|
Our research focuses on the integration of
manufacturing system CAD/CAM models with statistics-based methods
for design, control and diagnostics of multistage processes.
To this end our research addresses the following:
Increasing
complexity of current products with simultaneous higher
performance requirements place exceedingly high demands on modern
MMPs (convertibility, scalability, diagnosability) creating a
critical need for reusable/reconfigurable manufacturing systems.
Moreover, the developed techniques require generic MMP models with
capabilities to represent key product and process control
characteristics/features (KPC/KCC) with varying
resolution/“information granularity” such that they can be
utilized during design, launch and full production phases.
Current math-based manufacturing and design methods lack generic
MMP models with capabilities to represent
multi-resolution/granularity of KPCs/KCCs as they change during
new product and process development. In number of complex
manufacturing systems, about 70% of all engineering changes are
related to product dimensional variation due to the lack of
technology to accurately predict process performance/variation
during design phase.
Our research led to the development of a generic model for
dimensional variation of MMP, which integrates KPC/KCC using
state-space modeling framework [1,
4, C9,
S1, 17]. Additionally, we created a
“mode-based” approach to analyze varying “information
granularity” of the KPC/KCC as they change during new
product/process development (design, ramp-up and production
phases) [20].
The aforementioned methodology integrates KPC/KCC using
state-space modeling framework for rigid [17,
C9] and compliant part assemblies. These include
using 3-D elastic-beam [12,
13, 21];
FEM models [25,
28]; and
analytical models based on discrete cosine transformation, which
allows varying number of modes/granularity of the KPC/KCC as they
change during new product/process development [20].
This further led to: (a) Process-oriented tolerancing of MMP
systems, which integrates product variables with explicitly
represented process parameters ([21, C14])
and further synthesizes tolerance limits with tooling degradation
for maintenance design ([S4]).
The resulting methodology expands the concept of “part
interchangeability” into “process interchangeability,”
critical in increasing requirements related to suppliers
selection, benchmarking or outsourcing. (b) Design evaluation of
MMPs in early design phases ([6,
8, 19]).
(c) Diagnosability analysis of MMP systems [13,
16]. (d) Convertibility analysis of
reconfigurable/reusable MMPs using the developed rapid fixture
deployment approach [S2],
which is based on fixture workspace synthesis [24]
and tooling visibility analysis [S6]
|
