Naslov
DEFINITION OF NEW CRITERION IN DESIGN OF EXPERIMENTS BASED ON MODEL PRECISION

Autori
Cajner, Hrvoje ; Šakić, Nikola ; Ćosić, Predrag

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Management of Technology Step to Sustainable Production MOTSP2012 Conference Proceedings / Ćosić, Predrag - Zagreb : Croatian Association for PLM (CAPLM), 2012, 145-151

Skup
Management of Technology Step to Sustainable Production MOTSP2012

Mjesto i datum
Zadar, Hrvatska, 14.06.2012. - 16.06.2012

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
design of experiments; optimality criteria; optimal design; augmented design

Sažetak
Process of designing the experimental models consists of choosing the appropriate number and placement of experimental points. Taking into account the placement and replication of design points, the predicted variance and confidence interval of the model can be evaluated. The tendency is to achieve more precise model by adding more design points. Using the known optimality criteria, based on prediction of variance, it is possible to define new criterion of model precision. The paper will present the definition of incremental model precision criterion based on difference between the base model (initial) and modified model (the model with added design points - augmented design).

Izvorni jezik
Engleski

Znanstvena područja
Strojarstvo

POVEZANOST RADA