Skip to main content
Version: Next

Feature Overview

1. Capability map

Prodia presents a complete capability loop of:

ask → analyze → diagnose → act

Capability moduleCore capabilityBusiness effect
AI conversational analyticsNatural-language questions, follow-up clarification, result explanationLower the barrier so business users can ask data questions directly
Manufacturing BI analyticsOutput, quality, takt, fault, SPC, OEE analyticsProvide unified, professional, and reviewable manufacturing analysis
Agent diagnostic loopRoot cause localization, recommended actions, investigation paths, knowledge invocationLet the system move from reporting to diagnosing
Knowledge accumulationCase reuse, SOP accumulation, standardized workflowsTurn personal experience into organizational capability
Integration and deploymentConnect AI SCADA, MES, databases, and equipment dataImprove deployability while preserving governance boundaries

2. What users most often experience

Quickly see results

  • What was yesterday’s line output?
  • How did this week’s OEE change versus last week?
  • Which line or equipment contributed the most variation?

Quickly find causes

  • What caused the yield decline?
  • Which station is slowing down the entire line?
  • Which fault type has the greatest impact on downtime?

Quickly get guidance

  • What should be checked first?
  • Which parameter deserves attention?
  • What is the recommended next action for the current anomaly?

3. Capability levels

LevelFocusRepresentative capability
L1 DescriptiveWhat happenedQuery, statistics, trends, ranking, distribution
L2 DiagnosticWhy it happenedComparison, correlation, Pareto, root cause localization
L3 PredictiveWhat may happen nextRisk recognition, trend anticipation, degradation signals
L4 PrescriptiveWhat to do nextAction recommendations, investigation paths, optimization suggestions

4. How capabilities are used in practice

ScenarioUser actionSystem output
Daily reviewCheck yesterday’s key indicatorsKPI summary, anomaly points, focus items
Exception handlingAsk why the anomaly happenedCause candidates, affected scope, investigation suggestions
Process optimizationAnalyze takt, quality, and parameter relationshipsBottleneck identification, parameter window guidance
Maintenance supportAnalyze downtime and fault distributionFault ranking, MTBF / MTTR, response direction