Another type of model is correlation model, such as reasoning model, diagnostic
reasoning model or functional correlation model. The causal relationship between
model failure source and test or between test and test is needed to be expressed by
directed graph. Due to the inherent simplification characteristics of directed graph,
correlation model is the main modeling technology in testability analysis tools at the
moment.
The integrative technology of testability design and diagnosis for complex electronic information systems should focus on solving the problem of information modeling. Model is the quantitative description of system information and the foundation of
system analysis. In the aspect of testability modeling, there are three representative
models, which are testability block diagram description model, information flow
model, and multi-signal flow graph model. The testability block diagram description
model is a structured model, while the information flow model and the multi-signal
flow graph model are testability design model based on correlation. The multi-signal
flow graph model is also a directed graph model. The most representatives of directed
graph model are multi-signal flow graph model, Petri network model, and Bayes
network model. It can be said that in the field of testability modeling, correlation and
directed graph are the most representative theories. The correlation relationship can be
expressed as Fig. 1.
2 Digraph Model (Digraph = Directed Graph)
2.1 Definition of Digraph
The structure of directed “graph” is composed of the set of vertices and the set of
relations between vertices, which is:
G ¼ V; E
ð
Þ
G is the total set of the graphs; V is the finite nonempty set of the vertices; E is the
finite set of relations between vertices.
Multi-sign
al Flow
Model
Petri
Network
Model
Bayes
Network
Model
Correlation
Matrix
Fig. 1. The relationship between directed graph model and correlation model
392
L. Ou and F. Bai
reasoning model or functional correlation model. The causal relationship between
model failure source and test or between test and test is needed to be expressed by
directed graph. Due to the inherent simplification characteristics of directed graph,
correlation model is the main modeling technology in testability analysis tools at the
moment.
The integrative technology of testability design and diagnosis for complex electronic information systems should focus on solving the problem of information modeling. Model is the quantitative description of system information and the foundation of
system analysis. In the aspect of testability modeling, there are three representative
models, which are testability block diagram description model, information flow
model, and multi-signal flow graph model. The testability block diagram description
model is a structured model, while the information flow model and the multi-signal
flow graph model are testability design model based on correlation. The multi-signal
flow graph model is also a directed graph model. The most representatives of directed
graph model are multi-signal flow graph model, Petri network model, and Bayes
network model. It can be said that in the field of testability modeling, correlation and
directed graph are the most representative theories. The correlation relationship can be
expressed as Fig. 1.
2 Digraph Model (Digraph = Directed Graph)
2.1 Definition of Digraph
The structure of directed “graph” is composed of the set of vertices and the set of
relations between vertices, which is:
G ¼ V; E
ð
Þ
G is the total set of the graphs; V is the finite nonempty set of the vertices; E is the
finite set of relations between vertices.
Multi-sign
al Flow
Model
Petri
Network
Model
Bayes
Network
Model
Correlation
Matrix
Fig. 1. The relationship between directed graph model and correlation model
392
L. Ou and F. Bai