Wall-bounded flow is a crucial area of study in fluid dynamics, where turbulence plays a vital role in the interaction between fluid and solid boundaries. To effectively study turbulent behavior, appropriate boundary treatment is essential. Specifically, if wall functions are chosen to approximate asymptotic wall behavior, it can save computational costs compared to resolving the boundary layer. Extensive studies have been carried out on wall treatment and wall functions in mesh-based methods. In this study, we focus on Reynolds-averaged Navier-Stokes (RANS) turbulence modeling concerning a mesh-free Lagrangian method based on the Generalized Finite Difference Method (GFDM)[1,2]. Research and implementation of wall functions and boundary treatment for wall-bounded turbulence models in GFDM-based Lagrangian methods have received limited attention, leading to challenges in selecting the first point next to the boundary for wall function implementation. To address these issues, we propose three novel methods. The methods discussed will aid in boundary treatment and the application of wall functions for turbulence as well as the momentum equation. The first method involves selecting points based on their distance from the boundary rather than simply choosing the closest point, thereby avoiding holes or gaps in wall function implementation. The second method involves pseudo-shifting the boundary to a specified distance and treating these points as fluid points. The third method is an extension of the pseudo-shift method, where multiple pseudo-points are generated to capture large gradients at the boundary in the normal direction. The three novel methods are tested on simple 1D and 3D simulation cases, such as a flat plate and flow around a wing, using first-order turbulence models, namely: Spalart-Allmaras[3], k-ε, and k-ω[4] models. In this talk, we will highlight the challenges arising from incorporating boundary treatment with the proposed strategies and discuss the advantages and setbacks of each method.