keeping an eye on shareholder value and corporate profits.
In the face of this uncertainty, airlines have started to leverage what they have and control – their data – to improve their customer experience, operational efficiency and shareholder value. Let’s look at some of the key airline operations and the data challenges associated with them.
Key Airline & Airport Operations
Crew Operations
Between union contracts and government regulations, airline crew has a very strict framework in terms flying hours. A small delay in a single flight’s landing creates a cascading effect on a series of downstream activities – crew availability validation for the next flight, jetway bridges, baggable handlers, fueling and other equipment availability, etc. All of this leads to deep, complex analysis of the fallout from that one event, massive data processing has to be done, decisions have to be made and alternate options have to be generated to ensure minimal customer impact and limiting the airline KPIs and scorecard damage, all while maintaining regulatory compliance. And to add to the complexity, all of this has to be done in real-time, because the crew cannot board the next flight unless they are cleared to fly, given that they just spent additional unscheduled time on the previous flight. So all of these logistics models or operational and compliance related rules have to be executed in real-time to ensure no further disruption in airline operations from that one delay of that one flight.
Air Traffic Control
As part of ensuring safe & uncongested movement of aircrafts, Air Traffic Control covers two key areas of flight movement – in air and on the ground. It also has a third component, that of flight data and clearance delivery. This third component ensures timely exchange of information for the safety and timeliness of air traffic. This third component also plays a very important role when airports switch to procedural controls during routine or unscheduled downtime for the key radar-based air traffic control systems. So, two key areas of focus for such systems are:
Minimizing the downtime of the systems, and
Ensuring that a lot of historical and close to real-time data is available to support instantaneous critical decision making for aircrafts in the air and on ground moving between the runway and the gates.
Airport Operations
Knowing what is happening at the airport at any time is critical for smooth flow of traffic through the airport. Everything from lines at security checkpoints to the backlog on the conveyer belts to gate traffic to equipment movement impacts airport security, service delivery and consumer experience. Reaction to any unexpected event, whether it is a longer delay at a gate before boarding or delays in scanning baggage tags, etc. first requires full visibility of all operational data, and then the analysis of such data in real time. Getting that full visibility of all events across the airport means collecting data from various sensors, events, and technology systems across the airport operational technology stack. And then once data across the airport from various systems is available, business rules can be applied, and alerts can be raised to take corrective actions.
Customer Experience
Air travel, as a mode of transportation, has limited competition. However, amongst themselves, airlines are working hard to win consumer business – as our flight attendants admit, when it comes to flying, we have a number of options. Southwest successfully proved that there are ways for an airline to win consumers by providing them a different air travel experience. A big differentiator for air travelers is customer experience. Ubiquitous features like online booking and seat selection to more complex things like creating alternate options or accommodations in case of delayed flights, especially for loyalty program customers, all require information like flight status, connecting flights and gates, baggage movement status, customer loyalty status, etc. in real-time. From a customer safety & security perspective, knowing what flights have minors or senior citizens traveling who may require escorts or wheelchair access, etc. is critical as well. For airlines to manage all this and provide superior customer experience, they have to jointly process a variety of information available in different airline and airport management systems.
The Airline Data Challenge
Looking across the key airline and airport operations and the data needs to successfully execute these functions, we see a common set of data patterns –
New events and associated data is being constantly generated, whether it is from flight status to changes in local weather to movement of people, their bags or capital assets, etc.
Valuable information is spread across several legacy and modern data silos.
For a thorough analysis of events being generated, and to make optimal decisions, events must be combined with historical data.
Making real-time and historical data available is a great start, but processing these events within the appropriate SLA is critical, and some of this processing can be very complex, as we saw in the example of crew scheduling.
More importantly, all these data generation and processing patterns together apply to each one of those operational areas. It’s not that we can address crew management challenges by just analyzing streaming events or provide a superior customer experience by simply accessing data across silos. Each problem domain requires simultaneous addressing of these data challenges. And not only do these different data problems have to be addressed, they have to be addressed in a time window that is acceptable to provide the necessary value. There is no point in the flight attendant calling the airport staff after landing to tell them that appropriate personnel are needed to escort a minor to their next gate or to assist someone with health challenges.
A data solution that combines these different facets of the problem domain is needed to meet the demands of the airline industry. Let’s take a look at how GridGain’s Unified Real Time Data Platform addresses these needs and is the data platform of choice of the world’s leading airlines.
GridGain’s Unified Real-Time Data Platform
A unified real-time data platform combines real-time event streaming data and historical data with complex computing capabilities, along with durable data storage to provide enterprises with a unified platform that minimizes network latency between data stores, simplifies the data architecture and eliminates data inconsistencies within an enterprise data ecosystem.
Core Platform Capabilities
Combines data in-motion and at-rest: GridGain combines streaming event data with historical data at rest to enrich and provide relevant context to the event being processed.
Performs complex computations: GridGain executes complex business rules, mathematical or AI/ML models on the combined contextual data.
Provides durable data storage: Whether data is needed at ultra-low latency or is stored for longer term analytical needs, GridGain provides scalable, low-latency, in-memory data store, with a disk-based persistence layer for additional durability.
APIs for interaction with your data: GridGain supports APIs for all major development stacks, including full ANSI-99 SQL, with ACID compliance and immediate or eventual consistency with the persistent store.
Scalability, high availability & reliability: With its distributed architecture, data center replication, enhanced security, snapshots and flexible point-in-time recovery, GridGain provides its customers with a secure, scalable and reliable data processing platform.
Key differentiators
There are other platforms that also provide some of the capabilities that GridGain offers. A few things separate GridGain from other such technologies.
True unified real-time platform: GridGain combines streaming, transactional and analytical data processing with data aggregation, manipulation, or computing capabilities in a single platform, all synchronously, without the overhead of added latency caused by the need for data persistence.
Distributed computing: With its distributed architecture, supporting both row-based and columnar-based processing, along with various data models, data types and formats, GridGain enables massive scale with high-performance distributed, parallel processing of data.
Memory-first architecture: GridGain’s unique memory-first architecture allows enterprises to distinguish and manage various levels of data processing SLAs from ultra-low latencies to milli-second SLAs to longer processing times for after-the-fact analytics, all with simple configurations. This also gives enterprises control on their infrastructure costs and reduces their TCO.
Conclusion
Airlines across the world are looking at their data to provide their customers a superior and differentiated experience, minimize their operational costs and improve their market share. GridGain’s unified real-time data platform is trusted by leading airlines to provide them with a ultra-low-latency data processing problems for their multi-dimensional data processing needs across all their business and operational functions.