Multidisciplinary Design Optimization Overview

Principles

Definition

Multidisciplinary design optimization is a field of research and a methodological approach focused on applying formal optimization techniques to the design of complex engineering systems involving multiple interacting disciplines. It investigates methods to simultaneously optimize system performance by coordinating design variables and objectives across these disciplines, often employing advanced mathematical algorithms to manage interdependencies and achieve globally optimal solutions.

Ontological type

Core Methods

Problem Formulations

Architecture Decomposition

Key developments

Key Figures

Integrated MDO Architectures era

During the Integrated MDO Architectures era of 1991–1997, representative figures such as James N. Haftka, Sobieszczanski-Sobieski, and Jose E. Martins advanced formal problem formulations and reusable architectures that coordinated cross-disciplinary analyses. Sobieszczanski-Sobieski pioneered cooperative design and decomposition-based strategies that enabled coarse-grained parallelism and structured workflows across disciplines. Haftka contributed architecture-level frameworks and surrogate modeling approaches to improve tractability for mixed-variable formulations within integrated optimization. Martins helped articulate disciplined MDO problem formulations and interfaces that standardized cross-domain solvers, shaping the integrated design environment for aerospace and mechanical engineering.

Surrogate-Driven Computational Synthesis era

Timothy W. Simpson [1] is associated with the University of Pennsylvania [3] and Washington University in St. Louis [4] during the Surrogate-Driven Computational Synthesis era. His key contribution is the Kriging Models for Global Approximation in Simulation-Based Multidisciplinary Design Optimization [7] (2001), which introduced Kriging-based global metamodels that enabled rapid surrogate-driven exploration and cross-disciplinary co-design in this era. Farrokh Mistree [2] is linked with the Georgia Institute of Technology [5] and the University at Buffalo, State University of New York [6] in this era. Mistree's work, exemplified by the same Kriging Models for Global Approximation in Simulation-Based Multidisciplinary Design Optimization [7] (2001), highlighted how surrogate metamodels empower robust, high-dimensional MDO workflows under uncertainty, aligning with the era's design and analysis progression.

Geometry-Integrated Multiobjective MDO era

Paul M. Weaver [1] is associated with Maastricht University [3] and the University of Bristol [4] during this era. One of his key contributions in this era comes from the 2017 paper '12th World Congress of Structural and Multidisciplinary Optimization' [7], which highlights advances in multidisciplinary design optimization and the role of geometry-aware frameworks in enabling CAD–FEM integration and multiobjective design. Joaquim R. R. A. Martins [2] is affiliated with the University of Southern California [5] and Michigan State University [6] during this era. His 2016 paper 'Multidisciplinary design optimization of large wind turbines—Technical, economic, and design challenges' [8] discusses integrating MDO for large-scale energy systems, emphasizing technical and economic challenges and the importance of robust, manufacturable designs in geometry-integrated workflows.