eXpress

  Help
×
Menu
Index
Search
 
   
   
   
 
 

Multiple Fault: Maximize Functions Proven by Refinement

What is the purpose of this algorithm?
The fault isolation algorithm Multiple Fault: Maximize Functions Proven by Refinement, like the algorithm Multiple Fault: Half-Split Failure Probs. (refine where appropriate), uses Multiple-Failure isolation and does not constrain Test Set Test selection using Test Candidate Groupings. All Test Set tests are equally candidates-the first test to be performed will be the test that best satisfies the algorithm's weighting criteria. The weightings for this algorithm attempt to reduce the number of refinement tests by first using tests that prove the maximum failure-weighted percentage of functions and/or failure modes when they pass (a secondary set of criteria will privilege the tests that come closest to half-splitting when they fail).
How was this algorithm implemented?
The full set of test selection criteria for this algorithm are as follows:
A. Test Candidate Groupings (2)
     1. Test Candidate Grouping 1 of 2: Test Set Isolation
          a) Candidate Test Types (1)
               1) All Test Set Tests
          b) Weightings: uses algorithm defaults
          c) Cutoffs: uses algorithm defaults
     2. Test Candidate Grouping 2 of 2: Internal Isolation
          a) Candidate Test Types (2)
               1) All Net Functions
               2) All Output Flags 
          b) Weightings: uses algorithm defaults
          c) Cutoffs: uses algorithm defaults
B. Default Test Weightings (6)
     1. Test Weighting 1 of 6: Count Number of Items
          a) Priority:         50
          b) Entity:           Number of Items
          c) Type:            Count
          d) Domain:        Suspect Functions Proven
          e) Best Equals: Highest
     2. Test Weighting 2 of 6: Sum Failure Probability
          a) Priority:         50
          b) Entity:           Failure Probability
          c) Type:            Sum
          d) Domain:        Suspect Functions Proven
          e) Best Equals: Highest
     3. Test Weighting 3 of 6: Sum Failure Probability
          a) Priority:         40
          b) Entity:           Failure Probability
          c) Type:            Sum
          d) Domain:        Suspect Failure Modes Proven
          e) Best Equals: Highest
     4. Test Weighting 4 of 6: Count Number of Items
          a) Priority:         20
          b) Entity:           Number of Items
          c) Type:            Count
          d) Domain:        Suspect Functions Detected
          e) Best Equals: Half-Split
     5. Test Weighting 5 of 6: Sum Failure Probability
          a) Priority:         20
          b) Entity:           Failure Probability
          c) Type:            Sum
          d) Domain:        Suspect Functions Detected
          e) Best Equals: Half-Split
     6. Test Weighting 6 of 6: Sum Failure Probability
          a) Priority:         15
          b) Entity:           Failure Probability
          c) Type:            Sum
          d) Domain:        Suspect Failure Modes Detected
          e) Best Equals: Half-Split
C. Default Test Cutoffs (2)
     1. Cutoff 1 of 2: Count Number of Tests
          a) Entity:          Number of Tests
          b) Type:            Count
          c) Domain:        Isolation Path
          d) Modifier:        Test Usage
          e) Target:          Refinement
          f) Condition:       >= 3
          g) Action:          Ignore in Sequence
     2. Cutoff 2 of 2: Count Number of Items
          a) Entity:          Number of Items
          b) Type:            Count
          c) Domain:        Suspected Items
          d) Modifier:        <none>
          e) Target:          <none>
          f) Condition:       <= 1
          g) Action:          Terminate Sequence