Delhi Metro Rail Corporation

Delhi Metro Rail Corporation advances urban mobility with AI on Google Cloud and predictive analytics

Google Cloud results
  • Predicts failures early, keeping passenger journeys smooth, with Gemini in Vertex AI

  • Gives operations teams instant access to unified data with BigQuery

  • Detects faults in real time to improve service reliability with Dataflow

  • Aligns teams with 300+ shared performance dashboards using BigQuery and KPI dashboards

  • Cuts service disruptions, targeting 10-20% less downtime for passengers, with predictive analytics in Vertex AI and Dataflow

The Delhi Metro Rail Corporation uses Google Cloud to move to predictive maintenance and reduce downtime across operations.

Delhi Metro

Building a data-driven foundation for smarter operations

The Delhi Metro Rail Corporation (DMRC) is one of Asia’s largest transit networks, carrying more than seven million commuters every day across the Delhi-NCR region. With 395 kilometers of track and 289 stations, DMRC has long been a leader in transportation efficiency and safety in India’s capital. To boost rail operations and set global benchmarking standards for metro performance, DMRC wanted to move from traditional operations to a smarter, data-driven approach. But a major hurdle stood in the way: critical data was trapped in separate systems, and without a single view of the network, it was difficult to make fast, informed decisions.

Dr Amit Kumar Jain

To solve this, DMRC partnered with Deloitte to build a Data Center of Excellence (DCoE) on Google Cloud, selected for its ability to handle massive scale securely. For example, forecasting asset failures—across rolling stock, signalling gears, traction systems, lifts, and escalators—was a challenge due to a reliance on prevalent maintenance practice and manual monitoring of DMRC’s stations. This meant teams were often reacting to problems after they appeared, rather than preventing them, which increased the risk of service disruptions and delays for passengers.

With Google Cloud, DMRC is now able to process high-volume data from engineered assets in real time. This shift allows maintenance teams to work with the precision of an F1 pit crew, utilizing Vertex AI to identify early fault patterns, such as door malfunctions or wire sagging, before they cause delays. This transition is transforming maintenance from a manual chore into a strategic advantage, monitoring 300+ standardized KPIs to target a 10-20% reduction in downtime and ensure smoother rides for millions of passengers every day.

Enhancing safety and reliability with AI-driven maintenance

For these seven million daily commuters, a seamless journey depends on the reliability of thousands of moving parts—from lift doors to overhead traction wires, and assets that need to perform flawlessly around the clock. To achieve this level of reliability, DMRC deployed advanced AI/ML solutions, trained and hosted on Vertex AI, to interpret the massive streams of data generated by the metro's engineered assets.

We are using predictive maintenance for track, signalling gears, bearings of rolling stock, lifts, and escalators. Additionally, the Passenger Information Display System (PIDS) provides real-time updates, and the Communication-Based Train Control (CBTC) system on the magenta and pink lines improves train automation and safety. These projects collectively enhance efficiency, sustainability, and user convenience.

Dr. Amit Kumar Jain

Director (Operations and Services), Delhi Metro Rail Corporation (DMRC)

On the tracks, the focus goes beyond simple monitoring. For rolling stock, the system analyzes event data directly from the Train Control Monitoring System to identify early-stage anomalies in critical components, such as brakes and bearings. Simultaneously, ‘Point Machine’ monitoring data is processed through Dataflow and analyzed on scalable Google Kubernetes Engine (GKE) clusters to assess the health of signalling gears, predicting potential failures well in advance.

This data-driven intelligence extends into the depot, where DMRC is transforming manual workflows with Generative AI. Gemini in Vertex AI helps the system decipher complex depot maintenance logs to automatically assign the correct sub-systems for repair. This capability gives staff accurate asset histories instantly, contributing to a projected ~15% time savings in maintenance planning and ensuring that every repair is based on precise, historical context.

Beyond the engineering teams, the platform empowers leadership through an interactive app interface built on Google Cloud, which provides real-time visibility into operational performance and enables timely strategic interventions from anywhere. In the background, Cloud SQL is used for maintaining logs and configuration tables, while Cloud Run is used for all small transactional processing functions. Underpinning all of this is a modular cloud architecture that standardizes data integration, ensuring that as DMRC expands its network, this digital backbone is ready to scale with it.

Setting the stage for a scalable blueprint for the future of global mobility

This initiative by DMRC, with support from Deloitte, marks a transformative leap toward a smarter, more passenger-centric urban transit ecosystem. By moving from fragmented silos to a unified cloud-based platform, DMRC has achieved 100% faster access to operational information, creating a single source of truth for decision-making.

Looking ahead, the value of this DCoE extends well beyond its current daily operations. It is designed with future expansion in mind, given the platform utilizes a modular cloud architecture that acts as a reusable framework for new lines and projects. This strategy allows DMRC to leverage reusable cloud assets, significantly reducing redundant development efforts and costs as the network grows.

This shift to automated reporting also drives substantial workforce optimization, eliminating the need for ad-hoc dashboard generation and allowing teams to focus on strategic improvements rather than data extraction.

Ultimately, with Google Cloud solutions, DMRC positions itself as a standard-bearer for the global transit industry—optimizing travel for Delhi’s commuters while emerging as a global leader in metro analytics. This sets the stage for international collaboration, proving that a digital-first approach can redefine what is possible in public transportation efficiency.

The Delhi Metro Rail Corporation plans, builds, and operates one of Asia’s largest metro networks, serving 7M+ daily riders across the Delhi-NCR region with 395km, known for safety, efficiency, and innovation.

Industry: Transportation

Location: India

Products: Vertex AI, Gemini LLM, BigQuery, Dataflow, CloudSQL, Cloud Storage, Google Kubernetes Engine (GKE), Cloud Run


About Google Cloud partner — Deloitte

Deloitte is a leading global provider of audit and assurance, consulting, financial advisory, risk advisory, tax, and related services; and is a Google Cloud global partner.

Partenaires Google Cloud
  • Deloitte
  • Faites des économies grâce à notre approche transparente concernant la tarification
  • Le paiement à l'usage de Google Cloud permet de réaliser des économies automatiques basées sur votre utilisation mensuelle et des tarifs réduits pour les ressources prépayées. Contactez-nous dès aujourd'hui afin d'obtenir un devis.
Google Cloud