Strategic Partner QAMPO Win Rail Problem Solving Problem

Share on facebook
Share on twitter
Share on linkedin
Pictured: Award Winners: Evelien van der Hurk (PhD), Jørgen Thorlund Haahr (PhD) and Erik Hellsten, PhD candidate


Signature Rail’s vehicle optimizer strategic partner wins first place in Rail Problem Solving Competition

Over 40 teams participated in this year’s Rail Problem Competition, which seeks to predict delays in the Netherlands Railway passenger rail network. The three finalists used three very diverse methods.  The titles, abstracts and contributors are below. They will be presenting on Sunday, November 4, 2018 at the INFORMS Phoenix conference for first, second and third place.

Award Order:

First Place: Forecasting Train Delays in the Netherlands using Neural Networks

Abstract: We investigate to what extent low-maintenance and out-of-the box machine learning models can provide accurate predictions of train delays. We focus on predicting actual delay and but also the delay development. The results on real-life data from the Netherlands indicate that our models can outperform a constant prediction model.


  1. Jørgen Thorlund Haahr (PhD), Decision Scientist, QAMPO
  2. Erik Hellsten, PhD candidate, Technical University of Denmark
  3. Evelien van der Hurk (PhD), Assistant Professor, Technical University of Denmark

Second Place: Predicting Near-Term Train Schedule Performance and Delay Using Bi-Level Random Forests 

Abstract: Near-term train delays prediction is critical for railway management. We propose a bi-level random forest approach to predict train delays. The primary level predicts the delay category, and the secondary level estimates the delay (in minutes). The proposed model is compared with several alternative approaches, validating its superior accuracy. 


  1. Mohammad Amin Nabian, PhD Candidate, Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign,
  2. Negin Alemazkoor, PhD Candidate, Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign
  3. Hadi Meidani, Assistant Professor, Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign

Third Place: A Railway Delay Prediction Model Based on Non-Homogeneous Markov Chains

Abstract: Assuming the delay on a certain station depends only on delay attained in the previous station, we model the delay evolution over stations as a non-homogenous Markov chain. By discretizing the state space and retrieving transition matrices from historical data, we accurately predict delay using a probabilistic approach.


  1. Gao, Zheming,  Department of Operations Research, NC State University
  2. Luo, Haochen,  ISEN Department, Texas A&M University
  3. Wu, Qian,   ISEN Department, Texas A&M University
  4. Xu, Jin,  ISEN Department, Texas A&M University

Original Article –

More to explore...

Photo of a Vy train

Data and Integration for Efficiency, Passenger Experience and a Greener Future

“In rail, something always goes wrong. We often say that Murphy works in rail. But not only Murphy; his whole family too!”

This comment from Vy’s (formerly Norwegian State Railways, NSB) Peter Hausken perfectly captures his keen sense of humour.

But dig a little deeper and you realise it also underlines what has made him such a powerful advocate for change.

Vy case study image of child and man getting on train

Customer Case Study: Vy

With multiple technology solutions across different areas, Vy wanted to ensure all systems worked in harmony, supporting the production of effective plans that flow through to the day of operation. Special Advisor Peter Hausken explains how the team uses technology to manage disruptions, enabling better informed decisions and ensuring accuracy of information to passengers.

serious disruption for rail planners chart

The UK Rail Sector Visualised

Antony Chilton-Murphy identifies current disrupters and drivers of change, and outlines how technology can serve as an ‘enabler’ – with specific focus on decision science as a way to manage the incredibly diverse and complex variables faced by today’s train planners.