Quantitative Human Reliability Analysis (HRA) models aims to predict Human Error Probabilities (HEPs) for a given set of conditions. Owing to lack of empirical data most HRA methods use expert data. This paper presents a protocol to obtain empirical data from simulator experimentations. It aims to capture the effect of Performance Shaping Factors’ (PSF) states on human performance. Objective criteria is used to characterize the performance against system-level goals (safe operation and on-time service). This work’s domain of application is railway, mainly the train driver. Therefore, operational simulator implementing ERTMS (European Railway Traffic Management System) is employed to perform the experiments and gather data. Some experimental results with subjects are also presented. The data from such protocols is aimed at being usable for HRA modelling towards the quantitative risk analysis of human interaction on railway operations. © 2017 Taylor & Francis Group, London.