Singapore’s SMRT Corporation is advancing its digital rail strategy with a new AI-led maintenance pilot built on cloud technologies to enhance reliability and operational efficiency.
The initiative centres on JARVIS, an in-house intelligent analytics platform that brings together data from multiple standalone systems into a unified environment. It also enables engineers to move from reactive fixes to predictive maintenance.
SMRT’s rail network supports more than two million passenger journeys daily, placing increasing pressure on maintenance teams to detect faults early and minimise service disruptions. The AI platform is designed to address this by applying machine learning and a generative AI interface to continuously analyse operational and maintenance data.
SMRT’s group chief executive officer Ngien Hoon Ping said, “SMRT is committed to providing a safe, efficient, and high-performing railway network. We will leverage technology including AI to improve our safety, operations and reliability.”
“With JARVIS, we are making more intelligent use of our data and engineering expertise. We see strong potential to advance predictive maintenance, deepen engineering insights, and strengthen innovation across our teams,” he added.
Unified data, predictive insights
SMRT has deployed Oracle Cloud Infrastructure (OCI) Enterprise AI alongside the Oracle Autonomous AI Database to power JARVIS.
The autonomous AI Database aggregates and analyses a wide range of operational data from train performance metrics and sensor readings to asset lifecycle information into a single source of truth. This unified approach enables earlier fault detection and supports proactive interventions, helping reduce downtime and improve service consistency.
The platform also introduces a natural-language interface, allowing engineers to query systems more intuitively and access insights faster, particularly useful in time-sensitive operational scenarios.
“Rail operators depend on timely, accurate data to keep services running safely, reliably, and on schedule for millions of commuters each day,” said Chin Ying Loong, senior vice president and regional managing director, ASEAN & SAGE, Oracle.
The pilot marks an early step in SMRT’s wider effort to build a more resilient and data-driven rail network. If successful, the firm believes this approach could be extended across more assets and operational scenarios.
