WeMB | Digital Twin Expert Group

본문 바로가기
사이트 내 전체검색
navi
Solution/Production >Analysis/Prediction
Analyzing, Learning, and Detecting with the Deep Learning Algorithm
The analysis/prediction solution enables proactive responses by analyzing and detecting symptoms in real time.

Data-driven Smart Predictive Analysis Platform, TOBIT SPA

It is a system that studies the data collected from various sources by analyzing algorithms according to the characteristics of data to create a valid model, detects abnormal symptoms in advance, and analyzes the cause.
By providing a flexible integration engine that can be easily applied to the user’s various data occurring in the environment, it minimizes interface latency.
The analysis algorithm that utilizes the finance/public/manufacturing fields’ model is also provided in library format with multiple predictive analyses.

Key Feature

  • 01Detecting pre-failure signs

    구성도
    구성도
    구성도
    구성도
  • 02Various libraries

    Pre-failure symptom detection based on dynamic thresholds

    Failure symptom detection based on statistical modeling

    Fast failure and impact analyses based on a topology that links business services and IT resources

    Fast failure and impact analyses based on a topology that links business services and IT resources

    Organic and flexible management of the failure status and statistical analysis function according to the system equipment, disability type, and disability level

    Failure and abnormal symptom detection

    Threshold range exceeded/below average

    Proactive detection of abnormal symptoms through the statistical technique (algorithm)

    Incident occurrence

    Detection of an anomaly in advance → EVENT occurrence

    Primary-cause analysis of failure and abnormal symptoms

    Analyze the cause by linking the IT-resource-related information from the perspective of the business service → TOPOLOGY-based event linkage and filtering

    Analysis of similar disorders and abnormal patterns in the past

    Analysis of similar failure patterns and analysis using CMDB

    Secondary-cause analysis of rfailure and abnormal symptoms

    Analyze which services are affected based on the configuration information associations between IT resources from the perspective of the business services.

    Establishment of priorities based on the business impact

    Solution establishment and processing

    A stage where specific measures and activities are executed for problem-solving

    Failure is resolved, and the action history is managed.

    Actualizing the failure index and statistic model by analyzing the failure outcome

    Supporting the virtuous-cycle structure with the efficient actualization of the statistics model in the operation process

    Secondary-cause analysis of failure and abnormal symptoms

    Analyze which services are affected based on the configuration information associations between IT resources from the perspective of the business services.

    Establishment of priorities based on the business impact

  • 03Data visualization

    Providing the screen to link from the enhancement of the efficiency in big-data meaning interpretation and failure detection to the failure cause identification and cause analysis

    * The greatest value of a picture is when it forces us to notice what we never expect to see. – John Tukey, creator of exploratory data analysis

    Various explorations

    Advancement analysis

  • 04Timeline event

    A function that searches all the events of the set time interval (including specific information) sequentially according to the set sorting information and the impact analysis results of the configured items by time and priority

Architecture

구성도
구성도
확대보기