Unlike many other surface treatment techniques, such as shot peening (SP), low plasticity burnishing (LPB), ultrasonic surface rolling process (USRP), ultrasonic nanocrystal surface modification (UNSM), laser shock peening (LSP) is a non-contact process. This means that LSP achieves surface modification without direct physical contact, thereby eliminating the risk of media contamination and undesired wear on the peening equipment. As a surface hardening technology, LSP tends to produce a hardened layer with a high gradient of residual stress (RS) in the near-surface, significantly enhancing wear resistance. The well-known wear theories applied for numerical prediction, whether Archard model or dissipated frictional-energy model, contain a multiplier for contact pressure and relative slip amplitude p·ds or shear stress and relative slip amplitude q·ds, respectively. These multipliers are sensitive to the mechanical properties near the surface. Additionally, the wear profile is governed not only by the wear effect but also by the deformation of materials. Therefore, to accurately evaluate the wear performance for surface-hardened materials, the influence of RS is hard to be overlooked. In this work, the finite element method (FEM) is utilised to construct a 3D numerical model, which ignores and considers RS, for both as-received and LSP-treated samples. Two distinct RS distributions induced by different scanning strategies (one- and five-time), are introduced in the finite element (FE) model. The numerical simulation results are then compared with the experimental data to verify the accuracy of the fretting wear model built in this study. On this basis, the impact of different RS distributions on fretting wear characteristics, including wear profile and volume, is extensively investigated. The findings reveal that higher amplitude RS can reduce the degree of fretting wear to a greater extent, highlighting the beneficial role of RS in enhancing wear resistance. Furthermore, the numerical models incorporating RS demonstrate superior predictive accuracy compared to those that omit RS. This naturally underscores the importance of integrating RS in numerical simulations of LSP and other surface-hardened materials.