Hortilux cultivates an illuminating IoT approach with Mendix

Jessica Twentyman Profile picture for user jtwentyman February 8, 2018
Dutch manufacturer of grow lights Hortilux has developed a digital platform where IoT-sourced insights, harvested from commercial greenhouses, are shared with customers.

MLZ_7967 jako Obiekt inteligentny-s
Later this year, a new, environmentally friendly greenhouse will open for business at Ter Laak, one of the world’s largest growers of Phalaenopsis plants (moth orchids).

Based at Ter Laak’s headquarters in Wateringen in the Netherlands, the 5-hectare greenhouse will be warmed using solar energy and rely on a system of highly efficient LED grow lights from specialist supplier Hortilux.

In a further boost to the company’s green credentials, Ter Laak employees will be able to get an at-a-glance view into lighting efficiency via an online portal, Hortisense, enabling them to switch off lighting when it’s not needed, curbing energy consumption without impacting the eventual harvest.

Business transformation

These days, Hortilux provides access to the Hortisense platform to any company that, like Ter Laak, buys its grow lights, whether their installation comprises 100 lighting fixtures or 10,000, says Arno Wartewig, director of business development at Hortilux.

The digital platform’s launch in mid-2015 marked an important milestone in Hortilux’s development from ‘just’ a manufacturer of grow lights, to a company that also provides its customers – commercial growers – with installation services, online information and cultivation advice. It’s down to developments in Internet of Things (IoT) technologies, Wartewig explains, that Hortilux has been able to transform its business model in this way:

Our business is changing in several important ways and we needed to change with it. Growers are getting bigger, they’re becoming international and the growing business is attracting investors, too, so our customer base is changing. At the same time, growers are still very important, but they want to rely not just on their ‘green fingers’ - their instinct, their intuition - but also on data.

So what that means is that we have to be able to identify what’s happening in a greenhouse and share our findings with growers and to do that we need to collect and visualize data and help them to interpret it, with the end goal being that they achieve a higher yield per square metre of valuable greenhouse space.

Three-stage IoT

There are three elements to Hortilux’s IoT approach. First, there are the sensors that Hortilux installs in the professional greenhouses of its customers. These collect various types of information, including data on temperatures, carbon dioxide levels and power quality in the lighting installation.

Second, there is the platform on the Amazon Web Services (AWS) cloud where this data is stored and managed.

Third, there is the Hortisense platform, which provides growers with an interface by which they can view the information. This was (and continues to be) developed on Hortilux’s behalf by consultancy firm First Consulting, using low-code, rapid application development tools from Mendix.

The team here is led by First Consulting agile development practice lead, Andre van Teeseling, who says Mendix was chosen on the basis of its flexibility to support not just the initial idea for Hortisense, but also additional features and functions that were later added, as well as plans for its future development:

Three years ago, this was just an idea. Hortilux knew what they wanted to achieve initially and shared this with us, but we recognized from the start this was the beginning of a journey. And we were right, because we are still learning and adapting Hortisense as we go along.

Blooming benefits

Already, Hortisense provides a range of insights to the company itself and its customers, with two main benefits. First, the platform monitors and safeguards the hardware used in greenhouses, by delivering data on the workings of fixtures, transformers and electrical panels. It can predict the remaining life-time of a fixture and send automatic repair alerts when lighting fails.

Second, it delivers real-time insight into the performance of the wider greenhouse, so that growers can optimize their growing strategies. The application provides real-time insight, for example, into the amount of light reaching a crop, both from the sun and from grow lights. This knowledge, combined with weather forecasts and energy prices, for example, enables growers to optimize their lighting strategies for sustainability, energy cost and productivity. Says Wartewig:

As time has passed, Hortisense has become more sophisticated in terms of what it delivers. At first, it was really just about collecting relatively simple data and presenting it to the grower. Now, it’s more about sophisticated, big data analysis leading to deeper insights.

For example, First Consulting is working on analysing the data with machine learning to see if we can identify certain rules for particular crops and particular growers, to see how they can improve their growth strategy. Also we can now predict with the platform how many kilograms of tomatoes a client might harvest over eight weeks with a particular light strategy and that’s a big new insight that customers never had before. In addition, they are excited by the idea of being more sustainable - of controlling their lights so that they use energy more efficiently but still get the best yields.

In fact, Hortisense is now the main channel to Hortilux for many customers, who can also use it to keep track of their purchase history, download product information and use its e-commerce capabilities to request installation services or purchase replacement bulbs and reflectors. Adds Wartewig:

Customers are very enthusiastic about Hortisense. Not every grower uses all the modules and some are more focused on the predictive maintenance modules, while others focus on the growth modules, but we see more and more of them wanting the whole package as they see the value of using data to improve yields and, after that, start to see the results.

A grey colored placeholder image