SYSEN 5350 - Multidisciplinary Design Optimization

(crosslisted) MAE 5350

Fall. 4 credits. Student option grading.

Prerequisite: undergraduate linear algebra and knowledge of MATLAB, Python or R.

M. Haji.

This course presents a rigorous, quantitative multidisciplinary design methodology that incorporates the creative side of the design process. Through a topic of your choice, learn how to use multidisciplinary design optimization (MDO) to create advanced and complex engineering systems that must be competitive in performance and life-cycle value. Multidisciplinary design aspects appear frequently during the conceptual and preliminary design of complex new systems and products, where different disciplines (e.g. structures, aerodynamics, controls, optics, costing, manufacturing, environmental science, marketing, etc.) have to be tightly coupled in order to arrive at a competitive solution. This course is designed to be fundamentally different from most traditional university optimization courses which focus mainly on the mathematics and algorithms for search. Focus will be equally strong on all three aspects of the problem: (i) the multidisciplinary character of engineering systems, (ii) design of these complex systems, and (iii) tools for optimization. Students will demonstrate mastery of the subject by working in small teams on a term project to apply the multidisciplinary design optimization principles to design and optimize an engineering system of their choice.

Outcome 1: Subdivide a complex system into smaller disciplinary models, manage their interfaces and reintegrate them into an overall system model.

Outcome 2: Identify the most suitable optimization algorithm between gradient-based numerical optimization algorithms (i.e. sequential quadratic programming (SQP)) and various modern heuristic optimization techniques (i.e. simulated annealing (SA) or genetic algorithms (GA)) for their design problem and use it to find the optimal design for a single objective of their choice.

Outcome 3: Critically evaluate and interpret analysis and optimization results, including sensitivity analysis and exploration of performance, cost and risk tradeoffs.

Outcome 4: Utilize basic concepts of multi-objective optimization, including the conditions for optimality and Pareto front computation techniques, to optimize their design with respect to two objectives of their choice.

Outcome 5: Work as a team to formulate a realistic engineering design problem, optimize the design for a single objective and multiple competing objectives, and present the results in a final oral presentation and written report.