Big Data IoT

KWS: Digitalization reaches agricultural fields

Talking about agricultural work can lead us to concepts such as tradition, manual production, analog, etc. However, like all sectors, agriculture is changing and is joining the digital transformation to optimize its processes and achieve better harvest results.

KWS is one of the world’s leading seed suppliers and offers innovative solutions to farmers in 70 countries. One of their main lines of business is the extraction of sugar from sugar beet, but the manual process is slow, tedious, and more costly. To address this, they developed Beetrometer, an innovative system that automatically analyses the quality of harvested beets efficiently by pre-preparing and then analyzing the samples using a spectrometer.

Big Data, IoT
Azure IoT Hub, Azure Functions
In order to be able to use Beetrometer globally, KWS has relied on Plain Concepts as the developer of a solution capable of bi-directional connectivity, as well as a second stage where all data can be exploited, metrics visualized, and anomalies detected in near real-time.


The environments for which Beetrometer is designed include remote rural areas, where it is expected to have very limited or no mobile line. In addition, it operates in conditions (wet, muddy, etc.) that must be supported by any device provided to facilitate connectivity.

Therefore, the first challenge we faced was to achieve bi-directional connectivity to send different types of quality-related data obtained by Beetrometer to the cloud for further analysis and to receive parameter and configuration files from the cloud.

On the other hand, this data had to be exploited to extract metrics and obtain graphs that show them more visually and clearly.



Despite the complexity of resolving the connectivity issues, the connection between the industrial PC and GF-Shop has been established, achieving the desired remote access and uploading the data to the cloud for further processing and analysis, avoiding data loss.

From here, a data infrastructure has been created in Azure, which consumes data from the devices via IoT Hub and with Azure Functions. In addition, a web application has been developed from which user access, metrics, reports, graphs, averages, etc. can be managed.

In addition, it has the intelligence to generate new information when there are new developments in the same file thanks to the continuous checking of the data, which also allows the detection of anomalies and the statistical reaction to different pre-configured alarms.



This project has achieved several major milestones for KWS's Beetrometer solution: data storage by connecting the machine without having to install any software; automatic deployment of the infrastructure, configuration, and application with Beetrometer, creating the cloud infrastructure needed to extract data efficiently; more compelling visualization that processes the data and extracts the relevant information to make it consumable.

This project has already been implemented in the US. Still, it is intended to scale to the rest of the world, leverage cloud-based platforms to avoid duplication of resources, one-off projects, and technical debt, and embrace the use of IoT to connect other devices as needed.

One million data points are currently being monitored, and the volume of data in storage is already almost 100 GiB. Still, these numbers will multiply exponentially as more devices and locations are added.