Bringing economy and robustness in parameter testing: a Taguchi methods-based approach to model validation

This paper presents the application of an effective and practical method to complement the methodology for validating system dynamics (SD) models. The method is based on an analogy drawn between the SD model development process and the quality design and manufacturing product development process of an industrial context. Employing this metaphor, the revolutionary Taguchi principles of robust design can be utilized at the beginning of the validation process of SD models for parameter testing. The proposed Taguchi-based approach is applied and tested in the context of a SD corporate model. The results show that, with relatively few steps at the beginning of the validation process and in the sub-category of reference behaviour tests, the modeller can significantly reduce complexity of testing while increasing confidence in his model.