What challenges did SMRT face before AI implementation?

Unify distributed maintenance and condition-monitoring data across multiple standalone systems. Improve predictive maintenance, fault resolution, reliability, and operational responsiveness in a safety-critical rail network.

What was the impact of the AI solution?

Enables predictive maintenance and proactive intervention. Speeds up fault resolution. Improves rail reliability, commuter safety, and operational responsiveness.

Source

ProductionEvidence: Medium55/100

SMRT JARVIS rail maintenance pilot with OCI Enterprise AI and Oracle Autonomous AI Database

Use case typePredictive maintenanceUpdated Apr 14, 2026

SMRT, Singapore’s leading public transportation provider, is running a pilot project using Oracle Cloud Infrastructure (OCI) Enterprise AI and Oracle Autonomous AI Database to improve rail maintenance. STRIDES Technologies developed JARVIS, an intelligent analytics platform that unifies maintenance, train performance, sensor, and asset lifecycle data from multiple standalone systems into a single trusted source of truth. The platform uses machine learning through a generative AI chatbot interface and vector search to support predictive maintenance, faster fault resolution, and more informed operational decisions.

Organization: SMRT
Industry: Public Sector
Location: Singapore
Published: April 2026

Reported outcomes

Strategic outcomes

Cost efficiencyFaster fault resolution and maintenance responseRisk & complianceImproved safety and reliability in rail operations

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: SMRT
Provider: Oracle
Maturity: Production
Linked source: STRIDES Technologies

The platform uses machine learning through a generative AI chatbot interface and vector search to support predictive maintenance, faster fault resolution, and more informed operational decisions

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Predictive maintenance
2Operational analytics
3Conversational assistants

Unify distributed maintenance and condition-monitoring data across multiple standalone systems.
Improve predictive maintenance, fault resolution, reliability, and operational responsiveness in a safety-critical rail network.

Built JARVIS as an intelligent analytics platform on OCI Enterprise AI and Oracle Autonomous AI Database.
Consolidated maintenance, train performance, sensor, and asset lifecycle data into one platform.
Used vector search and a conversational interface to provide predictive fault detection and troubleshooting support.

Technologies

OCI Enterprise AI Oracle Autonomous AI Database Vector Search

Enables predictive maintenance and proactive intervention.
Speeds up fault resolution.
Improves rail reliability, commuter safety, and operational responsiveness.

Architecture

JARVIS is built on Oracle Autonomous AI Database as the core data platform, with OCI Enterprise AI and vector search capabilities supporting large language models and a generative AI chatbot interface. It consolidates maintenance and condition-monitoring data from train performance, sensor readings, and asset lifecycle systems across SMRT's rail network.

Sources & evidence1

Evidence: Medium55/100Evidence strength