Generative Adversarial Network-based Missing Data Handling and Remaining Useful Life Estimation for Smart Train Control and Monitoring Systems

Table 1

Existing time, monitoring, and data-driven maintenance applications in rail systems.

Existing research studies	Target railway components	Methods and characteristics	Maintenance type
			Time/distance	Monitoring	Data-driven
Faiz and Singh [14]	Railway track	(i) Detection of track geometry; (ii) Usage of rail profile-based regression model	O	—	—
Sharma et al. [7]	Railway	(i) Vibration sensor-based estimation of railway breakage	—	O	—
Shaikh et al. [15]	Solid axle wheel sets	(i) Installation of additional sensors (vibration sensors for capturing lateral and yaw dynamics); (ii) Vibration model-based simulation	—	O	—
Letot et al. [16]	Railway track point machine	(i) Degradation assessment and data-based RUL estimation	—	—	O
Corman et al. [12]	Train breaking system	(i) Work, maintenance, and failure data-based reliability estimation; (ii) Usage of Weibull distribution	O	O	O