Visually driven decision support tools for viticulture

Decision Support Tools/Systems (DSSs) are designed to assist users in decision-making scenarios that call for the processing of highly complex data. Examples of such a decision-making scenario include weather prediction, irrigation, yield prediction, and so on. In the field of agriculture, various state-of-the-art DSSs are being used by different stakeholders such as farmers, advisers, and policymakers to facilitate such activities. Depending on the type of decision support required, data is first gathered from multiple sources including sensors, satellites and in-field observations, and analysed using a series of statistical models. The output is then presented in a number of ways such as tables and/or graphs.

Visualisations are efficient in helping us make sense of complex data that might otherwise require significant cognitive effort. For instance, carefully selected visualisations may assist users to better understand not only farm data, cost, yield, etc., but also how and why a prediction model produces certain outcomes. In the field of viticulture, visualisation tools can enable a better vineyard monitoring, cost-profit optimisations and at the same time, generate a more transparent representation of the existent variability in the vineyard, which is valuable for the optimisation of the harvest and producing high-quality grapes. A few existing visually driven DSSs that are specifically designed for viticulture include Vite.net[1], GeoVit[2] and RFID/GPS-based grapevine management systems[3] [4]. The GUI of these systems can be seen in the figure below.

o.jpg

Vite.net is designed for the crop management of vineyards. Using sensors, it shows visual information about soil water content to help farmers monitor adequate water levels. Decision support modules provide information about vine growth, pest control and diseases in grape berries. The visualisation of real-time information supports farmers to make informed decisions and maintain a record behind each management action. The RFID/GPS-based grapevine management system by Blauth and Ducati introduces a map visualisation of the land usage in vineyards. The system monitors grape variety, production and inventory information, using a map to indicate vineyards, vegetation, bare soil or nearby urban areas. GeoVit provides support to farmers at a landscape level with an interactive dashboard that allows the selection and comparison of areas of interest together with multiple data layers on top of a map.

In the existing agricultural DSSs, dashboard visualisations have been seen widely used. Dashboard visualisations provide immediate access to various interactive components to monitor and display data, offering a set of visual tools to the decision-maker for interaction and exploration. Vite.net, for example, has a dashboard visualisation that shows the synthetic information concerning those aspects of vineyard management that require day-to-day support: vine growth, diseases, insect pest phenology and soil water content.

On the other hand, uncertainty visualisation has been one aspect that many existing DSSs have ignored but can unquestionably improve user understanding of the decisions provided by the system. By conveying the possibility that a point estimate may vary, uncertainty visualisations allow users to make more informed decisions based on prediction models. Despite the importance of illustrating uncertainty in visualisations, the DSSs in viticulture domain have not yet seen many applications of uncertainty visualisations.

For more information please contact Nyi-Nyi Htun – nyinyi.htun@kuleuven.be

 


[1] https://doi.org/10.1016/j.compag.2013.10.011

[2] https://doi.org/10.1016/j.compag.2017.05.028

[3] https://doi.org/10.1016/j.compag.2010.12.013

[4] https://doi.org/10.1016/j.compag.2010.01.007