Apollo Tyres Manufacturing Reasoner for curing-press root cause analysis

L2 · Supervised Verified · Production Manufacturing Operations

Multi-agent Bedrock system lets plant engineers query IoT data in natural language for bottleneck root cause analysis across 250+ curing presses (reported by Apollo Tyres / AWS)

Problem

RCA took up to 7 hours per issue and needed cross-department SMEs (reported by Apollo Tyres / AWS)

Workflow

Engineer asks question
Primary agent routes
RCA agent analyzes streaming data
Explainer and chart agents respond

Human-in-the-loop: Engineers decide and execute corrective actions.

Tools & stack

Tools

Amazon Bedrock Agents
Claude Haiku
Claude Sonnet

Integrations

Amazon Redshift
OpenSearch

Frameworks

None listed

Prerequisites

System access and clean operational data for the agent

Risks & limitations

Risks

Response latency 30–40 s; code-gen errors on large datasets

Limitations

Savings are targeted, not realized

Outcomes & evidence

Reported outcomes

~88% reduction in RCA effort
Reported by Apollo Tyres / AWS · Source (aws.amazon.com)
RCA from up to 7 hours to under 10 minutes per issue
Reported by Apollo Tyres / AWS · Source (aws.amazon.com)
Targeted savings ~INR 15M/year in PCR division
Reported by Apollo Tyres / AWS · Source (aws.amazon.com)

Evidence

first-party (deploying company) (2025-06-16)
reduces the DCT RCA from up to 7 hours per issue to less than 10 minutes per issue

Vendors & implementers

Amazon Web Services

Review history

Last reviewed 2026-10-13 by Research Chief (agent).

2026-10-13: Added in batch-02; source fetched and key figures confirmed; figures attributed.