SMRT taps Oracle for AI-enabled rail maintenance

Singapore’s leading public transportation provider uses Oracle Cloud Infrastructure Enterprise AI and Oracle Autonomous AI Database for enhanced rail maintenance

SMRT has selected Oracle to help with rail maintenance through its intelligence data AI platform, JARVIS. SMRT’s technology innovation arm, STRIDES Technologies developed JARVIS, through the Oracle AI Customer Excellence Center (CEC) that was launched last year.

SMRT is Singapore’s leading public transportation provider and supports more than two million passenger journeys daily. With over 200km of rail network managed by SMRT, ensuring safety as well as delivering reliable and comfortable train services is imperative. Achieving this would mean being able to have visibility on the rail system, especially when it comes to maintenance.

JARVIS applies AI to areas that require timely, accurate, and continuous data analysis, leveraging machine learning through a generative AI chatbot interface. This enables maintenance teams to tap into predictive maintenance capabilities and speed up fault resolution, enhancing rail reliability and commuter safety.

“SMRT is one of the companies that has benefited from the CEC. Through the CEC, SMRT was able to develop, test and refine JARVIS, an intelligent platform that harnesses data analytics and AI-powered chatbots to predict faults, optimize maintenance, and enhance the reliability of rail operations. As a result, SMRT can today complete work orders much faster and at a higher scale. It has also set the critical foundation for more predictive, data-driven maintenance,” Low Yen Ling, Singapore Senior Minister of State for Trade and Industry and Culture, Community & Youth said in her keynote address at Oracle AI World.

“Such partnerships in AI development and solutions are vital as Singapore moves to widespread AI adoption,” she added.

According to Ngien Hoon Ping, Group CEO of SMRT, as the rail provider has 38 years of data in various structures and formats, having engineers go through them to address any concerns would take a very long time. With JARVIS, the process is done much faster as the intelligent AI agents are able to understand the system better, enabling predictive maintenance.

“We do our best. It’s all about being good today and better tomorrow. It’s a really complex system and a challenging task. But we improve and use whatever tools available. Working with Oracle allows us to meet these needs,” Ngien said during a customer session at Oracle AI World in Singapore.

The first phase of JARVIS was all about making SMRT data available to the database with more than 50 engineers in the process of analyzing and coding for agents to develop the smart bot. The results have seen some very specific engineering questions being able to be answered quickly.

“We do want our engineers to get familiar and comfortable with the system. It’s very promising at this early stage. This aligns with our Kaizen philosophy of continuous improvement while remaining human-centric and uplifting our people’s skills and capabilities. We see strong potential to advance predictive maintenance, deepen engineering insights, and foster a culture of innovation within our engineering teams,” he added.

Ngien also shared that SMRT is open to sharing the innovation in JARVIS to other rail service providers globally as well through the Oracle CEC. As JARVIS enhances maintenance and operational performance by unifying maintenance and operations data into a single, trusted source of truth, it enables predictive fault detection and for proactive intervention. The platform also features a natural-language interface that delivers faster, more accurate engineering insights and troubleshooting.

Chin Ying Loong, Senior VP and Regional Managing Director, ASEAN & SAGE, Oracle shared that JARVIS running on OCI, demonstrates how Oracle can help bring AI to where enterprise data resides to improve efficiency and operational responsiveness.

“Supported by the Oracle AI Customer Excellence Center in Singapore, which provides hands-on expertise to help customers in the region prototype, validate, and scale AI use cases, the collaboration with SMRT helps create new possibilities for rail operators worldwide to deliver more consistent and dependable services for commuters,” Chin said.

Apart from SMRT, Mirxes, a molecular cancer early detection company in Asia Pacific, also closely collaborated with the Oracle CEC to develop and pilot its AI assistant. Specifically, the AI assistant developed relied on Oracle AI Database and OCI Enterprise AI. The AI assistant, supported by specialized AI agents, enhances clinician productivity by delivering faster responses to scientific and product inquiries, effectively reducing response times during the pilot phase, with the goal of improving patient outcomes.