On the Use of Analytical Target Cascading and Collaborative Optimization for Complex System Design

The methods of analytical target cascading (ATC) and collaborative optimization (CO) are studied with respect to their intended use and applicability. ATC was initially developed as a product development tool, while CO evolved from efforts to coordinate multidisciplinary analyses using a multidisciplinary design optimization (MDO) formulation. ATC and CO are used typically to solve object-based and discipline-based decomposed system optimization problems, respectively. Their mathematical formulations appear to be similar although they were developed with different motivations. The article defines and compares terminologies for each approach and shows that each has unique applicability and solution process. Two new analytical example problems are employed to illustrate the distinctions between ATC and CO. The first example elucidates how each method can be used to solve the same design problem. The second example uses a new optimization formulation, nested ATC-MDO, to illustrate their complementary nature. In this formulation a system is partitioned by object, and objects are partitioned by discipline. ATC coordinates the design of system elements and an MDO method, such as CO, coordinates the multidisciplinary design of an element. The formulation maps well to organizational matrix structures. The overall study demonstrates the benefits of using complementary solution strategies in solving complex system optimization problems.