Machine Learning for Paaspop

Paaspop is a three-day music festival that takes place every year during the Easter weekend in Schijndel, the Netherlands. The organization of Paaspop has two main objectives. The first is to create the most enjoyable experience possible for festival-goers. In addition, Paaspop is also looking at how they can increase sales.

Now of course, the festival-goer wants to enjoy his or her favorite artists as much as possible, such as Bløf or Kensington. So if you want to get a drink during performances, it is annoying if you have to queue for a long time at a bar. So the question was how to distribute people as optimally as possible over the available bars. In such a way that everyone waits in line as short as possible and can enjoy the performances for as long as possible.

A predictive model was created for this problem based on Machine Learning. A simple algorithm was chosen that can make a prediction based on sensor data. The model predicts the crowds at a bar and can direct festival-goers to the bar where you have the shortest queue.

Eventually, a screen above the bar can indicate where the wait time is shortest. The beauty of this solution is that it creates a win-win-win-win situation. Both the festivalgoer, the stand holder, the organization and the beverage suppliers will benefit from this solution because irritation about waiting too long will turn into “I’ll just get something quickly.

After all, if you as a festival-goer stand in line for a shorter time, you will be more likely to get new drinks. The stallholder can sell more this way, and part of that revenue is remitted by stallholders to the organization. After all, parties like Coca-Cola and Heineken also benefit from the increase in consumption.

This case study of Paaspop is about physical queues at a festival site, but also about being able to quickly validate the use of Machine Learning. The problem with the queue at Paaspop is not isolated, every organization has to deal with queues. Both for its own customers at a service desk, for example, and for internal processes; think of repetitive tasks waiting for each other.