Descriptive design structure matrices for improved system dynamics qualitative modeling

Abstract

Qualitative modeling approaches can be useful in system information collection, model analysis, and formal model development. This is difficult when the number of elements and their interactions in the system is large. System dynamicists need additional tools and methods to conceptually model these large tightly coupled systems. We propose and test the Descriptive Design Structure Matrix (DDSM) as a qualitative system dynamics modeling tool and approach for systems with many elements and more interactions that can reasonably be modeled individually using traditional system dynamics methods. A DDSM consists of four parallel and internally consistent matrices that describe system interactions with binary relations, nontechnical text, technical text, and literature support. By including and documenting system information in multiple forms, DDSMs facilitate multiple stages of system dynamics modeling, improve modeler communication with system participants and domain experts, and improve model rigor. DDSM construction is described. A case study of the 2014 flooding in Kashmir is used to illustrate and test a DDSM and its application. Due to their compact format, DDSMs provide a useful visual communication aid, intuitive reasoning tool, and foundation for formal system dynamics modeling and analysis. © 2024 The Authors. System Dynamics Review published by John Wiley & Sons Ltd on behalf of System Dynamics Society.