Normalized claim
Time: 99.5% decrease
Reduced processing time by 99.5% (from 4 hours to 1 minute per batch).
Grasim Industries doubles quality control efficiency with automated AI vision
Use case typeManufacturing quality monitoringUpdated Jun 13, 2026
Grasim Industries, a division of Aditya Birla Group in India, needed to improve its fiber quality control as manual procedures were slow, subjective, and labor-intensive. Defect detection depended on lab technicians performing lengthy, inconsistent manual analyses. Mandelbulb Technologies developed and delivered a deep learning vision model using Microsoft Fabric as the unified analytics platform. The solution integrated Fabric Real-Time Analytics, Data Activator, Power BI Embedded, OneLake, and Azure Kubernetes for highly scalable, automated QC. Processing time per batch was reduced from four hours to just one minute and defect detection accuracy reached 98%. Analysis throughput doubled, and Grasim now meets higher quality standards while cutting operational costs. The solution was prototyped and deployed in three weeks. Sustainability ambitions are supported by further leveraging Fabric and Azure for energy-efficient operations.
- Organization
- Grasim Industries
- Industry
- Manufacturing
- Location
- India
- Published
- August 2024
Reported outcomes
−99.5%
timeTime & speed
4 hourstime1 minutestime+98%accuracy+100%quantified impact
Strategic outcomes
New product / capabilityAutomated fiber quality analysisBetter decisions & insightReal-time defect detection alertsScale & capacityDoubled analysis throughputCost efficiencyLowered QC operational costs
Catalog median for time & speed deployments: −50% across 312 reported metrics. Compare benchmarks →
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Time: 4 hours decrease
Reduced processing time by 99.5% (from 4 hours to 1 minute per batch).
Normalized claim
Time: 1 minutes decrease
Reduced processing time by 99.5% (from 4 hours to 1 minute per batch).
Normalized claim
Accuracy: 98% increase
Improved defect detection accuracy to 98%.
Normalized claim
Quantified impact: 100% increase
Doubled analysis volume (100% throughput increase).
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Grasim Industries
- Provider
- Microsoft
- Maturity
- Production
- Linked source
- partner.microsoft.com
Analysis throughput doubled, and Grasim now meets higher quality standards while cutting operational costs
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 1 of 1
- 1Automated Vision-Based Quality Control in Fiber Manufacturing
- Manual QC processes were labor-intensive, slow, and inconsistent in fiber manufacturing.
- Traditional methods resulted in low defect detection efficiency and accuracy.
- Throughput was limited and costs high due to lengthy analyses and manual steps.
- Deep learning vision model automated fiber quality analysis using Microsoft Fabric platform.
- Integrated Real-Time Analytics for live data streaming and analysis.
- Data Activator enabled real-time alerts for defective fibers; Power BI Embedded delivered at-a-glance dashboards.
- Used OneLake for centralized scalable data storage and Azure Kubernetes for batch deployment and scalability.
- Reduced processing time by 99.5% (from 4 hours to 1 minute per batch).
- Improved defect detection accuracy to 98%.
- Doubled analysis volume (100% throughput increase).
- Enabled rapid deployment—solution delivered in 3 weeks.
- Lowered QC operational costs.
Architecture
The solution uses Microsoft Fabric as the backbone, combining Real-Time Analytics for streaming ingestion and querying, Data Activator for live alerts, Power BI Embedded for visualization, OneLake for data storage, and Azure Kubernetes for hosting scalable, containerized apps. Deep learning vision models analyze fiber images and trigger automated quality workflows.
Sources & evidence2
Evidence: High75/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
- Multiple corroborating sources available
Type: Case StudyPublished: Aug 7, 2024Publisher: partner.microsoft.comEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Was this useful?
Community
Comments
No published comments yet.