Human Factors Phase III: Effects of Train Control Technology on Operator Performance

Lanzilotta, E.; Sheridan, T. · 2005 · ROSA P / United States. Department of Transportation. Federal Railroad Administration

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study investigates the impact of train control automation on locomotive engineer performance, specifically focusing on how supervisory control systems affect operator response to emergencies. As high-speed rail operations increase the cognitive demands on engineers, automation is proposed to reduce workload by handling speed regulation and position control. However, this shift raises concerns about "out-of-the-loop" performance, where reduced direct interaction with vehicle controls may lead to complacency or degraded situation awareness, potentially compromising safety during fault conditions. The research aims to determine whether automation improves monitoring capabilities by freeing mental resources or degrades them by reducing engagement. The researchers conducted a human-in-the-loop simulation experiment using twelve subjects from the general student population. Participants underwent a six-hour training course to familiarize themselves with locomotive engineer tasks before completing nine hours of experimental testing across three sessions. The study compared three levels of automation: fully manual control, partial automation (cruise control and programmed stop), and full automation (autopilot). During the sessions, subjects were exposed to three types of unexpected, recoverable emergencies: brake failure, traction motor failure, and grade crossing obstruction. The primary dependent variables were response time and response accuracy to these failures. The results indicated a significant difference in the variance of response times for brake and traction motor failures under the partial automation condition compared to manual and full automation conditions. This increased variance suggests that engineers using cruise control tended to bias their visual attention toward the external environment rather than monitoring the instrument panel, leading to inconsistent detection of internal system faults. Conversely, there was no significant difference in response time or variance for grade crossing obstructions across any automation level, indicating that external monitoring remained effective regardless of control mode. No significant differences were found in response accuracy, though the sample size was insufficient for definitive statistical conclusions on this metric. Subjective evaluations revealed that participants perceived full automation as resulting in the lowest level of situation awareness, while manual control provided the highest. The findings imply that partial automation, specifically cruise control, may inadvertently reduce attention to in-cab instrument displays, potentially delaying the detection of internal system failures. However, automation did not adversely affect the ability to monitor external hazards like grade crossings, which represent a significant risk in rail operations. The study highlights the complex relationship between workload reduction and situational awareness, suggesting that while automation can streamline routine tasks, it requires careful design to ensure operators maintain adequate monitoring of critical system states. These insights inform the development of safety regulations and human-machine interface designs for high-speed ground transportation systems.

Key finding

Partial automation (cruise control) significantly increased the variance in response times to brake and traction motor failures, indicating a bias toward monitoring the external environment instead of the instrument panel.

Methodology

simulator

Sample size: 12

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• rail grade crossings
• automation complacency bias