Thu Jul 04 2024

Scheduling with Mathematical Optimization & Automated Feature Generation

Raamstraat 7, 1016 XL Amsterdam

It's summer time and there is nothing better than some networking and knowledge sharing at the heart of Amsterdam over a cold drink!

Join us on the July Data & Drinks event on July 4th at 17:30 at Xomnia's HQ in Amsterdam. Our guest speakers are Operations Research Engineer at ORTEC Eva van Rooijen, and Data Scientist Ekaterina Zvorykina.

This free event includes dinner, drinks and a lot of networking opportunities with data professionals from Amsterdam and beyond.

Abstract of the talks:

Talk #1: Happy Nurses: Solving the Nurse Scheduling Problem Using Mathematical Optimization

With the current shortage of labor forces, creating an employee schedule that meets the capacity requirements can be quite a puzzle to crack. This is especially challenging in light of labor rules, contract specifications for each employee, and the preferences and requests based on their personal circumstances. How can we achieve the best work schedules that get the work done without having a big impact on employee's work-life balance?

In this talk, Eva van Rooijen will explain how to use mathematical optimization to solve this puzzle and, walking us through the complications that arise when the "solution" to the puzzle affects personal lives (as well as the lives of patients).

Talk #2: Automated Feature Generation's Role in Unveiling the Mysteries of Biological Age

In the rapidly evolving field of machine learning, automated feature generation stands as a critical innovation, enabling researchers to delve into complex datasets with unprecedented efficiency and depth. In this talk, Ekaterina Zvorykina will illuminate the pivotal role of automated feature generation in the analysis that she uses in her work with time series (daily step count and heart rate data) for the prediction of biological age.

In the context of predicting biological age, time series data such as step count and heart rate provide a rich but challenging dataset, due to their high dimensionality and temporal nature. Automated feature generation emerges as a solution to these challenges, facilitating the extraction of meaningful patterns and characteristics from time series data without manual intervention and deep domain knowledge. Ekaterina will touch upon the most popular Python packages that empower researchers and data scientists to implement automated feature generation, such as Featuretools, TSFRESH, and Deep Feature Synthesis. These tools offer a range of functionalities from basic feature extraction to advanced, deep learning-based feature engineering.

About the speakers:

  • Eva van Rooijen is an Operations Research Engineer at ORTEC, with a strong passion for research on the interaction of mathematical optimization software and the people it affects. Specifically, she likes to combine the use of interviews and surveys to validate the mathematical translation of very practical problems, such as the scheduling of employees. She completed two master degrees touching upon this topic: Econometrcis and Operations Research (Erasmus University Rotterdam) and Communication Design for Innovation (TU Delft). She has experience teaching data science and data visualization to various audiences.
  • Ekaterina Zvorykina is a versatile data scientist researcher who contributed to in Singapore, where she collaborated with pharmaceutical and biotechnology companies and academia to develop therapeutics against diseases with the help of machine learning. With a foundation in computer science and cell biology and physiology from the National Research University Higher School of Economics and Lomonosov Moscow State University respectively, Ekaterina has a strong background in both practical and theoretical aspects of data science and molecular biology. Prior to her current role, she held positions as an associate data scientist at Janssen Biologics B.V. and a scientific data analyst at Petrovax, among others.


  • 17:30-18.30: Walk-ins & dinner
  • 18.30-18.35: Introduction to Xomnia
  • 18:35-19:10: Happy Nurses: solving the nurse scheduling problem using mathematical optimization
  • 19:10-19:25: Break
  • 19.25-20.00: Automated Feature Generation's Role in Unveiling the Mysteries of Biological Age
  • 20:00-21.00: Borrel