Bram van Rens is Principal Data Strategist at Blue Radix. In this interview he explains what data science is and how he applies it. He also reveals its added value for greenhouse horticulture, not just in autonomous growing, but also looking ahead to future data science developments in the greenhouse industry.

Tell us a bit more about your background and when you joined Blue Radix.
“I grew up in a village in Limburg in the Netherlands. As a teenager I had a brief part-time job in a small bell-pepper greenhouse, but I had nothing to do with horticulture for a long time after that. I studied Applied Physics in Enschede and then got my PhD at the University of Amsterdam. After that I wanted to work in a more applied way. A well-known medical technology company in Eindhoven tempted me with an interesting job in product innovation.

I worked there for ten years, then moved to AgroEnergy about five years ago. That’s where I encountered greenhouse horticulture again after a long break. As a Senior Data Scientist I worked on smart, data-driven energy solutions and I really saw how innovative, versatile and complex horticulture is.

That’s why I embraced the opportunity to start at Blue Radix last year. Here I can deepen my horticultural knowledge even further and use my data science knowledge and experience to tackle the complex tech challenges in the horticultural sector.”

What’s your role, and what are you involved in?
“My position at Blue Radix is that of Principal Data Strategist. I work on lots of different topics involving data and data science, to keep autonomous growing evolving. Data science is a multidisciplinary field where methods from mathematics, statistics and computer science combine with domain knowledge, to extract meaningful information from (large volumes of) data to create smart products and services.

One of my main focus areas at Blue Radix is to translate our customers’ challenges and business issues into innovative, practical data science solutions that can be applied in greenhouses worldwide. These solutions include optimizing the greenhouse climate for example, or generating setpoints for the best possible control of the installations, or predicting the yield. I’m a member of the product development team, so naturally I work closely with  everyone in it.

I also keep an eye on the latest data science and artificial intelligence developments to keep our knowledge and skills up to date. We delve into relevant new methods and where needed, we are supported by partner parties.

I also work continuously on our data and analytics strategy. We are looking ahead to what data, data quality, analyses and algorithms we need to realize future applications for autonomous growing. In this strategy I also include the results of scientific research projects I’m involved in, like FlexCrop (optimization of energy use in the greenhouse through flexible crop management).”

What are the biggest challenges in your work?
“That’s a tough question, because there are so many challenges! But to name a personal one, it’s finding the balance between developing something that’s perfect, versus something that already delivers value in an earlier stage. I’m a perfectionist by nature and often see ways to further improve a solution, even though we may already have something that’s very valuable to our customers. Over the years I’ve become better at finding this balance, but it does need my continued attention. What also helps here is the way we work together in the product development team and in Blue Radix as a whole. We really put effort into quality, while at the same time our goal is to bring any added value to the market as soon as possible.

One of the data science challenges is obtaining sufficient volumes of data, and of sufficient quality. This includes data from the climate computer, sensors and the weather for instance. In general, the more data and the better the data quality, the more optimally our solutions work. We’ve been able to tackle this challenge well thanks to the good cooperation with our customers, and of course we will continue doing so.”

How do you see the future of data science in the greenhouse industry. What’s on the horizon?
‘We’re only at the beginning with autonomous growing; there’s still a lot to be developed to manage all the processes involved. That apart, I envisage developments in several overarching themes where I believe data science will play an important role.

One of these themes is minimizing the ecological footprint. The horticultural sector is already working hard on this. I think data science will accelerate these developments significantly, because increasing volumes of data, from many sources, are becoming available and can be combined. This makes it possible to reduce the footprint significantly by optimizing the use of energy, water, nutrients and pesticides for instance, while of course maximizing yield and quality.

The second development is actually an extension of this: chain optimization. By this I mean the entire chain from production to consumption, so from seed, cultivation, production, storage, transport, sales and (super) market, up to and including the experience of the end consumer. This obviously involves large volumes of data with very complex relationships and influences between the various steps in the chain. Data science provides the means to perform in-depth analyses, predictions and optimizations on these data. For example, to provide insights into logistical bottlenecks, generate robust sales forecasts, improve quality and increase customer satisfaction.

I believe these developments contribute to the unique character and sustainability of horticulture as a whole, and of growers specifically, thus maintaining business continuity. In short, in greenhouse horticulture, data science is here to stay.”