Cause Point Analysis for Static Alarm Prioritization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Manual inspection of static analysis alarms is costly, time-consuming, and tedious, as users must classify each alarm as true or false, with repetitive reviews making the process less interesting and inefficient.

Innovation Solution

A system and method for cause point analysis that identifies and classifies the cause points of alarms generated during static analysis, using metrics like cause point type, contribution score, and spatial proximity to prioritize and rank alarms, allowing for more efficient manual inspection and error identification, with customizable ranking criteria and interactive querying to resolve alarms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual inspection of each alarm is performed to classify true errors from false warnings, then software quality assurance is improved, but time consumption and cost increase significantly

Engineering Contradiction:
Improvesoftware quality assuranceVSAvoidtime consumption for alarm inspection
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The alarm inspection process is segmented into multiple phases: automated static analysis generates initial alarms, cause point analysis identifies critical program points, and manual inspection focuses only on high-priority cause points rather than all alarms equally. This segmentation reduces the volume of manual work while maintaining quality assurance.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Cause point analysis acts as an intermediary between automated static analysis and manual alarm inspection. It processes alarm information and generates prioritized cause points, serving as a mediator that filters and prepares data for manual review, thereby reducing the time burden on inspectors.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If comprehensive manual review of all alarms is conducted to ensure no errors are missed, then detection precision is improved, but productivity decreases due to repetitive work

Engineering Contradiction:
Improvedetection precision of errorsVSAvoidinspection throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

Different levels of inspection effort are applied to different alarms based on their characteristics. High-priority cause points receive detailed manual review, while lower-priority alarms are handled through automated analysis or simplified review processes. This local quality approach maintains detection precision for critical errors while improving overall productivity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The inspection process dynamically adjusts parameters such as review depth, time allocation, and inspection methodology based on alarm priority and cause point characteristics. This allows the system to maintain high detection precision for critical issues while optimizing productivity across the entire alarm set.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If detailed analysis of each alarm is performed to accurately identify true errors, then measurement precision is improved, but the complexity of the inspection process increases

Engineering Contradiction:
Improveaccuracy of error identificationVSAvoidinspection process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Cause point analysis is performed as a preliminary action before manual alarm inspection. It pre-processes alarm data, identifies critical program points, and generates prioritized lists of cause points that require detailed analysis. This preliminary action simplifies the subsequent manual inspection process while maintaining accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The inspection process maintains continuity by seamlessly integrating automated cause point analysis with manual review. The output of automated analysis directly feeds into manual inspection workflows, creating a continuous process that maintains measurement precision without requiring complex transitions or rework between automated and manual phases.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP3285171B1System and method for cause point analysis for effective handling of static analysis alarms
Publication Date: 2021.11.24 TATA CONSULTANCY SERVICES LTD
  • EP3285171B1 patent drawingFigure 1
  • EP3285171B1 patent drawingFigure 2
  • EP3285171B1 patent drawingFigure 3

AI summary

A method and system is disclosed herein to identify causes of alarms generated corresponding to each alarm of the static analysis. The system models the basic reasons for alarms as alarm cause points of several types. Further, the system ranks these cause points based on different metrics such as unknown type-based ranking, grouping of cause points based on their lexical similarity or proximity and last but not least contribution score based ranking. The system also uses a workflow in which a user answers queries about the cause points and their answers are used in subsequent round of the static analysis. The collaboration between the answers of the user and the static analysis helps the tool to resolve the unknowns encountered during the analysis and weeding out the alarms. It also helps the user to expedite the manual inspections of alarms.