---

The Incremental Sheet Metal Forming (ISMF) process is a new and flexible method that is well suited for small batch production or prototyping. In this study, after the process simulation with ABAQUS software and verification of results through experimental tests, the effects of three parameters including friction coefficient, tool diameter and vertical step size on three objectives including vertical force, minimum thickness of deformed sheet and amount of spring-back are investigated. A neural-network model is developed based on simulation data and the effects of parameters are studied on each objective. Also multi-objective genetic algorithm is performed to get the Pareto front of optimum points.

---

Cracks due to manufacturing processes or in-service applications can propagate and cause failure in structures. Therefore, it is of interest to find a suitable fracture assessment method for predicting crack initiation. Main approaches for fracture assessment of structures are global approach and local approach. In the global approach, it is assumed resistance against fracture can be measured by a critical values of a far from crack tip parameter like K or J. In this study, Beremin model of local approach is used for predicting brittle fracture which studies stress and strain fields at the crack tip. The model introduces unknown parameters which have to be calibrated using experimental fracture data. The purpose of this study is evaluating of conventional calibration methods of local approach parameters using the experimental brittle fracture data of three point bending specimens, determining limitations, and finally presenting a new calibration method to produce suitable parameters for predicting brittle fracture of the specimens by using local approach to fracture. This study shows that conventional calibration method using experimental fracture data of three point bending specimens has limitation in some cases. Also, by introducing location parameter of Weibull distribution as a stress triaxiality criteria in Beremin model, a new rational method for predicting brittle fracture of the three point bending specimens with different constraints is presented.

---

Failure of mechanical structures may be caused by presence of some defects like cracks. These defects are generally created due to manufacturing processes or in-service applications. Failures due to cleavage fracture usually lead to catastrophic effects, thus studies of such failures are important. Main approaches for fracture assessment of structures are global approach (classical fracture mechanics) and local approach. In the previously presented models of local approach, unknown parameters are introduced which have to be calibrated using experimental fracture data. Despite of existing different calibration methods, obtaining suitable parameters for predicting brittle fracture based on local approach, has been limited in some cases. The purpose of this study is presenting a rational method for predicting brittle fracture in specimens of different shapes to transform it into full scale cases. In this paper, by considering the location parameter of Weibull distribution as a stress triaxiality criteria and modifying the Beremin model, predicting brittle fracture in specimens of different shapes are studied. Also, independence of the parameters from their shapes is shown and eventually a linear relation between location parameter of Weibull distribution and triaxiality factor for the material is presented.