By Eva A. Kyriakopoulou*
Ufortunately, answering this question is not that simple. We can invest in the latest technology for our cages or feeding system or use the most advanced feed for our livestock but we cannot overlook two key aspects:
The success of our production is affected by a range of different parameters, from environmental parameters, to feed types, feed composition, feeding rates and practices, as well as net changes, production management strategies, the people who are actually involved, as well as many other considerations.
Even though we capture and store large amounts of data, it is difficult to interpret the data and thus identify trends, issues or integrated patterns that may help us improve. It is very difficult (almost impossible) to understand and quantify the complex interactions between the different parameters and make decisions, just by looking at our data.
Aquaculture companies from all over the world are drowning in data, but starving for the knowledge and insights that will help them to reduce costs, improve operational efficiency and increase performance. How many times have we wondered about the efficiency of the feed we are using, the fry supplier or the feeding rates and practices?
In January 2015, Niordseas (Andromeda group) along with two other aquaculture companies, Grammos S.A (GRE) and Ardag (ISR), started working on a €3.1 million, 2 year long project called “AquaSmart“(www.aquasmartdata.eu), funded by the Horizon 2020 framework Programme of the EU. Aquasmart is run by a skilled and experienced consortium including top research institutes (WATERFORD INSTITUTE OF TECHNOLOGY - WIT from Ireland, INSTITUTO DE DESENVOLVIMENTO DE NOVAS TECNOLOGIAS - UNINOVA from Portugal, Q-VALIDUS LIMITED - Q-Validus from Ireland, INSTITUT JOZEF STEFAN - JSI from Slovenia) and a major player in the aquaculture IT sector (OLOKLIROMENA PLIROFORIAKA SISTIMATA - I2S from Greece).
AquaSmart will use big data, data mining and open data to assist companies in transforming data into knowledge, and then using this knowledge to improve efficiency, increase profitability and carry out business in a sustainable, environmentally friendly way. By providing production datasets and actively contributing to developing a better understanding of the way aquaculture companies process data the commercial co-operators are aiming to assist in the creation of global models, which will be used to evaluate production.
Globally, it is the only data mining and big data technologies initiative applied to the aquaculture sector. By using AquaSmart, we will be able to take decisions and improve the main Key Performance Indicators KPI’s of our production such as FCR, mortality rate, growth rate, production time, health, etc. In doing so, we should be able to better:
• Control the production process for maximum profitability,
• Respond to a wide range of production challenges, in real time,
• Identify, in a timely manner, production problems or trends,
• Evaluate feed and fry suppliers, feeding practices and fish management strategies and
• Continuously improve feeding and management practices.
AquaSmart is innovative because it enables us to perform data mining at the local level and get actionable results. It also provides highly accurate descriptive and predictive models to help understand not only what has happened, but more significantly what is most likely going to happen next!
However, the project goes much further: through the availability of multi-lingual Open Data, companies will be able to compare their specific results with other results that are stored in the cloud. By collecting and managing the data mining results from many companies and a large number of rearing cycles in variable conditions, with full respect to confidential data, Aquasmart will generate a knowledgebase that will be of maximum usefulness for all the stakeholders of the Aquaculture sector (i.e. companies, researchers, environmental agencies and governments).
Whether Aquasmart is used to drive new business, reduce costs or gain the competitive edge, data mining is a most valuable and transformational asset for every fish farming organization, be it a large or small business.
The secure and unobtrusive collation of data should enable analysis of huge volumes of historical data to deliver informed business driven knowledge from models built for prediction, estimation, and other inferences involving uncertainty.
The aim of data mining is to extract valid, novel, potentially useful, meaningful and understandable patterns accumulating in large and complex databases. It involves mathematical and statistical methodologies to explore the data, develop the models and discover previously unknown patterns. The project commenced in February 2016, and it has recently launched a free data exploration tool which is available from the Aquasmart website: http://www.aquasmartdata.eu/data-exploration-tool/)
The tool helps companies to gain insight into already available production data and is completely free, without any restrictions or limitations. You are most welcomed to register on the web site, use the tool and subscribe to the newsletter, which will provide you with up-to-date information on the project, as well as more free tools that will be provided in the future.
Added to this, the Aquasmart project also delivers a European standard CEN workshop agreement standard that complements the use of big data for the aquaculture industry. It is further supported by an intelligent training tool and programme that is designed specifically for the aquaculture industry and for professionals to up-skill their big data knowledge for the industry. In addition, Aquasmart delivers a certification assessment programme that complements the advancement of the use and skills of big data in the aquaculture industry.
Eva A. Kyriakopoulou is an Oceanographer - Ichthyologist from Greece. She comes from a seafaring family and the sea has always been her great passion. She graduated from the department of Marine Sciences of the University of the Aegean, in Lesvos island and she immediately started working for Andromeda S.A., one of the biggest sea-farming corporations in the Mediterranean. She is currently stationed in Preveza, eastern Greece.