In the week from 15 April to 22 April, 2016, a group of 25 enthusiastic young people started the first Data Challenge Week organized by the postmaster designer program Data Science. The program is part of the Jheronimus Academy of Data Science located in ‘s Hertogenbosch. The group consisted of 10 students of the designer program Mathematics for Industry and 15 applicants from all over the world. During the week, the applicants were assessed on their technical skills, competencies, attitude, and personality, to see if they fit the profile required to successfully complete the designer program Data Science.
The 25 students were split up into five teams. Together the students worked on a real data set kindly provided by Grodan, part of the international Danish company Rockwool. Grodan produces substrates used in precision growing solutions. The data set concerns a one year glass house experiment carried out in the year 2015, to which apart from Grodan, six other companies as Philips Lighting and Monsanto contributed.
During one full growth season, the conditions of the substrate (water content, temperature, electric conductivity, bio-culture), the climate conditions inside and outside the greenhouse (temperature, CO2, humidity, radiation), the quality of the tomato plants, and the production of tomatoes were measured by various sensors at sample frequencies varying from twenty per hour to once per week.
Five team leaders, students of the designer program Mathematics for Industry, prepared the data challenge and defined five themes on basis of which the data were supposed to be addressed: root zone, nutrients, climate conditions, crop quality, and water supply. The Data Challenge was concluded with team presentations on Wednesday 20 April. Thursday and Friday were reserved for the application interviews.
Peter Spoor, R&D director of Grodan, showed his gratitude for all the work done by the teams in such short period of time. The conclusions and recommendations offered by the teams give Grodan new insights in how various data science techniques can give focus to future projects in Grodan’s R&D department. The 15 invited candidates who started 15 April as individuals and the 10 MI students became one group, who managed to create a social structure in the bungalow park and a working atmosphere, five teams accepting each other’s roles.
As a result of the week, 11 applicants were positively assessed and offered a position in the program which starts September 2016.