What challenges did Novus Hi-Tech face before AI implementation?

Responding to unsafe driving alerts depended on command-center follow-up, making it difficult to scale timely and consistent safety interventions across fleets. The company needed a more intelligent approach that could interpret risk in context, trigger the next best action, and verify that interventions worked.

What was the impact of the AI solution?

In one of Novus's largest dangerous-goods logistics programs, the solution contributed to a 55% reduction in drowsiness alerts per 1,000 km and a 46% annual reduction in road-behavior violations. The company moved from concept to production-ready implementation in under 4 weeks versus an internal estimate of 4 to 6 months.

Source

Novus Hi-Tech enhances Fleet Management System using Amazon Bedrock

Use case typeAutomotive operations automationUpdated Jun 13, 2026

Novus Hi-Tech enhanced FleetGPT to improve fleet safety operations as its deployments expanded. The system assesses unsafe events in context, determines next-best actions, automates routine follow-up, and escalates higher-risk cases while human teams remain the validation layer. Amazon Bedrock powers the reasoning layer across video, telemetry, and driver-behavior data; AWS IoT Core, Amazon Kinesis Video Streams, Amazon S3, and AWS Step Functions support ingestion, storage, and workflow orchestration.

Organization: Novus Hi-Tech
Industry: Automotive
Location: India
Published: June 2026

Reported outcomes

55 percent reduction

drowsiness alerts per 1,000 kmOther quantified impact

46 percent annual reductionroad-behavior violations4 weeksbuild cycle

Strategic outcomes

Risk & complianceAutomated safety interventions with human validationNew product / capabilityAdded context-aware safety decisioningScale & capacityScaled timely fleet safety interventionsSpeed & agilityMoved from concept to production quickly

Primary read

Use case focus

Showing 3 of 5

1Agentic AI
2Fleet safety
3Predictive operations

Responding to unsafe driving alerts depended on command-center follow-up, making it difficult to scale timely and consistent safety interventions across fleets.
The company needed a more intelligent approach that could interpret risk in context, trigger the next best action, and verify that interventions worked.

Novus Hi-Tech enhanced FleetGPT as an agentic AI safety system within Novus OmniFleet.
It uses Amazon Bedrock with Claude Sonnet and Claude Haiku for reasoning, Amazon Titan embeddings for context retention, AWS IoT Core for edge device connectivity, Amazon Kinesis Video Streams for video ingestion, Amazon S3 and Amazon RDS for fleet data storage, and AWS Step Functions for multi-step workflows.
The system automates routine safety interventions while keeping people involved for validation and higher-risk cases.

Technologies

Amazon Bedrock Amazon Bedrock AgentCore Amazon Titan embeddings AWS IoT Core Amazon Kinesis Video Streams Amazon S3 AWS Step Functions

In one of Novus's largest dangerous-goods logistics programs, the solution contributed to a 55% reduction in drowsiness alerts per 1,000 km and a 46% annual reduction in road-behavior violations.
The company moved from concept to production-ready implementation in under 4 weeks versus an internal estimate of 4 to 6 months.

Architecture

FleetGPT is an agentic AI safety system that uses Amazon Bedrock for contextual reasoning over video, telemetry, and driver behavior, Amazon Titan embeddings for retaining context, AWS IoT Core for device connectivity, Amazon Kinesis Video Streams for video ingestion, Amazon S3 and Amazon RDS for storage, and AWS Step Functions for orchestration.

Sources & evidence1

Groundedness: 5/5Type: Customer StoryPublished: Jun 8, 2026Publisher: AWSEvidence: VendorConfidence: Medium

Primary source

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Customer evidence