Train driver experience: A big data analysis of learning and retaining the new ERTMS system

Area of Science:

• Railway Engineering
• Human Factors in Transportation
• Traffic Management Systems

Background:

• The European Rail Traffic Management System (ERTMS) Level 2 deployment necessitates significant changes for train drivers, including new interfaces and altered procedures.
• Effective training strategies are crucial for train operating companies to ensure drivers achieve and maintain ERTMS proficiency during phased roll-outs.

Purpose of the Study:

• To determine the minimum ERTMS driving hours/duties required for train drivers to become proficient (learning period) and maintain proficiency (retaining period).
• To provide quantifiable data for developing efficient training and rostering models for ERTMS implementation.

Main Methods:

• Big data analysis of train driver performance and HR data (2015-2017) from a large train operating company.
• Pseudonymization techniques were employed to maintain driver privacy while analyzing collective performance.
• Correlation analysis between operational delays and driver personal/training data to test hypotheses.

Main Results:

• Identified clear correlations between ERTMS driving exposure and driver proficiency, despite limitations in current ERTMS track availability.
• Quantifiable recommendations for the minimum amount of ERTMS driving per quarter are presented for the first year and subsequent years.
• Results are being translated into practical models for ERTMS training and rostering.

Conclusions:

• The study provides data-driven insights into the learning and retention curves for ERTMS driving.
• The findings support the development of optimized training schedules and operational rostering to manage ERTMS transition effectively.
• This research offers a unique, data-driven approach to address the human-factor challenges in large-scale railway system upgrades.