Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Production Engineering Archives

2 2018

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: Taguchi robust design

Sortuj według:

Ogranicz wyniki do:

The Response Surface Methodology revisited – comparison of analytical and non-parametric approaches

Osocha P.

Production Engineering Archives

2018

Vol. 20

49--53

Since G.E.P. Box introduced central composite designs in early fifties of 20th century, the classic design of experiments (DoE) utilizes response surface models (RSM), however usually limited to the simple form of low-degree polynomials. In the case of small size datasets, the conformity with the normal distribution has very weak reliability and it leads to very uncertain assessment of a parameter statistical significance. The bootstrap approach appears to be better solution than – theoretically proved but only asymptotically equal – t distribution based evaluation. The authors presents the comparison of the RSM model evaluated by a classic method and bootstrap approach.

The outline of the expert system for the design of experiment

Dwornicka R., Pietraszek J.

Production Engineering Archives

2018

Vol. 20

43--48

The design of experiment (DoE) is a methodology originated from early 1920s when Fisher’s papers created the analysis of variance and first known experimental designs: latin squares. It is focused on a construction of empirical models based on measurements obtained from specifically structured and driven experiments. Its development resulted in the constitution of four distinctive branches recognized by the industry: factorials (full or fractional), Taguchi’s robust design, Shainin’s Red-X® and a response surface methodology (RSM). On one hand, the well-known success stories of this methodology implementations promise great benefits, while on other hand, the mathematical complexity of mathematical and statistical assumptions very often lead to improper use and wrong inferences. The possible solution to avoid such mistakes is the expert system supporting the design of experiments and subsequently the analysis of obtained data. The authors propose the outline of such system and provides the general analysis of the ontology and related inference rules.