A Look at Mathematical and Computational Issues in Manufacturing Inspection Using Coordinate Measuring Machines
Craig M. Shakarji
Manufacturing Engineering Laboratory, Manufacturing Systems Integration Division
Tuesday, January 31, 2006 15:00-16:00,
Coordinate Measuring Machines are used widely for inspection in the manufacturing industry.
A critical component of these machines is mathematical software that must take a set of measured,
coordinate data points and extract information about the part itself.
At the core of this software are various least-squares and Chebyshev fitting algorithms.
Considerable work has been done at NIST to investigate the performance of these algorithms.
A special test service has been created to test several least-squares algorithms.
Research has also been done in the area of other fits, only to find serious issues exist there as well.
Mathematical reference algorithms have been created by NIST in seeking to address these issues.
This talk will give a brief background of these measuring instruments and why their embedded mathematical software is crucial to their performance;
no previous specialized knowledge is assumed.
The talk will then explain how various problems came to light in the industrial world and how NIST had a key role in the response to these problems.
Some mathematical challenges of creating reference fitting algorithms will be presented,
particularly in the area of problematic maximum-inscribed and minimum-circumscribed fitting.
At the heart of the method lies a simulated annealing optimization algorithm, which will be explained as well.
NIST North (820), Room 145
Tuesday, January 31, 2006 13:00-14:00,
Dr. Craig Shakarji is a Mathematician in the Manufacturing Engineering Laboratory of NIST.
He has been working for 10 years in the field of computational metrology, addressing problems in manufacturing inspection.
He actively participates in the international standards committee ISO 213 (on coordinate measuring machines)
as well as ASME B89 (on dimensional metrology) in the United States.
Craig received his PhD in Mathematics from UCLA in 1995.
Presentation Slides: PDF
Contact: P. M. Ketcham
Note: Visitors from outside NIST must contact
Robin Bickel; (301) 975-3668;
at least 24 hours in advance.