The use of fibre reinforced polymer (FRP) in the field of civil engineering has seen significant growth in recent years, providing efficient and reliable solutions for strengthening and repairing existing concrete structures. The performance of reinforced concrete (RC) beams strengthened with FRP using externally bonded (EB), near-surface mounted (NSM), and side near-surface mounted (SNSM) methods has been well documented in literature. However, all EB, NSM, and SNSM techniques have certain limitations. A hybrid technique combining EB and SNSM methods (EB-SNSM) has emerged to address these limitations by complementing each other and possibly mutually overcoming their limitations. Despite its potential, studies on the performance of the beams strengthened using EB-SNSM method are limited in the literature. This research aims to develop numerical models using FE-based software ABAQUS for FRP-strengthened RC beams using a novel approach considering the damage and failure of the FRP using the VUMAT user-subroutine. For different strengthening techniques, a comprehensive parametric study will be performed to find out the best configuration to effectively use FRP material, which will improve the load carrying capacity of the RC beam. Additionally, the research will incorporate machine learning (ML) to provide new insights into structural analysis and design. An ML-based prediction model will be developed to predict the strength behaviour of reinforced concrete beams strengthened using different techniques. An extensive and reliable database will be compiled from existing experimental data available in the literature for the ML model development. Present finite element (FE) models will also be used to generate more data required for the ML model. By developing all these models, the present research aims to enhance the understanding and performance of RC beams strengthened using different techniques.