Skip to content


Track: Data Mining

Big Data: Concerns and New Methods for Distribution Fitting

Monday, April 12, 1:45-2:25pm EDT

The Danish railway authorities (Banedanmark and DSB) offer compensation to passengers if trains are delayed beyond a threshold. Over two million data points are recorded every month, and a typical station has over 14,000 data points per month. However, much of the published research on train delays concerns data sets of less than 1,000 points, and many of the common statistics tools and methods return misleading results when applied to larger data sets. In this talk, Dr. Harrod explains how to evaluate large data sets, and how to avoid misleading results. Topics include: evidence to consider beyond statistical tests, expanding your horizons to mixed distributions, and the Wilcoxon ranked sum test.

Steven Harrod image

Steven Harrod

Steven Harrod

Associate Professor at the Technical University of Denmark

Dr. Steven Harrod is a specialist in transportation management, transportation analysis, and supply chain networks. He has published numerous research papers in journals such as Transportation Science, Transportation Research Part E, International Journal of Production Economics, and Simulation. Dr. Harrod also has extensive experience in industry, including many years in road transport and information systems. He is especially knowledgeable about railway management and railway technology. He has extensive experience with, for example, OpenTrack railway traffic simulation software. Dr. Harrod currently is Associate Professor at the Technical University of Denmark. Dr. Harrod is active in teaching and research in transportation management topics. He is also active in groups such as the Railway Applications Section of INFORMS and the International Association of Railway Operations Researchers. Dr. Harrod obtained his PhD from the University of Cincinnati, and an MSc from MIT. Dr. Harrod holds a certification from the prior American Society of Transportation and Logistics (today merged with APICS), and is a Certified Analytics Professional at INFORMS. He is a member of the Analytics Certification Board of INFORMS (2020 – 2022).