Pre-Grant Publication Number: 20090048815
Collaborate on the process of community review for this application. Posting will not be forwarded to the USPTO.
Flagging a post as an ACTION ITEM signals further research. Flagging SPAM and ABUSE helps to manage discussion.
Placing double brackets around a reference to a claim or prior art will create a hyperlink to the original
ex. [[claim 1]] and [[prior art 2]].
Facilitator's Comment 
Action Item Show without Noise
I did invite Professor Olson to participate via the P2P tool.
1) While automating decision making to facilitate use by non-expert users
is a worthy goal, this should not be patentable, any more than automating
decision making in the preparation of taxes for non-experts or automating
investment choices by optimizing the Black-Scholes PDE.
2) They should be more specific as to how the code decides on the
constitutive model. Does it compare the numerical
deformation/stress data with actual data?
"The method and system may then compare the output of the simulation
to a set of test results ..."
Fig. 3C, block 366 and 370 are "Fit to Test Data" and "Fit Optimized?"
which imply some comparison of data, but doesn't say exactly which.
More will be said about this below in 5).
3) One of the concepts of the "generalized constitutive model" can be
seen in the lines:
"it may be possible to use a common coefficient matrix for all materials."
The concept of having a very general model containing every conceivable
term with coefficients that can be tuned to match data is already a
well established in FEA. Most existing FEA software allows for the
mixing of different elements, with the ultimate effect that the entire
structure will have a degree of freedom for the corresponding loading,
and FEA software removes or includes terms based on the boundary
conditions.
4) This patent has parts that may involve expert interaction:
"As but one example of the manner in which knowledge, expertise and
experience is incorporated in to the system 100...
while following another structure for other materials."
They then describe possible examples where the choice of one material
leads to subcategories which then lead to further subcategories. But this
branching structure is essentially the decision making process which
experts have to do with or without software.
5) For expert users who may perform experimental testing to get data for
the software to choose the correct constitutive model, the description
gives the impression that it merely automates the process an expert
would carry out otherwise. Here, the issue of specificity arises again.
Looking at the paragraph:
"As mentioned previously, the analysis may be undertaken via equations
...
take advantage of a one dimensional model of the same behavior."
In FEA, a mesh is created that represents the numerical model of the object
to be analyzed. Typically, comparison of numerical, analytical, and experimental
data is made by matching points on the object that have the largest and
smallest deformations and/or stresses with the expectation that the locations
and values of the deformations and/or stresses should be nearly the same.
There should also be a general agreement among all other points.
So any true automation by software should itself be capable of making
these comparisons of the object in its entirety. How such software
would go about achieving this should be explicitly described.
While everything described would be highly valued in software, there is
not enough innovation or exactness in the ideas to warrant a patent.