A generic optimizer interface for programming-free optimization systems

In this paper, a new mechanism called the generic optimizer interface is presented to bridge different optimization problems with various optimizers for programming-free optimization systems. With this generic interface, an optimization problem is given independent of optimizers and can be solved by different optimizers linked to the interface. Several types of function representations for the generic interface are developed and analyzed, and the indexed binary tree representation (IBTR) is identified to be both efficient and convenient. Different formulations for solving multi-objective optimization problems are then implemented with high computational efficiency by taking advantage of the IBTR. The generic interface is object-oriented and can be easily extended to add new optimizers and formulations.