Airports generate a vast and varied array of data encompassing flight operations, passenger details, security measures, infrastructure maintenance, and commercial activities. The scale of this data is immense, involving real-time tracking, surveillance footage, biometric information, financial transactions, and environmental monitoring. This complexity makes it nearly impossible to tackle the data to create operationally meaningful insights that improve operations for the benefit of the passengers, service providers, concessionaires, airlines, and airport staff.
Some of the commonly seen problems that traditional big data projects are struggling with and how to avoid these hurdles in an AI-first world that is rapidly changing how we can tackle these.
Data fragmentation
Data fragmentation happens because airports have multimodal data
Airports handle a variety of fragmented data systems and types. The exact number can vary significantly depending on the size and complexity of the airport, but it typically includes systems such as passenger service systems, baggage handling systems, security checkpoints, and more. Types of data can include structured data, like passenger itineraries, and unstructured data, like real-time video streams from camera systems or even audio streams from ATC or active communications.
Data fragmentation often arises when different departments within an airport adopt isolated solutions to solve their specific problems without a unified vision. This siloed approach leads to pockets of data that do not communicate with each other, creating barriers to comprehensive data analysis and effective decision-making.
Evaluate the extent of fragmentation with an inventory of data stores by data type (video, text, time-series)
To assess if data fragmentation is affecting your airport, conduct a thorough audit of existing systems and data flows. Look for duplicated data entries, inconsistencies in reports, and challenges in correlating data from different sources. The IATA recommends examining the time and resources spent on data reconciliation tasks versus the value generated, as this can indicate inefficiencies due to fragmentation.
Operational business challenges due to data fragmentation
The repercussions of fragmented data solutions extend beyond immediate operational inefficiencies. They complicate the decision-making process, hinder coordinated responses to disruptions, and lead to suboptimal resource allocation.
For instance, a report by Forbes titled ”Data tells us that a data culture matters” in 2020 highlights that siloed data environments can reduce operational efficiency by as much as 25%.
Visibility risk issues due to data fragmentation
Limited visibility across data silos can hinder an airport's ability to gain a holistic view of operations. This can impact everything from passenger flow management to security protocols, leading to inefficiencies and increased risks. Breaking down these silos and integrating data systems is crucial for enhancing operational visibility and ensuring more effective, data-driven decision-making.
Integration Layer Complexity
Traditional integration layer challenge
In the highly dynamic atmosphere of airports, data integration is not merely a backend task but a cornerstone for transformative operations. A clear integration layer simplifies the process of connecting disparate systems such as passenger service systems, baggage handling, and security checkpoints. However, this has typically been a multi-year, complex IT project that is often not set up for success, as requirements are constantly evolving before the project is even complete.
According to 2023 Air Transport IT Insights report, both airports and airlines saw IT spend increase year-on-year into 2023, reaching an estimated US$10.8 billion and US$34.5 billion respectively, with over two-thirds of airport and airline CIOs expecting continued growth into 2024.
Recognizing the need to mix different types of data: Structured and unstructured
Airports handle a multifaceted array of data - from structured formats in passenger itineraries to unstructured data like real-time video streams. Leveraging these diverse data types requires a flexible integration strategy that accommodates different data schemas, formats, and sources.
Promise of advanced AI models to eliminate the traditional integration layer’s problems
Avoiding the trap of multiple point solutions requires a strategic shift towards comprehensive platforms that can serve various operational needs under a unified umbrella. Unlike conventional methods, which often require extensive customization and complex middleware to ensure seamless data flow, advanced AI models can autonomously manage data integration processes. These models are capable of learning and adapting to various data sources, formats, and workflows, significantly reducing the need for manual intervention and specialized skills.
By leveraging machine learning, algorithms and natural language processing, AI can efficiently harmonize disparate data sets, ensuring real-time synchronization and consistency. This not only streamlines operations but also results in substantial cost savings, faster implementation times, and enhanced data accuracy, ultimately fostering a more agile and responsive data environment.
Who leads data initiatives? Business, technology or both?
Need for a broad, holistic approach to tear down silos
Aligning business and technology goals in airports requires a broad and comprehensive strategy. This approach ensures all stakeholders—from IT and operations to security and management—are on the same page regarding priorities and processes.
Cross-functional leadership engagement is key for an integrated outcome
The digital future of airports hinges on the seamless integration of diverse data streams to create a harmonized operational environment. Leadership needs to encourage cross-functional engagement and collaboration to achieve the most meaningful results. As the aviation landscape evolves, data interoperability and effective IT integration are paramount.
Importance of operational data representation
Merely integrating data is insufficient without proper airport context and representation.
Merely warehousing data without contextual representation and analytical layers adds unnecessary complexity. Effective operational data representation enables end-users to glean meaningful insights without additional decoding, thereby enhancing decision-making processes and actionable intelligence. For example, have you considered correlating your flight plan data with the arrival patterns of the passengers at security? This question is a crucial step towards generating insights for forecasting and improving operational planning.
Airports often overestimate the technological sophistication and willingness of their operational workforce to understand the nuances in the data.
Data warehousing does not solve the end user's airport operational problems
Data warehousing alone does not resolve the end user’s airport operational problems due to the complex nature of these contexts. Simply aggregating raw data without proper contextualization can lead to misinterpretation and ineffective actions. Effectively addressing operational nuances, such as real-time passenger flow or baggage tracking, necessitates the merging of digital insights with physical operations. Additionally, successful integration requires robust change management to ensure that personnel adapt to new systems and processes efficiently, facilitating accurate data-driven decisions and tangible improvements in airport operations.
Focus on organization data representation and change control NOT just the data lake purchase
Integrating an effective representation layer involves creating a semantic layer that translates raw data into understandable formats for end-users. Employing AI algorithms that consider operational contexts can transform raw data into predictive insights and proactive alerts avoiding extensive data mapping projects. Change control processes help manage alterations in data structures, workflows, and systems, maintaining data integrity and compliance with regulatory standards. By prioritizing these areas, airports can achieve a more coherent data strategy, enabling better resource management, improved passenger experiences, and enhanced operational resilience, all while avoiding the pitfalls of an underutilized or mismanaged data lake investment.
Evolve to think in terms flexible software, not fixed hardware
High capex initiatives in fixed sensors reduce flexibility and lead to increased capex spent year after year
Deploying advanced sensor technologies like 3D cameras and LIDAR solutions entails significant costs beyond just the sensor purchase cost. The ACI World 2021 report ACI Sensor Technologies discusses the high costs associated with adopting advanced sensor technologies and suggests strategies for cost-effective implementation. These include hardware expenses, installation, and maintenance, along with the need to integrate these sensors into existing IT ecosystems. Defining an outcome-based strategy for data aggregation can streamline the requirements and lead to efficiency gains in capex initiatives while maintaining flexibility to adapt to changes in the fast-paced airport environment.
This is traditionally done as part of a new terminal buildout, but that means your terminal is outdated 3 years after its built
Software was Eating the World, data hungry AI will now feast on it even faster
The phrase "software is eating the world" encapsulates the dramatic shift where software and digital technologies have transformed and disrupted traditional industries by streamlining operations, reducing costs, and enabling new business models.
Now, as we enter the era of data-hungry AI, this transformation is accelerating even faster. AI thrives on vast amounts of data to create actionable insights and drive decision-making processes. By integrating existing data sources such as CCTV systems, WiFi networks, and other sensors with advanced AI models, industries can achieve unparalleled efficiency and innovation. This convergence not only enhances operational capabilities but also maximizes ROI, highlighting that data-powered AI is poised to continue and even accelerate the revolution that software initiated.
Looking into the future
In conclusion, making airports future-ready requires tackling complex integration challenges, data fragmentation, and aligning business and technology goals. By strategically addressing these hurdles, airports can fully harness their data ecosystems to drive innovation and operational excellence.
In the past 18-24 months, AI has radically changed the environment for data processing by providing new, powerful industry-trained AI models that generate operationally meaningful insights, predictions and forecasts much more easily than traditional big data projects. Zensors AI, jointly with its AI initiative allies AWS and NVDIA, is making AI easily accessible for airports around the world with a pre-trained Airport Foundation AI model.
Zensors AI is pivotal in this transformation. Built for airport operational excellence, it learns and adapts to various data sources, formats, and workflows, reducing the need for manual intervention. This next-gen AI mode transcends physical airport limitations enabling operators to understand ‘what has happened’ ‘what is happening now’, and ‘what is predicted to happen in the future.’ It’s the purest airport-focused foundation AI that learns from your unique context and data sources to translate raw data into understandable insights for end-users.
%20(1)%20(1)%20(1)%20(1).webp)
%20(1).png)