MIT Department of Electrical Engineering & Computer Science

EECS Spring 1998 Catalogue Supplement

6.291 Quantitative Methods in Systems Engineering

TR 2:30-4, 4-149
Visiting Prof. John Baras, 35-414, x2340 and Prof. Sanjoy Mitter, 35-308, x2160
3-0-9

Prerequisites: Basic understanding of engineering mathematics including calculus, probability and linear algebra, undergraduate education in any of the engineering disciplines, undergraduate background in control and optimization, basic knowledge in modeling and simulation, ability to use MATLAB to formulate and solve engineering problems.

Course Objective: The purpose of this course is to introduce graduate students in the School of Engineering to a set of structured methodologies for handling and managing the process of synthesizing complex engineering systems. The methods are quantitative in nature and have evolved over the last ten years through practice in industry and research in academia. They address the five key areas of systems engineering: system requirements capture, system architecture, system modeling, trade-off analysis, system verification and validation.

Rationale: Modern and future emphasis on cost reduction, quality improvement, time-to-market speed up, and linkage to market analysis as early as possible in the design of a product life-cycle have brought to the forefront the significance and criticality of Systems Engineering education for all branches of engineering. Systems engineering needs to be applied in all product design whether the product is a complex computer chip, or a chemical process reactor, or a telecommunication network or a smart material. Systems engineering needs to be applied in all complex systems operations and management whether the system is a telecommunication network, a chemical plant, a semiconductor fabrication facility or a satellite constellation.

The need for comprehensive education in systems synthesis has been further emphasized in recent reports on education by the National Research Council, and by the Council of Engineering Deans; the latter report effort was co-chaired by MIT President Charles Vest and by then Lockheed Martin Chairman and CEO Norman Augustine.

Text: No text. Extensive references will be provided as well notes by the instructors. Substantial part of the material will be presented for the first time in course format.

Guest Lecturers: The course will feature a few lectures by specialists on certain key topics.

Course Outline: Conceptually the course is based on fundamental ideas from control systems and optimization, communication and information systems, and relevant areas of computer science such as modeling languages, object oriented modeling, search planning and constraint based reasoning, databases. These techniques are blended together to produce a well integrated whole and a foundation of what todate has been more of an "art."