At any given point on a closed surface mesh, the Shape Diameter, a scalar function measuring the diameter of the largest fit circle or sphere, provides insight for shape analysis and, in our case, enforces a minimum feature size for a shape optimization problem. The shape diameter of a surface mesh and/or point clouds can be computed using the conical ray-cast approach, offset curve method, a point collision test, etc. Among various approaches available to calculate the shape diameter for both surface meshes, the current work implements a conical ray-casting-based approach due to its robustness. In the present work, the shape diameter represents a local thickness measure that is used to construct an optimization problem with a penalty term for minimum thickness violation. The current implementation integrates the ArborX library for constructing a bounding volume hierarchy with in-house optimization code for multiphysics optimization. Various numerical experiments highlight the sensitivity of current implementation with various parameters such as the number of rays within the cone and the angle of the cone. However, the improvement in terms of measuring local thickness just in the direction of normal compared to this current shape diameter approach reinforces the potential of using shape diameter for feature size control and regularization in shape optimization problems.