At All In in Eindhoven and Echt, Valdeludo Adventure Park, Kasteel Limbricht and De Hompesche Molen the cash registers, ticket sales and reservation systems generate a large stream of data every day. Igor: "You can do something with that, I knew that. I just didn't get around to actually doing something with it. Until I came into contact with MKB Datalab Limburg through the grapevine. They give companies practical advice and support on digitization. Because of the regulations that are in place, there are hardly any costs involved. That low threshold and the extra time that corona gave me, persuaded me to investigate together with MKB Datalab Limburg what I could do with my data.

Annet Schmeitz and Luke Xuereb, master students of Business Intelligence and Smart Services at Maastricht University and Jordy Slagter, student of Software Development at VISTA college were paired with Igor by Datalab.

Igor: "I wanted to start simple. That's why I was looking for an everyday problem. Flan, that is! Especially at Kasteel Limbricht and the Hompesche Molen, guests expect flan. So you have to have it in the house. But you never know how much flan you're going to need. So you either run short, or you have to throw flan away. And both of those are not good. That's how I came up with the 'flan predictor.'

Annet: 'By collecting and combining data we want to predict how much flan will be needed each day. We look at different factors that can have an influence. Think of the weather, the day of the week, the season, whether it's a holiday and whether there are events in the area, for example.'

Luke: "I'm involved in web scraping. Using automatic scripts, I collect data from public sources; the weather forecast, for example, but also event calendars and social media. I match this data with data from Entertrain Group, such as box office data, ticket sales and reservations. This is how I predict the number of guests. Annet focuses on the cash registers and the purchasing data of De Hompesche Molen and Kasteel Limbricht. By combining all that data, we can predict the required number of flans.'

Jordy ensures that all of the collected data can be used. Jordy: "The data comes from different systems, so you can't use it right away. I ensure that the Excel files with data are converted to SQL - a standardized language. Programming languages can work with SQL. It's a challenging job, because I've never worked with so much data before. Annet: 'Luke and I are happy to have Jordy on our team. His work allows us to go faster. We complement each other very well.'