Project
1:
Title:
Stream-of-Variation Modeling for Multi-Station Manufacturing
Processes. Modeling Infrastructure for Virtual Assembly
Sponsor:
National Science Foundation (NSF) -CAREER.
Collaborators:
DCS and Ford
Description:
This project focuses on development of modeling; analysis and
control of dimensional variation in complex multistage assembly
processes (MAP) with compliant parts such as in automotive,
aerospace, appliance and electronics industries. The goal is to
develop a generic MAP model with capabilities to represent key
product and process control characteristics/features (KPCs/KCCs)
with varying resolution/"information granularity" that
can be utilized during design, launch and full production phases
of a new manufacturing system. The model will be based on a
generic Computer-Aided Design and Manufacturing (CAD/CAM) system
integrated with statistical analysis to predict manufacturing
process performance in early design phase. A challenge facing
the proposed plan is the diversity of required information and
lack of physical and geometrical relations between KCC and KPC.
The research will focus on developing: (1) variation propagation
models integrating statistical and CAD/CAM information; (2)
multi-resolution/granularity of KPC/KCC as they change during
product development; (3) computational efficiency for simulation
of multistage assembly processes; and (4) generic issues
pertaining to new process-oriented modeling, design and control.
Details in poster1,
poster2.
Project
2:
Title:
SOVA: Stream-of-Variation Analysis System for Multistage
Assembly Processes
Sponsor:
Advanced Technology Program, National Institute of Standards and
Technology (NIST-ATP).
Collaborators:
DCS, DaimlerChrysler, General Motors, Ford, Boeing, Northrop
Grumman
DCS PI: R. Kumar
UW-Madison Investigators: Drs. Ceglarek and Zhou, IE
Objective:
Develop a widely applicable computer simulation system for
modeling, analyzing, predicting, and optimizing the performance
of multistage manufacturing processes requiring accurate parts
alignment to improve production and product quality.
Description:
"Who could have predicted?" This is heard often
during the launch of a new manufacturing and assembly process
designed to deliver a wonderful new product to the consumer
market. The joining of rigid parts with compliant or flexible
parts often leads to unanticipated misalignments and other
dimensional variations that accumulate and are increasingly
compounded as the product moves down the assembly line.
Dimensional Control Systems (DCS) and the University of
Wisconsin-Madison will develop Stream-of-Variation Analysis (SOVA)
to eliminate most of the costly trial-and-error fine-tuning of
new-product assembly processes attributable to these unforeseen
dimensional errors. SOVA, a modeling, analysis, synthesis and
process control software system for variation management of
multistage manufacturing processes, is intended to be a widely
useful tool-set to be used throughout the assembly process from
design through production. Implemented during the product design
phase, the software will produce math-based predictions
of potential downstream assembly problems, based on evaluations
of the design and a large array of process variables. By
integrating product and process design in a pre-production
simulation, SOVA can head off individual assembly errors that
contribute to an accumulating set of dimensional variations,
which ultimately result in out-of-tolerance parts and products.
Once in the ramp-up stage of production, SOVA will be able to
compare predicted misalignments with actual measurements to
determine the degree of mismatch in the assemblies, diagnose the
root causes of the errors, isolate the sources from other
assembly steps, and then, on the basis of the SOVA model and
product measurements, recommend solutions. These analytical,
predictive and diagnostic capabilities are enabled by new
variation modeling research by DCS and the University of
Wisconsin-Madison. If transferred to the manufacturing sector,
such tools would deliver major benefits in terms of cost
savings, productivity and quality improvements, and shortened
product development cycles. For more information please see
[27] http://www.3dcs.com/sova.html.
Current progress:
(1)
Time-based Competition in Manufacturing: Stream-of-Variation
Analysis (SOVA) Methodology- Review
Summary: Frequency of model
change and the vast amounts of time and cost required to make a
changeover,also called time-based competition, has become a
characteristic feature of modern manufacturing and new product
development in automotive, aerospace, and other industries. This
paper discusses the concept of time-based competition in
manufacturing and design based on a review of on-going research
related to stream-of-variation (SOVA or SoV) methodology. The
SOVA methodology focuses on the development of modeling,
analysis, and control of dimensional variation in complex
multistage assembly processes (MAP) such as the automotive,
aerospace, appliance, and electronics industries. The presented
methodology can help in eliminating costly trial-and-error
fine-tuning of new-product assembly processes attributable to
unforeseen dimensional errors throughout the assembly process
from design through ramp-up and production. Implemented during
the product design phase, the method will produce math-based
predictions of potential downstream assembly problems, based on
evaluations of the design and a large array of process
variables. By integrating product and process design in a
pre-production simulation, SOVA can head off individual assembly
errors that contribute to an accumulating set of dimensional
variations, which ultimately result in out-of-tolerance parts
and products. Once in the ramp-up stage of production, SOVA will
be able to compare predicted misalignments with actual
measurements to determine the degree of mismatch in the
assemblies, diagnose the root causes of errors, isolate the
sources from other assembly steps, and then, on the basis of the
SOVA model and product measurements, recommend solutions.
(2) Tolerance Analysis for Design
of Multistage Manufacturing Processes using Number-Theoretical Net Method (NT-net)
Summary: Recent developments
in modeling stream of variation in multistage manufacturing
system along with the urgent need for yield enhancement in the
semiconductor industry has led to complex large scale simulation
problems in design and performance prediction, thus challenging
current Monte Carlo (MC) based simulation techniques. MC method
prevails in statistical simulation approaches for
multi-dimensional cases with general (i.e., non-Gaussian)
distributions and/or complex response functions. A method is
proposed based on number theory (NT-net) to reduce computing
effort and the variability of MC’s results in tolerance design
and circuit performance simulation. The sampling strategy is
improved by introducing NT-net that can provide better
convergent rate over MC. The new method is presented and
verified using several case studies, including analytical and
industrial cases of a filter design and analyses of a four-bar
mechanism. Results indicate a 90–95% reduction of computation
effort with significant improvement in accuracy that can be
achieved by the proposed technique.
(3) Statistical
Modal Analysis Methodology for Form Error Modeling with Implementation
to Assembly and Stamping Systems with Compliant Parts
Description: Current
geometric tolerancing (GT) techniques define tolerance boundary
solely based on deterministic geometry perspectives. For parts
with geometrically complex features such as sheet metal parts,
current tolerancing techniques are neither able to model random
form or surface errors nor allow for statistical tolerancing in
design. The Statistical Modal Analysis (SMA) methodology
developed in this paper attempts to
resolve the aforementioned challenges. The SMA methodology
includes: (i) mode representation of form error field; (ii) mode
significance test; (iii) mode truncation/selection criteria; and
(iv) sampling strategy. A discrete-cosine-transformation (DCT)
based decomposition method is proposed for modeling part form
error, which decomposes the form error into a series of
independent error modes. Compression, which ensures a compact
model, is achieved by correlation reduction and mode compression
based on four criteria: (i) statistical significance (SSC); (ii)
variance significance (VSC); (iii) energy significance (ESC);
and (iv) Hausdorff distance (HDC). The proposed SMA methodology
is expected to serve two purposes. First, in the design phase,
it allows to statistically represent a population of form or
surface variations, thus, providing the ability to simulate
compliant part variation in statistical tolerance analysis.
Second, in the manufacturing phase, it allows for modal model
representation of major error patterns that lead to an
easy-to-explain interpretation for dimensional fault diagnosis
during manufacturing.
