Human Brains Need Help Making Decisions

Coen Jonker, Data Scientist / Analytics Department

Coen Jonker works in a small, fun team of eight young people, who help clients understand their data, exploit it to improve decisions, and create a more prosperous future.


Coen Jonker: ‘I like rain forecasting apps as a metaphor for the way data are (not) being used most of the time: the information is there, everybody can see rain coming on their phone, but people still get rained on. Humans simply have a hard time taking all data into account during a day: grocery lists, appointments, traffic and then weather forecasts.

When it comes to organizations, decision makers face the same overflow of information. To deal with this, humans follow their ‘gut feeling’ based on partial information and poorly chosen vanity metrics. For example, if you launch an app and it is downloaded forty thousand times on the first day, you may think it is a huge success. But download counts do not say anything about whether people will be using your app next week. Key is to make decisions based on the right data, and the correct interpretation of this data.`


Cool example

‘A cool example of the way we approach data to support decision making at Mobiquity is what we did for the Passenger Boarding Bridges at Schiphol Airport, which is one of our clients. A major pain for everyone at airports is waiting to get off the plane. And of course any unnecessary time that planes are on the ground is considered expensive waste. That was the problem at hand. Schiphol wanted us to find out how much time they could save by automating the docking process of the Passenger Boarding Bridges. Schiphol wanted to know how much time could be saved if the connection process could be reduced to 1 minute. So we dove into their data lake. The first thing we were asked to check is the average time it takes personnel to do it. But — easily overlooked — even a small technical failure increases that average time dramatically, because averages are very sensitive to extreme values. Averages are usually not a very good statistic to use for processing times.

We needed a deeper analysis of the whole industrial process. In collaboration with Schiphol we analysed the millions of log entries and really calculated the time that could be saved by robotizing the bridges. It turned out, automated bridges could reduce handling time with almost a minute. At first glance that does not seem a lot, but if you take into account that more than 60 million passengers travel through Schiphol yearly, minutes sum up to many years of passenger waiting time. Passenger waiting time in itself is not a bad example of a vanity metric (laughs), because what does a year of waiting time that actually mean? However, a more reliable docking process also improves transfer reliability, so more people make their connecting flights, which strengthens Schiphol’s function as an international flight hub.

We could clearly show from data that installing robotic jet bridges is worthwhile and Schiphol is now moving forward with this innovation. `


The essence

‘An important aspect of data science from Mobiquity’s point-of-view is to spot ‘invisible’ business opportunities for our clients; we often create business opportunities from scratch. For example, DTG (‘De Telefoongids’, the Dutch telephone book) is another one of our clients. They have a huge amount of data and we help them benefit from it. We created an algorithm that ranked the most interesting commercial leads for them based on the information they have available.

We aim to create value for our client’s clients. For an automotive website we created a price prediction tool for its car dealer clients. For example, a used Volkswagen can be cheaper in the south of Holland, and a vintage Porsche can be cheaper in the north. Having a market price prediction for cars helps car dealers to lower their cost and raise their profit.`

‘We have a Dutch saying ‘regeren is vooruitzien’ which can be translated as, ‘to rule is to look into the future’. In other words: to prosper is to anticipate and make wise decisions for the future. The amount of available data is overwhelming nowadays and incorporating all available information in your decisions is no longer feasible for human brains — and that’s where we come in… With all the data at hand in the world, and the right skills and tools to process them, you can really improve companies and society as a whole.’


We hope you enjoyed reading this, and don’t forget to follow us if you’d like to read more interesting stories.

If you are looking for a new opportunity as a Data Scientist, come join us at Mobiquity: