MF2024 Robust and Probabilistic Design 6.0 credits

• Educational level

  Second cycle

• Academic level (A-D)

  D

• Subject area

• Grade scale

  A, B, C, D, E, FX, F

• Course home page

• Course page on KTH Social

Learning outcomes

Probabilistic design is an engineering design methodology with the aim to produce high-quality products, by systematically studying the effects of variations in the design parameters on product performance. Robust design is a methodology for optimising this quality by making the performance of the product insensitive to variations in the manufacturing, material, operational, and environmental properties.

A student that has completed the course shall be able to:

• describe the characteristic properies of various design characteristics in statistical terms,

• assess the confidence interval of the assessed reliability of a technical system,

• find the type of probability distribution for a given set of data,

• describe the purpose for, the methodology of, and the output from Design of Experiments,

• define a testplan for a set of physical and numerical experiments,

• describe the purpose and steps for performing a Monte-Carlo simulation.

• use Monte Carlo simulations to analyse how the uncertainties in a models input variables affects the results from the model;

•  describe the purpose of Robust design and how it relates to optimization approaches,

• use the Robust design methodology to minimize the sensitivity of a technical response parameter to variations in a set of component design parameters,

• use the Robust design methodology to minimize the sensitivity of a technical response parameter to variations i a set of technical interaction parameters,

• use the Robust design methodology to minimize the sensitivity of an interactive response parameter to variations in a set of ergonomic parameters.

Course main content

Engineering statistics; distributions, Normal, exponetial, Weibull, confidence intervals

Design of experiments: physical and simulation experiments, suspended or censured tests

Probabilistic design; Monte-Carlo simulation (Matlab, Ansys) of performance variations caused by variations in design (manufacturing tolerances, material properies, geometric configuration), user (anthropometric data), and environmental parameters (humidity, electromagnetic fields, temperature, dust)

Robust design; minimizing performance variation due to variation in design parameters, human properties and environmental conditions

Disposition

• Twelve lectures (12 x 2 hours)

• Four laborations (4 x 2 hours)

• Two group assignments

• Two individual assignments

• A written examination

Eligibility

Qualified for studies in grade  3 and P: MF1012/4F1812, MF1013/4F1813, MF1014/4F1814 ellerM: MG1003/4G1162, MG1004/4G1163 T: MF1015/4F1815

Design and product realization-methods/Productrealization-design/Product realization for T and MF101X/MF102X/MF104X/MF111X/MF112X/MF114X/MF1025/MF1026 and MF2018 or a Bachelor in Mechanical Engineering.

Literature

• Handouts

• Clyde M. Creveling "Tolerance Design: A Handbook for Developing Optimal Specifications .

Examination

• INL1 - Assignments, 3.0 credits, grade scale: P, F
• TEN1 - Examination, 3.0 credits, grade scale: A, B, C, D, E, FX, F

• INL1 - Individual assignment, 4.0 credits, grade scale: A, B, C, D, E, FX, F

• TEN1 - Examination, 2.0 credits, grade scale: A, B, C, D, E, FX, F

Final grading based on INL1 and TEN1 requires passed laboratory work and group assignments

Offered by

ITM/Machine Design

Contact

Sergei Glavatskikh, 08-790 63 82, segla@kth.se

Examiner

Sergei Glavatskikh <segla@kth.se>

Version

Course plan valid from: Autumn 11.
Examination information valid from: Spring 08.