Composite materials are widely utilized in manufacturing process of principal load bearing elements of various engineering structures. However, due to damage accumulation over the service life of the composite structures, their structural integrity are likely to diminish. To avoid from these undesirable failures by the way of condition-based maintenance scheduling, a structural health monitoring system that can perform real-time shape and stress sensing can be integrated to composite structures. For this purpose, in this study, a new four-node inverse-plate element (iRZT4) is developed using a computational algorithm based on the inverse finite element method (iFEM) coupled with the refined zigzag theory (RZT). The iFEM/iRZT4 formulation presented herein minimizes a weighted least-squares functional that uses membrane, bending, transverse shear, and zigzag section strains of RZT. The main benefit of the present approach for shape sensing is that it does not require any loading information and uses only strain measurements obtained from the on-board sensors. As for another advantage is that it is applicable to a general class of thin or thick laminated composite and sandwich plates. The high sensing capability and accuracy of the iRZT4 element is demonstrated by performing various numerical analyses of composite plates with different lamination sequences and comparing iFEM results against reference solutions.