Table 1 Comparison of key modeling... | Emerald Publishing

Table 1

Comparison of key modeling methodologies

Modeling methodology	Model accuracy	Applicable scenarios	Advantages	Disadvantages
BIM-GIS Integration	High	Railway station, line planning and design, 3D visualization management	Strong spatial expression ability, good compatibility with engineering data, convenient for full-life cycle management	Weak mechanical behavior simulation ability, high modeling cost for large-scale lines
Finite Element Method (FEM)	Very high	Bridge, tunnel structural health monitoring, stress and strain simulation	High mechanical simulation accuracy, suitable for complex structural analysis	Large computational load, slow real-time response, high requirement for parameter setting
Data-Driven (CNN/LSTM)	Medium-High	Track defect prediction, equipment performance degradation monitoring	Fast real-time response, strong adaptability to complex data, low modeling difficulty	Dependent on large-scale high-quality data, poor interpretability, weak physical constraint
Multi-Body Dynamics	High	Vehicle-track coupling system, turnout operation simulation	high consistency with actual operation state	Complex model establishment, high computational cost for long-term simulation
Hybrid Model (Physical-Data Driven)	Very high	High-speed railway key sections, complex equipment integrated monitoring	Balances simulation accuracy and computational efficiency, strong interpretability and adaptability	Complex model fusion technology, high requirement for cross-disciplinary knowledge

Modeling methodology	Model accuracy	Applicable scenarios	Advantages	Disadvantages
BIM-GIS Integration	High	Railway station, line planning and design, 3D visualization management	Strong spatial expression ability, good compatibility with engineering data, convenient for full-life cycle management	Weak mechanical behavior simulation ability, high modeling cost for large-scale lines
Finite Element Method (FEM)	Very high	Bridge, tunnel structural health monitoring, stress and strain simulation	High mechanical simulation accuracy, suitable for complex structural analysis	Large computational load, slow real-time response, high requirement for parameter setting
Data-Driven (CNN/LSTM)	Medium-High	Track defect prediction, equipment performance degradation monitoring	Fast real-time response, strong adaptability to complex data, low modeling difficulty	Dependent on large-scale high-quality data, poor interpretability, weak physical constraint
Multi-Body Dynamics	High	Vehicle-track coupling system, turnout operation simulation	high consistency with actual operation state	Complex model establishment, high computational cost for long-term simulation
Hybrid Model (Physical-Data Driven)	Very high	High-speed railway key sections, complex equipment integrated monitoring	Balances simulation accuracy and computational efficiency, strong interpretability and adaptability	Complex model fusion technology, high requirement for cross-disciplinary knowledge

Source(s): Compiled from [Djordjević, Krmac, Lin, Fröidh, & Kordnejad, 2024; Doubell, Kruger, Basson, & Conradie, 2021; Doubell, Basson, Kruger, & Conradie, 2023]