To evaluate the seismic performance of single-degree-of-freedom (SDOF) systems under mainshock–aftershock (MS–AS) seismic excitation, nonlinear time-history analyses were conducted on SDOF systems with various parameter combinations, using 50 sets of real MS–AS sequences and 150 sets of artificial sequences generated by repetition, random, and attenuation methods. The results indicate that the ground motion characteristics of MS–AS sequences generated by the repetition, random, and attenuation methods differ from those of real MS–AS sequences, with the repetition and random methods tending to overestimate the peak ground motion parameters and acceleration response spectra of MS–AS sequences, and the attenuation method potentially underestimating them, while all three methods for generating MS–AS sequences are prone to overestimating the ground motion duration of MS–AS sequences. Residual deformation is influenced by relative yield strength coefficient (η), aftershock relative intensity (χ), post-yield stiffness ratio (r), natural vibration period (T) and the hysteresis model under MS–AS seismic excitation, and residual deformation exhibits a positive dependence on aftershock intensity (χ) and a negative dependence on post-yield stiffness ratio (r), while the relationship between residual deformation and relative yield strength coefficient (η) is influenced by the natural vibration period (T), showing a positive correlation in the short-period range and a negative correlation in the mid-to-long period range. A log-normal distribution can be adopted to describe the probability distribution of the ratio of residual deformation to peak elastic-plastic deformation subjected to MS–AS seismic excitation with different parameters. Finally, a probabilistic prediction model for residual deformation under MS–AS seismic excitation was proposed which can effectively predict residual deformation under MS–AS seismic excitation.

Against the backdrop of low-carbon transportation and urban logistics transformation, metro-based freight is regarded as an important pathway for emission reduction. This paper constructs a tripartite evolutionary game model involving the government, logistics enterprises, and metro operators, and analyzes multi-agent strategy evolution and the influence of key parameters using replicator dynamics equations and numerical simulation. The results show that well-designed subsidies and penalties can effectively promote a stable state characterized by “active government intervention, active response from logistics enterprises, and low-carbon integrated passenger and freight transportation by metro operators”. Reducing the cost of transformation can improve evolutionary efficiency, while excessively high subsidies may weaken the government’s willingness to intervene. This study provides insights for optimizing low-carbon transportation policies and supporting the development of metro-based freight systems.