Shortcoming in the current design guidelines on externally bonded FRP shear strengthened members has initiated a motivation to relook the whole shear design approach. It is understood that effective FRP strain models used in present design guidelines are basically calibrated from the experimental data based on the conservative and unrealistic 45-deg truss model. This paper is intended to propose an optimized truss model that derived from the principle of minimum total strain energy theorem to improve the present 45-deg truss model. The proposed optimized truss model is characterized with limiting failure criteria that reflects truly to the actual FRP strengthened beam behaviour. One of the most important failure criteria is the FRP debonding failure. To characterize it, limiting effective FRP strain εfrp,e model is incorporated into the optimized truss model. Six most recent effective strain models are chosen for the analysis, included three of the international design guidelines. Performance of each effective strain model will be evaluated along with the optimized and 45-deg truss models in order to assess their respective accuracy in predicting the FRP contribution to the shear strength. The validation of optimized truss model is done through comparing with experimental test results collected from the literature. The results obtained indicated that the optimized truss model is indeed more viable representative to the actual internal stress distribution and accurate than existing 45-deg truss model. So it might have a great potential to be used in the derivation of a new effective FRP strain model that can be implemented in the current design guidelines. © 2015 Elsevier Ltd. All rights reserved.