(1) No prerequisite in optimization is required.
MEDICINE: How does a doctor optimally radiate a brain tumor and not kill the surrounding brain tissue?
ENGINEERING: How does an engineer optimally size and arrange the truss elements of a bridge so that it is as strong as possible for a given allowable weight and cost, and avoids wind-induced resonance?
BUSINESS: Where should Wal-Mart optimally place its warehouses and distribution centers to maximize profits?
BIOLOGY: How do we optimally predict that a cell-sample is malignant or benign?
PLANNING: How does a powerplant optimally choose to produce electricity from coal, gas, and nuclear sources?
These are some of the question that you will learn how to formulate and solve in Systems Optimization: Models and Computation. Students from all departments are welcomed and encouraged to explore this exciting subject. The examples and topics are relevant to problems in business, medicine and biology, and engineering.
Final projects and hands-on applications and exercises facilitate the practical use of the techniques described in the lectures. You will develop your problem formulation and solution skills on real-world problems in homework assignments. You will formulate and solve a large-scale optimization problem according to your own interests in a final project.
The Official Description: This course is a computational and application-oriented introduction to optimization modeling of large-scale systems in a wide variety of decision-making domains. We focus on using state-of-the-art optimization software to solve models and understanding the limitations of the modeling and solution techniques. Application domains include transportation, logistics, and telecommunications system planning; pattern classification and artificial intelligence; structural and engineering design; and financial engineering. Model formulation techniques include linear optimization, nonlinear convex and non-convex optimization, as well as discrete optimization and conic semidefinite optimization. Solution techniques covered include decomposition methods, column and constraint generation, continuous mappings, stochastic programming, and conic optimization and solution methods via interior-point methods. Students will develop formulation and solution skills in homework assignments and will formulate and solve a problem aligned with their interests for a final project.
MIT Units: 3-0-9, Graduate H-level Grad Credit
Prerequisites: This course does not have required prerequisite in optimization.
Course Meeting Time and Location:
Lecture: Tuesday and Thursday, 2:30 PM - 4:00 PM in 3-333. Recitation: To be scheduled on an as necessary basis.
Course Materials: Text: Introduction to Linear Optimization, D. Bertsimas and J. Tsitsiklis, Athena Scientific, Belmont, MA, 1997, on reserve at Dewey Library (MIT).
Instructor: Brian W. Anthony TA : Sisir Koppaka
MIT Room 35-130 E-mail : koppaka@mit.edu
Telephone: 324-7437
E-mail: banthony@mit.edu