Diagnostic method and system

• US 20060150018A1
• Filed: 11/22/2005
• Published: 07/06/2006
• Est. Priority Date: 12/21/2004
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Neyt, Jean-Lois, Cousin, Benoit
• Granted Patent: US 7,409,317 B2
• Time in Patent Office: Days
• US Class Current: 714/26
• CPC Class Codes: G05B 23/0248 Causal models, e.g. fault tree; digraphs; qualitative physics
• CPC Class Codes: G05B 23/0278 Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA
• CPC Class Codes: G06F 11/2257 using expert systems

Abstract

A method and system comprising a diagnostic symptom tree system for diagnosing a failing system element causing a symptom in a system-under-test. A diagnostic symptom tree comprises symptom roots and dependent lower function nodes and sub-function nodes. Element nodes depend from the function or sub-function nodes, and a plurality of penultimate failure mode leaves depend from the element nodes. The function and sub-function nodes and the failure mode leaves comprise test information. Responsive to positive test results, the diagnostic symptom tree is configured to identify a function or sub-function or element node associated to a lowest symptom tree node or failure mode leaf for which a test is positive, or advise that no failing function or sub-function or element is found. In one aspect of the invention, nodes may include parameter values allowing successive selections of the nodes of the symptom tree for test iterations according to the parameter values.

Claims

1. A method to diagnose a failing system element causing a symptom in a system-under-test, said method comprising the steps of:

a. defining a diagnostic symptom tree comprising a plurality of symptom roots and a dependent plurality of lower nodes, the plurality of lower nodes comprising a plurality of function nodes depending from the symptom roots, a plurality of sub-function nodes depending from the function nodes, a plurality of element nodes depending from the function or sub-function nodes, and a plurality of penultimate failure mode leaves depending from the element nodes;

b. associating test information with each of the function and sub-function nodes and the failure mode leaves;

c. selecting one of the plurality of symptom roots responsive to identifying a symptom;

d. performing the test information associated with the selected symptom root, thereby generating a symptom root test result;

e. responsive to a positive symptom root test result;

e1. performing the associated test information of each of the symptom tree depending nodes and failure mode leaves responsive to each positive test result; and

e2. identifying a function or sub-function or element node associated to a lowest symptom tree node or failure mode leaf for which a test is positive; and

f. if no test result is positive, advising that no failing function or sub-function or element is found.

2. The method of claim 1, wherein the node and leaf test information comprises system-under-test type application information, further comprising the steps of:

a. identifying the system-under-test; and

b. defining the symptom tree by selecting a depending plurality of the nodes and leaves responsive to a correlation of the system-under-test identify to each node and leaf system-under-test type test information.

3. The method of claim 1, wherein the test information further comprises an intelligent component configured to collect system-under-test variables from system-under-test elements, further comprising the steps of:

a. test information in a first plurality of depending nodes comprising an automatic test defined as a logical expression of the system-under-test variables, a system-under-test context and an associated manual-confirmation test for confirmation of the automatic test;

b. test information in a second plurality of depending nodes comprising a manual-node test; and

c. setting a system-under-test context;

wherein the step 1d. of performing the test information further comprises the steps of;

d. performing the first node automatic tests responsive to the variables values and by computing the logical expressions;

e. if a logical expression is true on a node;

e1. executing the associated manual-confirmation test on that given node;

e2. if the associated manual-confirmation test is positive, then performing the other automatic and associated manual-confirmation tests of lower depending nodes; and

e3. reporting the element associated to the lowest node in the tree for which an associated manual-confirmation test is positive as the failing part;

f. if no automatic test or associated manual-confirmation test is positive;

f1. applying the associated manual-node test in each selected nodes and lower depending nodes responsive to positive manual-node test results; and

f2. reporting an element associated to the lowest node for which an associated manual-node test is positive as the failing part; and

g. if no manual-confirmation or manual-node test result is positive, reporting that no failing part is found.

4. The method of claim 3 wherein the symptom tree automatic tests are performed sequentially on different groups of associated tests, all the associated tests of each group having the same context.

5. The method of claim 1, further comprising the steps of:

a. storing a parameter value in each of the nodes; and

b. ordering the diagnostic symptom tree responsive to the parameter value in selected nodes.

6. The method of claim 5, further comprising the step of computing the parameter value of a first node responsive to the result of a previous diagnostic session performed on the first node.

7. The method of claim 5, wherein the step of ordering the diagnostic symptom tree further comprises the step of selecting nodes having a first parameter value, further comprising the steps of:

a. if no test result is positive in step 1f., ordering a different diagnostic symptom tree comprising nodes having a different parameter value;

b. if a different diagnostic symptom tree manual-node test result is positive;

b1. performing lower depending node manual-node tests; and

b2. reporting an element associated to the lowest node in the different diagnostic symptom tree for which an associated manual-node test is positive as the failing part; and

c. if no different diagnostic symptom tree manual-node test result is positive, then repeating at least one iteration of the steps 7a, 7b. and 7c., wherein the different parameter value of each iteration is divergent from a parameter value of any other iteration.

8. The method of claim 7 wherein the parameter value comprises a test cost of the node.

9. The method of claim 8 wherein the step 5b. of ordering a diagnostic symptom tree further comprises the step of selecting nodes having a low test cost value, and the step 7a. of ordering a different diagnostic symptom tree further comprises the step of selecting nodes having a higher test cost value relative to the diagnostic symptom tree node low test cost value.

10. The method of claim 5, further comprising the step of computing a frequency count of failure findings for each node at the end of a diagnostic session, wherein the step of computing the parameter value is performed responsive to the frequency count.

11. A diagnostic symptom tree system for diagnosing a failing system element causing a symptom in a system-under-test, said system comprising:

a. a diagnostic symptom tree comprising a plurality of symptom roots and a dependent plurality of lower nodes, the plurality of lower nodes comprising a plurality of function nodes depending from the symptom roots, a plurality of sub-function nodes depending from the function nodes, a plurality of element nodes depending from the function or sub-function nodes, and a plurality of penultimate failure mode leaves depending from the element nodes; and

b. the function and sub-function nodes and the failure mode leaves comprising test information;

c. wherein the diagnostic symptom tree system is configured to perform the test information and generate test results;

c1. wherein responsive to a positive symptom root test result, the diagnostic symptom tree system is configured to perform depending node and depending failure mode test information responsive to each previous positive test result and identify a function or sub-function or element node associated to a lowest symptom tree node or failure mode leaf for which a test is positive; and

c2. if no test result is positive, advise that no failing function or sub-function or element is found.

12. The invention of claim 11, wherein the node and the leaf test information further comprise system-under-test type application information, and wherein the diagnostic symptom tree is defined responsive to a correlation of each of the node and leaf system-under-test type test information.

13. The invention of claim 12, wherein the test information further comprises an intelligent component configured to collect system-under-test variables from system-under-test elements;

a. wherein test information in a first plurality of depending nodes comprises an automatic test defined as a logical expression of the system-under-test variables, a system-under-test context and an associated manual-confirmation test for confirmation of the automatic test;

b. wherein test information in a second plurality of depending nodes comprises a manual-node test;

c. wherein the diagnostic symptom tree has a system-under-test context; and

d. wherein the diagnostic symptom tree is configured to perform the test information by;

1. performing the first node automatic tests responsive to the variables values and by computing the logical expressions;

2. if a logical expression is true on a node, executing the associated manual-confirmation test on the node;

3. if the associated manual-confirmation test is positive, then performing the automatic and associated manual-confirmation tests of lower depending nodes, and reporting an element associated to the lowest node for which an associated manual-confirmation test is positive as the failing part;

4. if no automatic test or associated manual-confirmation test is positive, then performing the associated manual-node test, and performing the associated manual-node tests of lower depending nodes and leaves responsive to a positive manual-node test result, and reporting an element associated to a lowest node for which an associated manual-node test is positive as the failing part; and

5. if no manual-confirmation or manual-node test result is positive, reporting that no failing part is found.

14. The invention of claim 13 wherein the diagnostic symptom tree is configured responsive to a system context.

15. The invention of claim 14, wherein the diagnostic symptom tree is further configured to store a parameter value in each of the nodes, and the diagnostic symptom tree is ordered responsive to the parameter value.

16. The invention of claim 15, wherein the diagnostic symptom tree is further configured to compute the parameter. value of a first node responsive to the result of a previous diagnostic session performed on that first node.

17. The invention of claim 15, wherein the diagnostic symptom tree is defined by selecting nodes having a first parameter value, and wherein if no 13d5. test result is positive, the diagnostic symptom tree is configured to:

a. order a different diagnostic symptom tree by selecting nodes and leaves having a different node and leaf parameter value;

b. if a different diagnostic symptom tree manual-node test result is positive, perform the manual-node tests of the different diagnostic symptom tree lower nodes and leaves and report an element associated to the lowest node for which an associated manual-node test is positive as the failing part;

c. else if no different diagnostic symptom tree manual-node test result is positive, then repeat at least one iteration of the steps 17a., 17b. and 17c., wherein the different parameter value of each iteration is divergent from a parameter value of any other iteration.

18. An article of manufacture comprising a computer usable medium having a computer readable program embodied in said medium, wherein the computer readable program, when executed on a computer, causes the computer to:

order a diagnostic symptom tree comprising a plurality of symptom roots and a dependent plurality of lower nodes, the plurality of lower nodes comprising a plurality of function nodes depending from the symptom roots, a plurality of sub-function nodes depending from the function nodes, a plurality of element nodes depending from the function or sub-function nodes, and a plurality of penultimate failure mode leaves depending from the element nodes, the function and sub-function nodes and the failure mode leaves comprising test information;

perform a symptom root test information and generate a symptom root test result;

responsive to a positive symptom root test result, perform depending node and failure mode test information responsive to each positive test result, and identify a function or sub-function or element node associated to a lowest symptom tree node or failure mode leaf for which a test is positive; and

if no test result is positive, advise that no failing function or sub-function or element is found.

19. The article of manufacture of claim 18, wherein the node and leaf test information comprise system-under-test type application information, wherein the computer readable program, when executed on a computer, causes the computer to:

identify the system-under-test; and

define the symptom tree by selecting a depending plurality of the nodes and leaves responsive to a correlation of the system-under-test identity to each node and leaf system-under-test type test information.

20. The article of manufacture of claim 18, wherein the computer readable program, when executed on a computer, causes the computer to:

a. set a system-under-test context; and

b. order the diagnostic symptom tree responsive to the system-under-test context;

wherein the test information further comprises an intelligent component configured to collect system-under-test variables from system-under-test elements;

the test information in a first plurality of depending nodes comprises an automatic test defined as a logical expression of the system-under-test variables, a system-under-test context and an associated manual-confirmation test for confirmation of the automatic test;

wherein test information in a second plurality of depending nodes comprises a manual-node test; and

wherein the computer readable program, when executed on the computer, causes the computer to;

c. perform the test information by the steps of;

1. performing the first node automatic tests responsive to the variables values and by computing the logical expressions;

2. if a logical expression is true on a node, reporting a positive test result and advising an operator to execute the associated manual-confirmation test on that given node;

3. performing the other automatic and associated manual-confirmation tests of lower depending nodes responsive to indication of positive associated manual-confirmation test results;

4. reporting an element associated to the lowest node for which an associated manual-confirmation test is positive as the failing part;

5. if no automatic test or associated manual-confirmation test is positive, reporting an element associated to the lowest node in the tree for which an associated manual-node test is positive as the failing part; and

6. if no manual-confirmation or manual-node test result is positive, reporting that no failing part is found.