Report

Teaser, summary, work performed and final results

Periodic Reporting for period 2 - BigDataOcean (BigDataOcean - Exploiting Ocean\'s of Data for Maritime Applications)

Teaser

The challenge that the maritime-related industries and organisations face nowadays is twofold. From the technological point of view, big data come with common recognised challenges. Carefully designed systems are needed in order to be able to integrate heterogeneous resources...

Summary

The challenge that the maritime-related industries and organisations face nowadays is twofold. From the technological point of view, big data come with common recognised challenges. Carefully designed systems are needed in order to be able to integrate heterogeneous resources coming from different sources. It goes without saying that these systems should be cloud-based, in order to ensure seamless and universal data access; Another common obstacle is the pursued “collaboration” and integration amongst the various fragmented data types constituting the “big data stack” ; legitimate concerns can be raised to achieve cross-sector data collaboration. However, the aforementioned do not constitute the only challenges one can identify; Selecting exploitation of proper data to solve the problem at hand, clarifying the correct model for storing and possessing the incoming data, identifying how to deploy the data for serving real-life applications, covering the “semantic gap” between the targeted knowledge and incoming (either structured or unstructured) data constitute only some of the challenges faced in the pre-application stages. Even data visualisation can prove to be challenging when data are aggregated from diverse sources in a huge volume and imported to a model allowing decision-making in minutes rather than weeks.
From a practical point of view, although a plethora of individual and disconnected applications can be found serving the “data exploitation for marine-related applications” profile, the fact that there is a lack of networked initiatives bringing together organisations and knowledge from different scientific and policy domains, as well as geographical areas is quite disappointing. Combined data from port authorities, buoys, large and small ships, satellites, but also available open datasets on weather, fauna and flora population etc. can open a wide spectrum of unprecedented services, if effectively tackled by modern technologies. As it can be easily supposed, when stakeholders try to combine all aforementioned data in order to extract knowledge and conclusions, they are confronted with a challenge; to homogenise and cope with huge volumes of data. The abundance of such data volumes calls for new, innovative, more efficient and more effective applications and novel calculation methods. These applications and methods, if correctly designed and implemented, are capable of renovating industries operating in the maritime environment.

Work performed

The activities of BigDataOcean summarised below:

Architecture, Data Ingestion and Development
• 72 vocabularies have been collected for the harmonisation of data in BigDataOcean
• A significantly renovated Data Ingestion system has been developed to allow the automated ingestion, semantic enrichment, parsing and normalisation of data from different type of sources including real-time data. This system enabled the fast and timely ingestion of large amounts of data on the platform.
• By the time of the report, more than 170 TB of data from 209 data sources are available.
• The BigDataOcean backbone was renovated and in a layered, cloud infrastructure, with several virtualization and performance capabilities that would enable executing data processing tasks over the new large amounts of collected data (billions of rows).
• A new data exploration section has been integrated in the platform that allows users to browse, query,visualise, search, preview the datasets, their metadata and their variables on the platform.
Pilots and Evaluation
• A detailed description of the evaluation framework and validation methodology was provided, including an extensive description and justification of the models used in the development of the evaluation framework. Additionally, a validation plan has been included addressing the target of the platform validation from at least 1000 users.
• The set of scenarios associated with each pilot and respective test cases were enhanced and improved to demonstrate the added value of the BigDataOcean platform both in terms of scientific innovation and business situation. The stakeholders involved in each pilot were described, giving a special focus to their current and envisaged business situation in the context of the designed scenarios and the envisaged solution that the BigDataOcean platform is providing through the Pilot applications.
• The results of the external evaluation have been documented and used as feedback to be taken into consideration for the development of the lessons learnt while formulating them into adoption guidelines for the users as well as the impact assessment of the project into the business market.
Networking and Dissemination
• The dissemination and communication activities enabled the BigDataOcean platform evaluation by more than 1100 individuals, 70% of the ones engaged in the last evaluation phase were related with stakeholders from the maritime and energy industries.
• The number of industry contact points has grown more than 400% (from 59 at M15 to 309 at M30). BigDataOcean communication has also reflected a positive impact. For instance, the number of followers in social channels increased nearly 45% (from 176 to 255 in a similar period, reaching 431 followers in total) and the number of unique visitors to the website increased by 124% (from 2707 visitors at M15 to 6051 at M30).

Exploitation
• A detailed business plan has been defined for afore-mentioned BDO commercial partnership (HOLISTIC, UBITECH, MARINETRAFFIC), important market hypotheses have been validated while several negotiations are ongoing for converting prospects into paying customers.
• The exploitation strategy has rolled-out in each business and a ROI quantification was performed for each of them.
• Thanks to BDO business-oriented dissemination, 1000+ prospects tested BDO platform in a hands-on manner and provided their feedback.
• During the grant period, the Consortium has been able to cultivate a community of maritime organisations that recognise the value of BDO solutions and are eager to adopt them as a working tool. At the project completion, in fact, 116 prospects have expressed their interest in a commercial follow-up.
• 4 deals have been signed by Consortium partners in the second reporting period. 2 of them, in particular, allow UBITECH to continue providing BC1 solutions to ANEK (i.e., Fuel Consumption Reduction Investigation Service) and FOINIKAS (i.e., BDO Fault

Final results

• Final BDO Architecture
• Harmonisation Tool
• Integration of all state-of-the-art analytics algorithms
• Data Ingestion System
• Query Designer
• Visalisation Engine
• Dashboard Builder
• Realisation of the pilots (evaluated by more than 1000 users)
• Dissemination and Communication to more than 1100 Individuals
• More than 300 Industry Contact Points
• Final BDO Business Model
• 116 Business Prospects, 4 Deals

Website & more info

More info: http://www.bigdataocean.eu/.