The conventional design sensitivity analysis (DSA) methods have been widely used for gradient-based optimization, under the limitation of small design perturbations [1]. Using the first order Taylor series expansion with respect to a design parameter, configuration designs are additively decomposed into shape and orientation variations, for which the configuration DSA is independently performed. Therefore the configuration design sensitivity is valid only for small design perturbations, and sometimes provides the inaccurate direction of design updates for configuration design optimization. Moreover, it is challenging and even impossible to decompose the design velocities of shape and orientation variations for curved structures. We propose an isogeometric design gradient tensor for the accurate and efficient computation of configuration DSA. The design gradient tensor is defined as the gradient of line elements of the perturbed design with respect to original design. The polar decomposition of design gradient tensor then provides the design rotation and stretch tensors under arbitrary design variations. Since the polar decomposition is valid regardless of perturbation amount, the proposed approach is not limited to the small variations. When combined with the isogeometric framework [2], the perturbation of control points provides the design gradient field, where the exact rotation and stretch tensors are multiplicatively coupled. It is well known that the exact representation of geometry is crucial for the shape DSA [3]. Therefore, the isogeometric design gradient tensor naturally leads to accurate configuration DSA for arbitrary perturbation field without the need to decompose the shape and orienation design variations. We show that the proposed method fits with the conventional approach for flat elements under small design perturbations. Moreover, we demonstrate the feasibility of the proposed method for curved elements under arbitrary design variations by comparing with the fintie difference sensitivity. REFERENCES [1]K.K. Choi, N.H. Kim, Structural Sensitivity Analysis and Optimization 1, Springer, 2005. [2]T.J.R. Hughes, J.A. Cottrell, Y. Bazilev, Isogeometric analysis: CAD, finite elements, NURBS, exact geometry and mesh refinement, Computer Methods in Applied Mechanics and Engineering. 194 (2005). [3]S. Cho, S.H. Ha, Isogeometric shape design optimization: Exact geometry and enhanced sensitivity, Structural and Multidisciplinary Optimization, 38 (2