A bundle of services and tools that enable the design, execution and monitoring of the data analytics workflows. This bundle of services provides value to the data scientists, offering an user friendly and interactive UI that supports the whole data analysis life cycle, from the design of the analysis to the execution, with a plethora of data analysis algorithms and visualization types that can be customized according to their needs and scheduled to be periodically executed. This bundle of services is also relevant for business users as they may gain insights on the results of aviation-related analytics through intuitive visualizations.
The ICARUS Data Analytics and Visualisation offers a novel bundle of services that facilitate the analysis of 1st and 2nd tier aviation data with the help of machine learning algorithms, deliver intuitive reporting and interactive visualizations and grant access to the secure experimentation analytics sandbox in order to generate new insights and knowledge.
One of the main goals of ICARUS is to support the analytics algorithms workflow design and execution, by making the best known and widely accepted algorithms in the aviation industry available, so as to allow all aviation-related stakeholders to analyse and visualize results downstream of big data applications and generate new knowledge and insights.
In ICARUS, Data Analytics is viewed from three major perspectives: a) descriptive which refer to methods that attempt to describe raw data and extract some form of useful information interpretable by humans, b) predictive which aims to forecast the future and make predictions based on discovered patterns in the given dataset and prescriptive and c) prescriptive analytics which aims at recommending particular courses of action that lead towards a solution. In the context, the ICARUS Data Analytics and Visualisation bundle of services offers algorithms for Basic Analytics, Machine Learning, Deep Learning and Visual Analytics.
The ICARUS Data Analytics and Visualisation undertakes all the aspects related to the processing and presentation of the results of the performed analysis over the selected the data. The user is offered a simple and intuitive user interface in order to build up analytics workflows in the form of ICARUS applications that are composed by a set of selected: (a) datasets that the user owns or has legitimate access (based on a smart contract), (b) data analysis algorithm’s and (c) parameters values for the selected algorithm’s. The bundle of services enables the execution or scheduled execution of the ICARUS applications in a secured “sandboxed” environment, as well as the visualisation of the results from an extended list of interactive charts that can configured based on the needs of the user.
Overall, data analytics in aviation is rising, with the aviation value chain stakeholders implementing their own, isolated solutions, which may improve particular processes, but fail to capture the whole picture and face a real obstacle when attempting to scale up and combine data sources. It is clear that in big data domain, including aviation, there is a strong need for solution and technologies that offered innovative and effortless execution of the data analytics over multiple datasets with the required security and performance. The list of algorithms included in the services bundle includes variations of existing algorithms that have been further customized for the aviation domain, however the majority of the algorithms are domain agnostic and the list can be further enriched with additional algorithms emerging from the evolvement of the open source libraries.
The objective of this bundle of services is to provide added value to the operations and services of the aviation data value chain. This particular solution on data analytics and visualisations is integrated into a unique data and intelligence platform for the aviation industry, taking into account the concrete stakeholders’ needs and requirements. Progress beyond the state-of-the art is substantiated along the following dimensions:
- An innovative and intuitive user interface in order to build up analytics workflows
- The execution, immediate or scheduled, of analytics workflows in the secure “sandboxed” environment
- A complete data visualisation toolset with offering multiple visualisation options spanning from line charts, time series, column charts, to histograms, tree maps and dendrograms.
The novelty and added value of the ICARUS Data Analytics and Visualisation has been acknowledged by all stakeholders with whom the ICARUS consortium interacted till now in different engagement activities (in events, through dedicated interviews and online questionnaires).
The ICARUS Data Analytics and Visualisation offers an easy to use graphical interface where the user is able to compose simple or advanced data analytics workflows which constitutes as a pipeline of machine learning (ML) algorithms that can be dragged and dropped in an interactive canvas and are allowed to changing the algorithm behaviour and execution providing the input datasets and algorithms’ parameters based on his/her needs.
Additionally, the ICARUS Data Analytics and Visualisation offers the scheduling mechanism for the execution of the aforementioned data analytics workflows that are executed into an isolated and secure experimentation environment.