US 11,797,871 B2
Predictive methodology to identify potential unknown sweet spots
Hossein Jacob Sadri, Novi, MI (US); Steven Torey, Macomb Township, MI (US); Stephen Juszczyk, Commerce Township, MI (US); Lance David Marsac, South Lyon, MI (US); and Sueha Elmir Salame, Dearborn, MI (US)
Assigned to Ford Global Technologies, LLC, Dearborn, MI (US)
Filed by Ford Global Technologies, LLC, Dearborn, MI (US)
Filed on Nov. 22, 2022, as Appl. No. 17/992,405.
Application 17/992,405 is a continuation of application No. 16/781,310, filed on Feb. 4, 2020, granted, now 11,537,923.
Prior Publication US 2023/0089283 A1, Mar. 23, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 7/01 (2023.01); G06F 16/9035 (2019.01); G06N 20/00 (2019.01)
CPC G06N 7/01 (2023.01) [G06F 16/9035 (2019.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
receiving, from a measurement system, measurement values for a subset of a plurality of criteria attributes for each of a subset of a plurality of variants of an object being tested to generate a plurality of measurement value sets, each measurement value set comprising the measurement values for each of the subset of criteria attributes for a variant of the plurality of variants of the object;
prioritizing the subset of criteria attributes;
setting cross-functional variants for at least a portion of the plurality of criteria attributes;
filtering the plurality of measurement value sets based on objective values for each criteria attribute of the prioritized subset of criteria attributes;
for each criteria attribute among the filtered prioritized subset of criteria attributes:
compare the criteria attribute against other criteria attributes in the filtered prioritized subset of criteria attributes using a sequential computing algorithm to identify parallel relationships within and between the filtered prioritized subset of criteria attributes; and
identify a sweet spot for each criteria attribute in the filtered prioritized subset of criteria attributes for each variant of the object based on the identified parallel relationships.