LOGIN
Singapore–MIT Alliance
Skip navigation
Skip navigation Skip main site navigation
 
Skip section subnavigation Skip breadcrumb trail
Go to main navigation

HPCES

New Frontiers in High Performance Computation


Description

The High Performance Computation for Engineered Systems (HPCES) degree programme is the most technologically advanced and critically acclaimed computational engineering coursework available in the world today. Through a powerful combination of state-of-the-art distance learning technology and premiere academic collaboration, the HPCES programmes are graduating the very best high performance computation professionals.

High performance computation for engineered systems is a crucial component in the modeling, simulation, design, optimisation, control and visualisation of engineered systems in a wide range of technology and service industries. HPCES courses promote creativity as well as hands-on experience in an effort to study the improvement of both product and systems design. The programme's unified approach combines engineering science and systems optimization:

Engineering science
A keen focus on modeling and simulating physical phenomena and product behavior helps students to uncover shorter design cycles and improve functionality. Such virtual testing allows industries to design innovative, quality products with a minimum of costly physical prototypes.

Systems optimisation
Careful attention to modeling and designing complex systems allow students to identify optimal configurations for maximum operational performance. The study of efficient process automation and integration is also emphasized. Such virtual design tools are widely used by industries to construct innovative solutions to complex tactical and strategic decisions.

The SMA academic programmes are also unique in their close affiliation with the IHPC, a premiere research institute in Singapore's Science Park. The IHPC specialises in research involving simulation and visualisation using advanced computational techniques. The Institute maintains close ties with the academia to undertake upstream research for the development of new technology, and at the same time supports local companies in industry-inspired research to enhance their capabilities and productivity.

| top of page |

Degrees

The Master of Science (S.M.) in HPCES
A professional master's degree programme that prepares graduates for careers in simulation, design and optimisation for engineered systems. This one-year programme focuses on the critical and effective application, modification, and integration of existing simulation and optimisation software. In addition, a two-week course of study at MIT is required.

The Doctor of Philosophy (Ph.D.) in HPCES
A research doctorate degree programme which emphasises the formulation, analysis and implementation of new computational methods for the simulation and optimisation of engineered systems for advanced technical careers in research and development. Completion of the Ph.D. programme may require three or more years. All Ph.D. students will have the opportunity to spend a semester at MIT to take courses and conduct research with MIT students and faculty.

| top of page |

Career Paths

The SMA programme in HPCES is the first of its kind to deliver a unified perspective on simulation and optimisation techniques in the domains of engineering science and systems optimisation. Students learn to develop and apply advanced techniques for a diverse range of applications in:

Courses are primarily for people with an interest in, and passion for, modern and sophisticated high performance computation tools as the means to improve product and systems design. Careers might include employment in companies or research institutes in which modeling, simulation, design, and optimisation play a critical role. With a unified perspective on simulation and optimisation techniques, graduates are poised to accept high-level professional or research positions with thriving industries or entrepreneurial businesses around the globe.

| top of page |

Programme Requirements

The S.M., M. Eng. and Ph.D. degree programmes contain the core curriculum, which includes the following courses:

SMA Project Course

Many of the courses will involve extensive hands-on experience. In addition to the SMA core curriculum, the M. Eng. degree requires a Master's thesis, and the Ph.D. degree
requires a Ph.D. thesis, as well as several additional advanced courses.

| top of page |

Courses

Summer Session

SMA 5200 Numerical Linear Algebra (12 Units)
A two-week long intensive review of linear algebra with an emphasis on topics related to numerical computation. The course meets for two hours of lectures in the morning and a three-hour project lab in the afternoon. Topics covered include a review of vectors, matrices, norms, range and null spaces, and orthogonality, least squares problems and the QR algorithm, Gaussian elimination, eigenvalues and eigenvectors, the singular value decomposition, and iterative solution methods.

SMA 5202 Computing Technology and Tools (12 units)
A hands-on course on the technology and software tools for high performance computation. Software packages used include Mathematica, C++ and/or Java, Excel, PETSs, AVS, PVM and MINDSET.

Fall Session

SMA 5211 Introduction to Numerical Simulation (12 units)
An introduction to computational techniques for the simulation of a large variety of engineering and engineered systems. Applications are drawn from aerospace, mechanical, electrical, and chemical engineering, as well as materials science and operations research. Topics include mathematical formulations; network problems; sparse direct and iterative matrix solution techniques; Newton iteration for nonlinear problems; solution techniques for eigenvalue problems; discretisation methods for ordinary differential equations and differential-alegebraic equations; discretisation methods for partial differential and stochastic partial differential equations; methods for the solution of integral equations; and Monte Carlo techniques and higher dimensional problems.

SMA 5212 Numerical Methods for Partial Differential Equations
Covers the fundamentals of modern numerical techniques for a wide range of linear and nonlinear elliptic, parabolic, and hyperbolic partial differential and integral equations. Topics include: mathematical formulations; finite difference, finite volume, finite element, and boundary element discretisation methods; and direct and iterative solution techniques. The methodologies described form the foundation for computational approaches to engineering systems involving heat transfer, solid mechanics, fluid dynamics, and electromagnetics. Computer assignments requiring programming.

| top of page |

SMA 5213 Optimisation Methods (12 units)
This course is an introduction to the principal methods for linear, network, discrete, nonlinear optimisation, as well as dynamic optimisation and optimal control. Emphasis is on methodology and the underlying mathematical structures and their connection to computational procedures.

Topics include:

SMA 5214 Numerical Algorithms on Advanced Computer Architectures (6 units)
The course introduces basic concepts of vector and parallel computer architectures and their influence on numerical algorithms that are commonly used in the numerical modeling of large-scale engineering simulation and optimisation problems. Elements of MPI/PVM are covered to enable the implementation of numerical algorithms written in FORTRAN and C/C++ on parallel computers. Hands-on parallel computation exercises on selected numerical algorithms are included to give insight into performance measures and the influence of computer architectures on the programming and performance of selected numerical algorithms.

SMA 5215 Integrated Simulation and Optimisation of Engineered Systems (6 units fall/6 units spring)
An integrated presentation of simulation and optimisation techniques for engineered systems. Emphasis is on understanding and exploiting underlying mathematical frameworks and associated computational methodologies common to both simulation and optimisation; and on the systematic application of optimisation methods to problems for which the forward analysis requires extensive simulation. Particular topics include optimisation formulations of unconstrained and constrained equilibrium problems; methods for unconstrained and constrained optimal control of systems governed by ordinary differential equations; formulation, approximation, and solution of inverse and parameter estimation problems; and optimisation of systems described or simulated by statistical or stochastic simulation techniques. Applications will be drawn from the engineering, operations research, and management and finance domains.

| top of page |

Spring Session

SMA 5212 Numerical Methods for Partial Differential Equations (12 units)
A presentation of the fundamentals of modern numerical techniques for a wide range of linear and nonlinear elliptic, parabolic and hyperbolic partial differential equations and integral equations central to a wide variety of applications in science, engineering, and other fields.

Topics include:

SMA 5216 Integrated Simulation and Optimisation of Engineered Systems Part II
An integrated presentation of simulation and optimisation techniques for engineered systems. Emphasis is on understanding and exploiting underlying mathematical frameworks and associated computational methodologies common to both simulation and optimisation; and on the systematic application of optimisation methods to problems for which the forward analysis requires extensive simulation. Particular topics include optimisation formulations of unconstrained and constrained equilibrium problems; methods for unconstrained and constrained optimal control of systems governed by ordinary differential equations; formulation, approximation, and solution of inverse and parameter estimation problems; and optimisation of systems described or simulated by statistical or stochastic simulation techniques. Applications will be drawn from the engineering, operations research, and management and finance domains.

SMA 5223 Systems Optimisation: Models and Computation
An applications-oriented course on the modeling of large-scale systems in decision-making domains and the optimisation of such systems using state-of-the-art optimisation tools. Application domains include: transportation and logistics planning, pattern classification and image processing, data mining, design of structures, scheduling in large systems, supply-chain management, financial engineering, and telecommunications systems planning. Modeling tools and techniques include linear, network, discrete and nonlinear optimization, heuristic methods, sensitivity and post-optimality analysis, decomposition methods for large-scale systems, and stochastic optimization.

SMA Project Class
Capstone subject providing hands-on technical skills and an opportunity for group project work and technical communication.

| top of page |
Go to top of pageGo to main navigationGo to additional information