how AI can help Agritech

Feb 8, 2023 | by S. Zamboni, Use Cases

by S. Zamboni

Climate change has been receiving growing attention in the last few years, with more and more people concerned about the human contribution to global warming. An aspect of climate change that is already affecting numerous countries and causing great distress is the scarcity of drinking water. With growing populations and increasing demand for water resources, draughts are a serious concern that needs to be addressed immediately.


Innovative technologies to reduce energy and water waste are one of the ways in which a more sustainable future can be achieved. In this regard, the combination of real-time data and artificial intelligence is a game-changer, as it offers exceptional opportunities to optimise consumption.

Radicalbit offers a viable solution with our MLOps Platform designed to unleash the full potential of AI and Machine Learning for numerous use cases, including water management for agriculture. Radicalbit’s Platform can support the implementation of a data-driven approach leveraging smart systems to analyse real-time and historical data, such as soil and weather conditions.. This information can then be used to make well-informed decisions on water usage, resulting in a more sustainable and efficient use of such a precious resource.

Water waste scenario in agriculture

According to an analysis by The Future We Don’t Want, a minimum of 10% decrease in freshwater availability will affect 685 million people in over 570 cities due to climate change by 2050. Some cities such as Amman, Melbourne, and Cape Town may face a decrease in freshwater availability ranging from 30-50%. Concerning Cape Town, for instance, the city has already been affected by water shortage issues – in 2018, a “Day Zero” drought was caused by an exceptional 3-year rainfall deficit. For months, Capetonians were forced to survive on less and less water as severe restrictions brought extreme hardships for many, and even the loss of livelihood for some.

As for agriculture, according to a report by High Tide Technologies farms around the world account for 70% of all water consumed annually. Of that 70% used by farmers (over 2 quadrillion gallons), 40% is lost to the environment due to poor irrigation systems, evaporation, and overall unsatisfactory water management. As the world population keeps on increasing, the demand for food and water supplies will consequently rise. By understanding how water is used in agriculture, we can continue to learn and discover new methods of farming to maximise production as well as conserve the water supply. To tackle this global challenge, it is necessary to start developing strategies that will include a combination of structural, behavioral, programmatic, and nature-based solutions.

How AI & Radicalbit Platform can help

To effectively reduce water waste and generate accurate predictions through an automated solution, multiple elements must seamlessly work together to support decision-making activities. Indeed, artificial intelligence can be an incredibly powerful tool in this field: it is possible to build a model to interpret large amounts of real-time and historical data from different environments and soils and to generate accurate predictions in real-time.

With its cutting-edge technology, our MLOps Platform takes the lead in data-driven solutions’ development that can be integrated into smart systems to improve water and energy saving. Our Platform offers the necessary features to provide an automated tool that effectively minimizes consumption and enhances accuracy in predictions.

First of all, it incorporates a powerful streaming engine that simplifies the process of sending data to the model and collecting predictions in a dedicated Stream. To ensure the model can interpret the information, pre-processing of raw data is necessary to transform it into a usable format. A Pipeline can then be created to manage the flow of data, from raw input to output prediction, with operators in place to perform any necessary transformations.

Moreover, Radicalbit’s MLOps Platform allows multiple pipeline creation, each tailored to meet specific requirements and applications. Pipeline management is facilitated through Helicon’s Spaces, which provide a structured and organized approach to handling different situations. Multiple pipelines provide significant flexibility and customization in addressing different challenges and needs. The use of Spaces as a tool of work management further increases the platform usefulness. Radicalbit’s platform guarantees that all tasks are handled neatly, and helps to keep the workflow streamlined and efficient.

Another significant advantage of using our Platform lies in the ease of setting up data operators’, which can be done with just a few clicks and no coding skills. This makes the models’ update and monitoring process much more manageable and efficient.

In the platform’s back office, models publication, monitoring and updating is super easy and user-friendly, users can do this with just a few clicks. Considering our agritech use case, Radicalbit’s solution presents an unique opportunity for real-time predictions of field water requirements, employing data from sensors, which are placed at key locations in the field. The sensors collect data such as soil moisture and type, rainfall and wind potential, and then they will send their readings to the MLOps Platform’s streaming platform, where they will be processed and fed into our predictive model for real-time predictions.

Incorporating a webhook into the solution further improves Raidicalbit Platform’s capabilities. Webhooks allow alerting messages to be sent based on predefined conditions, such as stopping irrigation in the case of rain or high soil moisture levels, or increasing irrigation if imbalances are detected in the monitored field.

Such an added level of control ensures that resources are used efficiently, while also reducing waste. With its real-time water requirement prediction capabilities and webhooks for responsive adjustments, our Platform becomes a valuable tool for the agriculture industry, allowing for efficient and effective data-driven management of water resources, helping to preserve supplies and minimise waste.

To sum up, the use of data and machine learning can play a crucial role in tackling future challenges, including the pressing issue of water scarcity. Having a platform that efficiently collects and processes data and generates predictions for decision-making in agriculture or water management can result in significant environmental and economic benefits through reducing water waste and conserving resources.