The micro mechanical model by Gurson-Tvergaard-Needleman is widely used for the prediction of ductile fracture. Some material properties (Gurson parameters) used as material input in this model for simulation are estimated experimentally from specimen level. In this article an attempt has been made to tune the values of some of these Gurson's parameters by comparing the simulated results with the experimental results in the specimen level (axisymmetric tensile bar and CT specimens). An elastic plastic finite element code has been developed together with Gurson-Tvergaard-Needleman model for void nucleation and growth. The initial value of f(c) is determined from Thomason's limit load model and then tuned on the basis of best prediction of the failure of one-dimensional tensile bar. Then the load versus load line displacement and J versus Delta a results for CT specimen are generated with the same code and the value of f(n) is tuned to match the simulated J versus Da results with the experimental results. Lastly the same code and the Gurson's parameters obtained are used to simulate the load versus load point displacement and crack growth for pipe with circumferential crack under four point bending. The simulated results are compared with the experimental results to assess the applicability of the whole method. In the proposed material modelling, post-yielding phenomena and necking of the tensile bar are simulated accordingly and strain softening due to void nucleation and growth has been taken care of properly and drop in stress is implicitly simulated through a model. Incremental plasticity theory with arc length method is used for the nonlinear displacement control problem.