Efficient baggage tracking systems are indispensable in the aviation industry and help to provide timely and intact delivery of passengers’ belongings. Baggage handling and tracking errors can trigger a chain of complications, from flight delays and missed connections to lost luggage and dissatisfied customers. Such disruptions tarnish the airline’s reputation and can result in significant financial losses. Consequently, airlines devote substantial resources to develop and deploy accurate, efficient, and reliable baggage tracking systems. These systems help to improve customer satisfaction through near real-time bag location updates and optimize operational workflows to support punctual departures. The critical role of a baggage tracking system is evident in its ability to effectively track packages, digitize operations, and streamline corrective actions through re-routing triggers.
In this blog post, we discuss a framework that IBM created to modernize a traditional baggage tracking system using AWS Internet of Things (AWS IoT) services and Amazon Managed Streaming for Apache Kafka (Amazon MSK) that aligns with the airline industry’s evolving requirements. Before discussing the solution’s architecture, let’s discuss the traditional baggage tracking process and why there’s a need to modernize.
Traditional baggage tracking process
The baggage tracking system involves manual and automated barcode-based scans to monitor how checked luggage moves within an airline and airport infrastructure. The baggage tracking system can be subdivided into capabilities, as depicted in Figure 1, to support the products and services that airlines offer.
Figure 1: High-level baggage tracking capabilities
Baggage tracking starts with the customer check-in and progresses through several stages. At check-in, luggage is tagged and associated with the passenger using a barcode or radio-frequency identification (RFID) technology. Then the luggage gets sorted and routed to the right pier or a bag station. Sorting gateways communicate with backend systems using protocols such as TCP/IP, HTTP, or proprietary messaging protocols. The luggage then goes through bag rooms where they are stored and then pier areas where they are loaded onto the flight by the airport staff. In some cases, luggage is sorted into containers inside the flight.
When the flight arrives at the destination, luggage is offloaded from the flight and routed to the baggage claim area or onto the next flight. Unclaimed luggage is then routed to the baggage service office area, as necessary. Throughout this process, luggage is scanned at every stage for accurate and near real-time tracking. If luggage is mishandled or misplaced at any stage, tracking information becomes vital to recover the luggage.
Figure 2: Traditional baggage tracking architecture
As depicted in Figure 2, the traditional baggage tracking architecture relies extensively on application programming interfaces (APIs), which are commonly implemented using either the REST framework or SOAP protocols. Since most airlines leverage a mainframe as the backend, using APIs follows two primary pathways: direct data transmission to the mainframe or an update to a relational database.
A distinct offline process retrieves and processes the data before sending it to the mainframe through other APIs or message queues (MQ). If device information is received, it’s typically limited and may require another background process to orchestrate additional calls to transmit the information to the mainframe.
This entails manual interventions which may result in potential service disruptions during the failover periods.
The need to modernize
A traditional baggage tracking system is significantly hindered by several critical business and technical challenges.
- Inability to scale with the high volume of baggage tracking data and telemetry for on-site and on-premises infrastructure.
- Challenges in handling sudden bursts of data volume during irregular operations (IROPS).
- Connectivity concerns in airports, such as bag rooms, claim areas, pier areas, and departure scanning.
- Lack of required resilience for mission-critical systems affecting continuity.
- Inability to quickly adapt to changing baggage tracking regulatory requirements related to mobility devices.
- Integration with systems like kiosks, sortation gateways, self-service bag drops, belt loaders, fixed readers, array devices, and IoT devices for comprehensive tracking and data collection.
- Latency concerns for global operators affecting operational efficiency and passenger experience.
- Lack of monitoring and maintenance for tracking devices potentially leading to operational disruptions and downtime.
- Cybersecurity threats and data privacy concerns.
- Absence of near real-time insights of baggage tracking data. This hinders informed decision-making and operational optimization.
Modernizing the baggage tracking system is crucial for airlines to address these issues, supporting scalability, reliability, and security while improving operational efficiency and passenger satisfaction. Embracing advanced technologies will position airlines to stay competitive and support growth in a rapidly evolving industry.
The solution
Figure 3 depicts a solution to the challenges in the traditional baggage tracking process.
Figure 3: Baggage tracking cloud solution architecture
Devices like scanners, belt loaders, and sensors communicate with their respective device gateways. These gateways then connect and communicate with the AWS cloud through AWS IoT Core and the MQTT protocol for efficient communication and telemetry. This design uses MQTT because it can provide optimal performance, particularly in environments with limited network bandwidth and connectivity.
The AWS IoT Greengrass edge gateways support on-site messaging for inter-device and system communications, local data processing, and data caching at the edge. This approach improves resilience, network latency, and connectivity. These gateways provide an MQTT broker for local communication, and sending required data and telemetry to the cloud.
AWS IoT Core is particularly useful in scenarios where reliable data delivery is more critical than time-sensitive delivery to backend systems. In addition, it offers features like the device shadow that allows downstream systems to interact with a virtual representation of the devices even when they are disconnected. When the devices regain their connection, the device shadow synchronizes any pending updates. This process resolves issues with intermittent connectivity.
The AWS IoT rules engine can send the data to required destinations like AWS Lambda, Amazon Simple Storage Service (Amazon S3), Amazon Kinesis, and Amazon MSK. Required device telemetry and baggage tracking events are sent to the Amazon MSK to stream and temporarily store the data in near real-time, Amazon S3 to store telemetry data long-term, and Lambda to act on low-latency events.
This event-driven architecture provides reliable, resilient, flexible, and near real-time data processing. AWS IoT Core and Amazon MSK are deployed across multiple regions to provide the required resiliency. Amazon MSK also uses Kafka MirrorMaker2 to improve reliability in the event of regional failover and synchronizes the offsets for downstream consumers.
Baggage tracking data must be persisted within a central baggage-handling datastore. This supports downstream applications, reporting, and advanced analytical capabilities. To ingest the required telemetry data, the solution uses Lambda to subscribe to the respective MSK topic(s) and process the scans before ingesting the data into Amazon DynamoDB. DynamoDB is ideal for a multi-region, mission-critical architecture that necessitates near-zero Recovery Point Objective (RPO) and Recovery Time Objective (RTO).
During baggage loading, devices like belt loaders and handheld scanners often require bi-directional communication with minimum latency. If you require publishing data to similar IoT devices, then Lambda could publish messages directly to AWS IoT Core.
With the vast amount of device telemetry and baggage tracking data being collected, the solution uses Amazon S3 intelligent tiering to securely and cost-effectively persist this data. The solution also uses AWS IoT Analytics and Amazon QuickSight to generate near real-time device analytics for the fixed readers, belt loaders, and handheld scanners.
As depicted in Figure 3, the solution also uses service to collect, process, and analyze the incoming MQTT data streams from AWS IoT Core and store it in a purpose-built timestream data store. Amazon Athena and Amazon SageMaker are used for further data analytics and Machine Learning (ML) processing. Amazon Athena is used for ad-hoc analytics and query of large datasets through standard SQL, without the need for complex data infrastructure or management. Integration into Amazon SageMaker makes it convenient to develop ML models for tracking bags.
Conclusion
In this article, we discussed using AWS IoT, Amazon MSK, AWS Lambda, Amazon S3, Amazon DynamoDB, and Amazon QuickSight, airlines can implement a scalable, resilient, and secure baggage tracking solution that addresses the limitations of traditional systems. The modernized solution, powered by AWS services, ensures near real-time tracking, enhancing operational efficiency and passenger experience through accurate tracking, reduced mishandling, and efficient recovery of misplaced luggage. Additionally, it addresses cybersecurity threats, data privacy concerns, and regulatory compliance while enabling data analytics and reporting for informed decision-making and operational optimization.
To learn more about the components in this solution, see the Further reading section. Also to discuss how we can help to accelerate your business, see AWS Travel and Hospitality Competency Partners or contact an AWS representative.
Further Reading
IBM Consulting is an AWS Premier Tier Services Partner that helps customers use AWS to harness the power of innovation and drive their business transformation. They are recognized as a Global Systems Integrator (GSI) for more than 17 competencies, including Travel and Hospitality Consulting. For additional information, please contact an IBM representative.
About the authors:
Neeraj Kaushik is an Open Group Certified Distinguish Architect at IBM with two decades of experience in client-facing delivery roles. His experience spans several industries, including travel and transportation, banking, retail, education, healthcare, and anti-human trafficking. As a trusted advisor, he works directly with the client executive and architects on business strategy to define a technology roadmap. As a hands-on Chief Architect AWS Professional Certified Solution Architect and Natural Language Processing Expert, he has led multiple complex cloud modernization programs and AI initiatives.
Venkat Gomatham is a Sr. Partner Solutions Architect at AWS helping AWS System Integrator (SI) partners excel. He has worked as an IT architect and technologist for more than 20 years to lead innovation and transformation. He serves as a subject matter expert (SME) and Technical Field Community (TFC) member at AWS in the Internet of Things (AWS IoT) with specialties in Automobile and AI/ML.
Subhash Sharma is Sr. Partner Solutions Architect at AWS. He has more than 25 years of experience in delivering distributed, scalable, highly available, and secured software products using Microservices, AI/ML, the Internet of Things (IoT), and Blockchain using a DevSecOps approach. In his spare time, Subhash likes to spend time with family and friends, hike, walk on beach, and watch TV.
Vaibhav Ghadage is an AWS IT Specialist at IBM with several years of IT experience and is currently working in IBM Consulting. He is an AWS Professional Certified Solution Architect and primarily focuses on cloud.
