VANL
VANL - Internship Data Science

Studiefase

WO Bachelor 1, WO Bachelor 2, WO Bachelor 3, WO Pre-Master, WO Master

Regio

Noord-Brabant

Dagen

4

Vereiste talen

Engels OF Nederlands

Functie

ICT

Sector

Information & Communication Technology (ICT)

Maanden

3, 4, 5, 6, 6+

Werkvorm

Deels op kantoor

Bedrijfsinformatie

VANL is een Nederlands softwarebedrijf, opgericht in 2005. Wij ontwikkelen programma’s die bedrijven in staat stellen hun data te analyseren en vervolgens ondersteunen we ze om met deze analyse hun bedrijfsprocessen te volgen, stimuleren, optimaliseren en nauwkeurig te onderhouden.

Functie omschrijving

We are looking for someone that has the following qualities:

• You are an HBO/WO master student in the fields of Data Science, Process Mining or Software Science.

• Or; you have recently finished one of the above mentioned masters and are looking for work experience.

• You have an affinity with the public domain and the data involved.

• You are available for a period of 3-6 months. Depending on the time we can either shrink/broaden the project scope.

• You have experience with (some of) the following tools: PROM, RapidMiner, or something similar.

• You are able to work independently, manage your own workload and ask questions when needed.

• You are able to take a step back and look at the project from a helicopter view.

• You like working in an small and socially active team

Functie eisen

Imagine; it’s the end of your internship at VANL. The worst of the corona sorrows are over, your project was a success, ADE (Amsterdam Dance Event) is allowed to take place and there’s a party at Museumplein which starts around 3 o’clock PM. The first attendees will arrive in Amsterdam around 1 o’clock. At 2 o’clock the city is already so crowded that cars are not allowed into the city center anymore. At GVB (Amsterdam’s public transport company) people are working around the clock to manage the hustle and bustle.

Public Transport Replay, the tool you’ve been working on, allows one to visually replay a chosen day on a map, showing pop-up messages with actual context information that add an extra layer of meaning. You can fast forward or rewind time and see all PT-vehicles move on a map. The vehicle’s color shows its punctuality: green meaning timely, red meaning delayed. But you’ve added an extra feature during your internship; not only does it show the big dynamic of the public transport that day, it can also predict the effects of such an event. Where and when will vehicles get stuck in the crowd? When will the departure punctuality turn late? When will things go back to normal? And what are the triggers for these anomalies?

All timetables and real-time data from all busses, metros and trains in The Netherlands are already available in our database. The largest table has over 4.3 billion rows. We have been adding more context information since the beginning of this year, such as weather conditions, emergency alerts, bridge openings, events, etc. All data needed for this project is already available.

In this project, you will start developing the process mining side of this application. You will have to begin with defining the used and needed concepts and some data-cleaning to prevent accidental logging mistakes from cluttering the analysis. A base-line has to be established; calculated statistics to use and test the simulated model on. For example; what is a ‘normal’ driving time? What is an average punctuality? Secondly, it’s time for model extraction; finding the most accurate model and the triggers that make a driving time deviate from this model. Based on this analysis you will have to find an algorithm that is able to “predict” the need for flexible public transport. “Snow is expected in three days, it is advised to add X minutes to the driving times of these lines: (…)” or “An event was just scheduled for the … area, make sure to add X extra trips to the timetable due to extra passengers”.

Arbeidsvoorwaarden

• A small, fun and invested team.

• A highly specialized environment.

• Chances of a job after graduation

• High degree of independence

• Monthly compensation of EUR 350

• Travel allowance

• An internship in the inspiring surroundings of the TU/Eindhoven, as our office is situated on campus.

Solliciteren