Figure 9 shows the valuation editor. Because we are trying to simulate the effect of the old model without conditioning on live steam, we want the probability of failure to be the same for both values of the conditional value. To achieve this effect, we can group the conditions. We set the values for the probabilities and save the valuation.
Figure 9. Valuation Editor for "Steam-Mov-Rule". Note that the conditions are grouped to simulate an unconditional probability.
To make "Steam-Mov-Rule" a class, we drag it into the template area to the left of the graph. This changes the instance into a prototype (class). At the same time, Graphical-Belief creates an new instance "Steam-Mov-Rule-1" of the new prototype. This new instance will inherit the information we just set, thus when we make changes to the prototype the instance will be affected too.
Figure 10. Making "Steam-Mov-Rule" a class. We create the prototype by dragging the "Steam-Mov-Rule" icon back into the template area. This promotes the instance to a prototype (class) and creates a new instance.
We are now ready to finish the connections. We connect the first rule to "MOV-25-B" and "Live-Steam". We create another instance of the "Steam-Mov-Rule" and connect it to "MOV-26-A" and "Live-Steam". Note that the icons for these rules turn yellow after connection. This is because they can now inherit their valuation from their prototype. Figure 11 shows the final shape of the model. We are now ready to compile and analyze the model.
Model Construction. Go back to the
begining of this example.
Return to
the main example page.
Back to overview of Graphical-Belief.
View a list
of Graphical-Belief in publications and downloadable technical
reports.
The Graphical-Belief user
interface is implemented in Garnet.
Get more
information about obtaining Graphical-Belief (and why
it is not generally available).
get
the home page for Russell Almond , author
of Graphical-Belief.
Click
here to get to the home page for Insightful (the company that StatSci
has eventually evolved into).