A statistical dimension and tolerance design for mechanical assembly under thermal impact

A. Jeang*, Tin-Chih Chen, C. L. Hwan

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

25 Scopus citations


Most materials change length as they change temperature. As as a result of this change, the dimensions and tolerances of a product become at variance with the design values. Hence, thermal effects must be taken into account when designing a product that will undergo temperature cycling. In this regard, a simultaneous optimisation of component dimensions and tolerance values at various temperatures is needed. The approach first generates a set of experimental data through Monte Carlo simulation, based on various combinations of parameters and tolerance levels as inputs. Then, the data are converted into a total cost as response values before applying response surface methodology (RSM) for statistical analysis and optimisation. The response value includes quality loss, tolerance cost, and failure cost, which reflect the combined effect of the parameter and tolerance values assigned. The results provide designers with optimal component parameters and tolerance values, and the critical components and the response function. The approach can also guarantee that the parameter and tolerance values found remain within tolerance for the temperature variation. Then, the product can function as intended under a wide range of temperature conditions for the duration of its life.

Original languageEnglish
Pages (from-to)907-915
Number of pages9
JournalInternational Journal of Advanced Manufacturing Technology
Issue number12
StatePublished - 1 Dec 2002


  • Design
  • Parameter
  • Robust
  • RSM
  • Thermal impact
  • Tolerance

Fingerprint Dive into the research topics of 'A statistical dimension and tolerance design for mechanical assembly under thermal impact'. Together they form a unique fingerprint.

Cite this