Suite of metrics for software quality assurance and product development

• US 7,003,766 B1
• Filed: 06/19/2001
• Issued: 02/21/2006
• Est. Priority Date: 06/19/2001
• Status: Expired due to Fees

• Current Assignee: AT&T Corporation (AT&T, Inc.)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Hong, Yi
• Primary Examiner(s): Khatri, Anil
• Application Number:
  US09/884,723
• Time in Patent Office: 1,708 Days
• Field of Search: 717/108, 717/116, 717144-147, 717101-103, 717162-164
• US Class Current: 717/144
• CPC Class Codes: G06F 9/4488 Object-oriented

Abstract

Based on software developed with use-cases, a suite of metrics for the Product/Process (PP) complexity of software systems as well the PP complexity of each stage in the software development process—analysis, design, and testing—is provided. The metrics suite is important for software quality assurance—Validation and Verification (V&V) as well as for understanding new complexities brought into software development by new technologies. The metrics can be collected based on the two most important documents of use-case software development—scenario diagrams and state diagrams. All metrics are CASE tool computable. Thus, each iterative or incremental development process can be quantitatively measured and compared without incurring extra cost or load on the project team. Efficiency metrics for implementation and test stages, normalization rules and logic-metrics for business tables are also presented. A three-level software visualization approach may have impact on graphical programming or direct transformation.

Claims

1. A computer-implemented method for measuring a complexity of nested object state transition diagrams that are represented as data structures in the computer, the method comprising:

a) using the computer, determining a plurality of graphs of object state transitions at K levels l_k, wherein;

(A) 0<

k<

K;

(B) one or more of said graphs at level l_k+1 are expansions of one or more of said graphs at level l_k; and

(C) said graphs comprise a plurality of nodes to represent a corresponding plurality of states of use-cases and a plurality of edges to represent a corresponding plurality of transitions between the states; and

b) using the computer, determining a complexity for said plurality of graphs.

2. The computer-implemented method of claim 1, wherein:

the state of an object at level k is a super state of an object state at level k+1, and

the state of an object at level k is a sub state of an object state at level k−

1, and

the method further includes identifying the state transion relationship between super state and sub state as one of a group consisting of “

in,”

“

out,” and

“

inout”

.

3. The computer-implemented method of claim 2, further comprising:

selecting said level l_k=l_K;

expanding a graph L from said selected level l_k with at least one graph from said level l_K+1;

determining paths in said expanded graph; and

determining a number of conditions in said transition paths.

4. The computer-implemented method of claim 3 further comprising:

removing an unnecessary path from said at least one graph.

5. The computer-implemented method of claim 4 wherein said one or more unnecessary paths comprise one or more numbers of a group consisting of:

(exception→

fallout),

(exception→

exception),

(one-repetition loop→

fallout),

(one repetition loop→

cancelled), and

(one repetition loop→

exception).

6. A computer-implemented method for measuring a complexity of nested object state transition diagrams that are represented as data structures in the computer, the complexity measuring method comprising:

a) using the computer, determining a plurality of graphs of object state transitions at K levels l_k, wherein;

(A) 0<

k<

K;

(B) one or more of said graphs at level l_k+1 are expansions of one or more of said graphs at level l_k; and

(C) said graphs comprise a plurality of nodes to represent a corresponding plurality of states of use-cases and a plurality of edges to represent a corresponding plurality of transitions between the states; and

b) using the computer, determining a complexity for said plurality of graphs, the complexity determining step including measuring a nested object state transition complexity between two super-states by recursively applying Equation 6;

$\begin{array}{cc}{\mathrm{STPC}}_{k,p,q}=\underset{i=1}{\overset{m_k}{\sum <br><br>}}<br><br>\left(\underset{j=1}{\overset{n_{k,i}}{\prod <br><br>}}<br><br><br><br>\left(C_{k,i,j}+\left({\mathrm{STPC}}_{k+1,i,j}-{\mathrm{sub}}_{k,i,j}\right)\right)+{\mathrm{mul}}_{k,i}<br><br>\left(N\right)\right)& \mathrm{Equation}<br><br><br><br>6\end{array}$

wherein;

m_k=a number of transition paths of level-k object (k>

1) between two super-states or a number of transition paths of level-1 object;

n_k,i=a number of states along path i for level-k object;

C_k,i,j is a number of conditions to bring level-k object state from Sk,i,j−

1 to Sk,i,j;

STPC_k+1,i,j=Substate Transition Complexity between state Sk,i,j−

1 and Sk,i,j;

sub_k,i,j=0 if STPC_k+1,i,j=0 and sub_k,i,j=1 if (STPC_k+1,i,j≠

0);

if a multiplicity between level-k object and level-_(k+1) is 1;

N, then mul_k,i(N)=0 if (N=1) and mul_k,i(N)=r if (N>

1); and

r is a number of times that level-_k+1 objects transit back to level-_k in path i.

7. The computer-implemented method of claim 6, wherein:

the state of an object at level k is a super state of an object state at level k+1, and

the state of an object at level k is a sub state of an object state at level k−

1, and

the method further includes identifying the state transion relationship between super state and sub state as one of a group consisting of “

in,”

“

out,” and

“

inout”

.

8. The computer-implemented method of claim 7, further comprising:

selecting said level l_k=l_K;

expanding a graph L from said selected level l_k with at least one graph from said level l_K+1;

determining paths in said expanded graph; and

determining a number of conditions in said transition paths.

9. The computer-implemented method of claim 8 further comprising:

removing an unnecessary path from said at least one graph.

10. The computer-implemented method of claim 9 wherein said one or more unnecessary paths comprise one or more numbers of a group consisting of:

(exception→

fallout),

(exception→

exception),

(one-repetition loop→

fallout),

(one repetition loop→

cancelled), and

(one repetition loop→

exception).