Manufacturing
processes are an excellent real-world example of the applicability of
compositional causal reasoning. A set of components form the process by
interacting with one another in well-defined ways to create a long chain from
cause to effect. The process, as a system of interacting components, is
amenable to computer-assisted study because of the degree to which the
behavior can be encapsulated in numbers and rules. Our system, because it so
centered around the assembly of composite entities, causal reasoning, and
equation-based models of reality, has a lot to offer manufacturers in
arranging and managing processes, and in understanding the tradeoffs they are
making in their decisions.
The model-based
applications of our technology to manufacturing processes include:
 | Process design, considering its ability to cope with various
occurrences of concern.
|
 | Model-based diagnosis of conditions of interest, including multiple
simultaneous faults.
|
 | Process configuration, and rapid reconfiguration in response to
monitoring and diagnosis.
|
 | Identification of situations that expose the process' vulnerabilities,
and of improvements that reduce their deleterious effects.
|