An enhancement implemented in Creo Elements/Pro 5.0 (Wildfire 5.0) brings the capability to execute statistical studies. This capability is part of the Behavioral Modeling functionality.
In previous Creo Elements/Pro releases, you can define Multi-Objective Design Studies (MODS) to help find optimal solutions that satisfy several design goals. For example, you can investigate possible shapes of a fuel tank that let you maintain the volume of fluid inside and certain structural properties (less stress or deformation) within a desired range.
In Creo Elements Pro 5.0, Statistical Design Study allows you to assign statistical distributions to dimensions and parameters that are design variables and to parameters that are design goals of a Multi-Objective Design Study (MODS). Using the response surface corresponding to a MODS and the statistical distributions assigned to the corresponding design variables; Creo Elements/Pro runs a large number of approximate experiments to determine the distribution parameters of the design goals of MODS. Thus, you can use the results of a statistical design study to find a ‘close to optimal’ operating point for your design goals.
In Creo Elements/Pro there are five available types of statistical distributions (Exponential, Lognormal, Normal, Uniform and Weibull). You can select a type of statistical distribution for the design goals of a MODS. Also, there are two methods of conducting a statistical design study: Single and MODS. There are more details about them in the Creo Elements/Pro Help documentation.
Let’s try this in a simple example. The goal is to understand the distribution of the height of the spring due to the tolerances of the length and angle dimensions.
- First, calculate the Standard Deviations of the dimensional tolerances (length and angle). Mathcad or any other mathematical tools can help you accomplish that. The computed values will be used in the Creo Elements/Pro Statistical Design Study
- We prepare the dimensions in the Creo Elements/Pro model and create a Distance analysis feature for the height
- We execute an MODS analysis (found under >Analysis >Multi-Objective Design Study) in Creo Elements/Pro to build a ‘response surface’. This tells how sensitive the height measurement is to changes of the spring length and angle. Save the study
- Then, in Creo Elements/Pro select >Analysis >Statistical Design Study. Set up the design variables and assign Standard Deviations to them
- Finally, compute the distribution of height. We now can do that since we know the Standard Deviations of the driving dimensions and the sensitivity of height to changes for these variables. We learn the Mean Value of the height (given the statistical distribution of length and angle) is 2.217 mm and the Standard Deviation is 0.02133 mm
- Optionally - if we have design limits of what is acceptable for height- we can enter these values and learn how many defects per million units (DPMU) one would expect given these limits. Also, how many sigma’s (Standard Deviations) these limits are away from the Mean Value which is important for quality initiatives in a company