Technologies at the service of farmers and livestock farmers
The farming world is experiencing major changes. According to McKinsey & Company, about a third of food produced is lost or wasted every year, amounting to a $940 billion economic hit, at the same time when 50% more and better food will be needed over the next 20-30 years. Inefficiencies in planting, harvesting, water use, reduced animal contributions, as well as uncertainty about weather, pests, consumer demand and other intangibles contribute to the loss.
To that extent, the goal of the CYBELE project is to demonstrate how the convergence of HPC, Big Data, Cloud Computing (services) and the IoT can revolutionize farming, reduce scarcity, increase the food supply and bringing social, economic, and environmental benefits.
The CYBELE project has been started at the beginning of this year for 3 years in total.
Atos Bull is delighted to be part of this project to help by providing technologies the Aquaculture, Livestock and Agriculture to increase the efficiencies in planting, harvesting, water use and to reduce the loss. Just to name a few, nine demonstrators covering the Agriculture, Livestock and Aquaculture have been defined in the CYBELE project involving in diverse tasks of data curation and preparation for a variety of data and model sources, as well as modeling and simulation. CYBELE relies on High Performance Computing infrastructure, to provide the compute power required to advance models and methods of the demonstrators. A holistic set of security and privacy-preserving services will be embedded throughout the entire operational lifecycle of the CYBELE platform.
With the related stockholders in the project, Atos Bull have been working on the demonstrators to understand the challenges and needs for increasing efficiencies, define the key features in order to i) advance models and methods by focus on forecasting future events and behaviors, enabling businesses to conduct what-if analyses to predict the effects of potential changes in business strategies, ii) deliver an advanced tool for the analysis and management of the results and predictions on a large scale allowing the monitoring and analysis of recommendations for user consumption in order to facilitate decision making by the end users of the data.
Taking Open sea fishing demonstrator as example, their main challenge is the state of the largest part of the marine ecosystem, including most fish stocks, remains largely unknown causing that little ecosystem-based management has been put in place despite the success of fisheries management in the EU during the last decade in rebuilding overfished stocks and preventing overfishing thanks to the increased availability of data and better analysis methods that enabled to assess, and thus provide more precise management for an increasing number of commercially exploited fish stocks.
An important reason for this is that most marine data is collected by means of scientific surveys on research vessels. Such surveys are expensive, and consequently, it is practically impossible to provide a full spatiotemporal data-coverage of the seas. In contrast to research vessels, commercial fishing vessels have a much wider spatiotemporal coverage of the seas. Moreover, the increased usage of sensors and IT equipment on board of commercial fishing vessels allow these vessels to collect many data. However, due to the lack of sufficient processing capacity and adequate database systems, nor fishers e.g. to optimize their operational decisions, nor fisheries managers make optimal use of these data.
To answer to the Open sea fishing challenges, the CYBELE platform is going to be used to integrate different data sources stemming from satellites, individual fishing vessels and auctions to assess the distribution and exploitation status of fish stocks at high spatiotemporal resolution. As such, fisheries managers can use the output to optimize quota uptake, fishers can optimize their operational decisions at sea, while fish auctions and buyers will have a better idea on how much from each fish species will be landed by the fishing fleet. The solution will allow to i) exploit the plethora of existing relevant datasets on the CYBELE platform and minimize the cost of maintaining these datasets in-house, ii) provide out-of-the-box cloud infrastructure for state-of-the-art libraries for statistics, algorithms, deep learning and neural networks that will minimize the cost for running those in-house, iii) provide sufficient memory for evaluating spatial abundance simulation models, iv) implement more accurate and high quality spatially explicit fish prediction models, v) deliver the results in a timely manner (utilizing the HPC power of CYBELE) that is critical especially in case of optimizing fishing effort allocation, vi) provide efficient and cost-effective data collection and management.
To conclude, CYBELE project will demonstrate increase of innovation and productivity in the main target sectors as well as the market share of Big Data technology providers. It will also allow i) the effective integration of HPC/BD/Cloud/IoT technologies in the main target sectors, resulting into integrated value chains and efficient business processes of the participating organizations, ii) widening the use of and facilitating the access to advanced HPC, big data and cloud infrastructures stimulating the emergence of the data economy in Europe and iii) stimulating additional private and public target investments in HPC and Big Data technologies from industry, Member States and Associated Countries, and other sources, as referred to in the contractual arrangements of the HPC and/or the Big Data Value Public Private Partnerships.