Search

Big data in agriculture: challenges, opportunities and solutions

The CYBELE project brings together technical, commercial and research partners to build big data solutions for precision agriculture and precision livestock farming. The Wageningen University team interviewed and surveyed use case representatives and stakeholders to identify the current state of big data solutions in agriculture and what they expect to achieve in the future. We found that the use cases are in various stages of technological maturity. We also identified the drivers of change and the big data challenges associated with each use case. Agricultural domain is diverse and influenced by many factors, and the challenges are not easy to overcome considering the fact that there are no off-the-shelf solutions for agriculture.


Despite the challenges, CYBELE partners have been collaborating and sharing experiences. We organized a workshop for use case representatives to share their progress and experiences. In the workshop, they found out that they shared many similar challenges and that they can learn from each other.


We conducted a SWOT analysis to understand the strengths, weaknesses, opportunities and threats in building big data solutions for agriculture. CYBELE partners have both domain and technical expertise and good relationships with stakeholders. Availability of data and computation resources makes it possible to explore solutions to real business needs using techniques such as machine learning. However, it is difficult to scale solutions built in local research settings to an HPC environment and tailor them to the needs of end-users.


Going forward, we plan to organize a workshop including the end-users and stakeholders. This workshop will provide ideas to CYBELE use case representatives to fine-tune their solutions to address the needs of stakeholders.


----------------------------------------------

The link to the original document is here:

https://docs.google.com/document/d/18Quh3VKC9ateG5skK9dp0OxV-bpmGlaWX4uT9vbjWMQ/edit?usp=sharing


42 views