This work package will explore the best practices borrowed from data science and information management fields for the collection and aggregation of data in order to evolve towards a dynamic monitoring of the network situation in real-time – instantaneous operational efficiency indicators –. WP04 will define the technical requirements for an advanced data model that could cover the needs for the dynamic monitoring of the efficiency indicators.
Architectures and tools already applied in other domains will be analysed. Benchmarking will allow reaching a general consensus on which is best for what kind of tasks, and whether one framework will come to dominate in terms of general adoption. Literature research and preliminary testing will be performed to select the most suitable technology or the combination of them. Apache Storm, Storm Trident, Apache Spark and Apache Spark Streaming are initially identified and will be analysed. Other stream processing solutions that should be assessed are Simple Scalable Streaming System (S4), IBM InfoSphere Streams, TIBCO StreamBase, Software AG Apama and Amazon Kinesis.
WP04 will also evaluate the minimum data and infrastructure needed to have a set of live indicators of operational performance achieved and realistic achievable targets for decision making support.
Based on previous findings, WP04 will implement a stream-based data model that will be used in the second stage of the experimental testing – WP03 – where the air traffic data are streamed in real-time as the flights evolve.