SMRT uses automation to boost train overhaul capacity in S$7 million Bishan Depot upgrade

SMRT has completed its two-year, $7 million project to upgrade Bishan Depot, doubling the facility’s monthly train overhaul capacity from two to four.

The upgrade streamlines work processes and introduces advanced equipment at the 39-year-old depot, which services the North-South Line operated by SMRT.

New autonomous vehicles and robots have cut manpower needs at Singapore’s oldest train depot by 30 per cent, SMRT said on Nov 21.

The project, dubbed Depot 4.0, is intended to guide the design of future depots and upgrades to existing ones, the train operator said when it announced the plans in July.

The expansion of Singapore’s rail network and addition of more trains mean that larger maintenance facilities and more manpower are required, but this comes amid a shrinking and ageing workforce, SMRT said.

With the upgrade complete, all train overhaul works will now be carried out at Bishan Depot, which is about the size of 55 football fields. This allows Tuas West Depot, the other train depot where overhauls were done, to focus on other types of maintenance work.

Typically, trains are overhauled for every 500,000 km and 1 million km travelled, with different parts of the train restored at each mileage milestone.

The upgraded depot was launched by SMRT chairman Seah Moon Ming. Also in attendance were Acting Transport Minister Jeffrey Siow, the guest of honour, and former transport minister Khaw Boon Wan.

Depot 4.0 is a joint initiative with Chinese train manufacturer CRRC Qingdao Sifang, which has produced trains for the North-South and East-West lines, as well as the Thomson-East Coast Line in a joint venture with Kawasaki Heavy Industries. It will also supply trains for the upcoming Cross Island Line.

The upgraded depot uses automation to clear longstanding bottlenecks in the overhaul process. This halves the time needed to overhaul a train, allowing the depot to double output without expanding its footprint.

It brings together autonomous vehicles, robotic systems and digital tools to speed up the movement and handling of heavy components, including a new control system that coordinates these machines across the workshop and allows engineers to plan, monitor and analyse maintenance activities in real time.

Replacing overhead cranes with robots

The depot has replaced its overhead cranes with a fleet of 11 rail-guided vehicles (RGVs), or autonomous robots, which are used to move heavy train components such as bogie frames and can each carry loads of up to five tonnes. A bogie is the undercarriage of a train. Each train car has two bogies, so a six-car train has 12.

Previously, these bulky parts often sat idle in queues till a crane was available to lift them.

The RGVs’ height-adjustable platforms allow staff to work more comfortably on the frames, reducing strain and the likelihood of back injuries, SMRT said. The robot’s controlled movements also improve safety by reducing the risks of manual lifting and overhead operations.

Another addition is the new bogie disassembly and assembly system, which automates the labour-intensive process of separating bogie frames from wheelsets and reassembling them.

Previously, the task relied heavily on the overhead cranes, leading to delays as engineers waited for access. The work also required repeated manual adjustments to align wheelsets with the frame.

The new system links directly to the RGVs, which deliver bogie frames to workstations and collect them once the process is complete.

It has reduced manpower needs from four engineers to one, while shortening the time it takes to disassemble and reassemble each bogie, SMRT said.

Driverless vehicles used to deliver smaller train parts

The depot now uses two automated guided vehicles (AGVs) to handle smaller train components such as traction motors and tread brake units. Previously, these items were moved manually or transported by forklifts, creating delays as parts waited for forklift access.

The AGVs can carry up to one tonne and deliver parts directly to designated stations. The driverless vehicles navigate using on-board sensors and proximity detection systems that allow them to stop safely around people and obstacles. Staff can also input routes using a built-in control panel.

By removing the need for forklifts and manual handling, the AGVs reduce queue times, improve ergonomics, and lessen the physical strain on staff, SMRT said.

SMRT has also rolled out a dry-ice cleaning machine that its staff use to clean train components, instead of wash cloths and buckets of water.

The method saves water, eliminates the time required for parts to dry, and removes the risk of slipping on wet floors.

Centre showcasing ground-up staff initiatives launched

On Nov 21, SMRT also launched its new Kaizen Centre, which showcases ground-up staff initiatives to improve productivity and safety.

“Kaizen” is the philosophy of making steady, meaningful improvements.

Launching the centre, Mr Seah, the SMRT chairman, said that since 2018, employees have completed nearly 9,000 projects, which have collectively generated about $700 million in savings over seven years.

“With these savings, we can also do more for the community,” he said.

SMRT will partner SG Enable to distribute 18,000 Nets FlashPay cards, which can be used to pay for bus and train fares, to people with disabilities and their caregivers. Each card has a stored value of $38, in celebration of SMRT’s 38th anniversary.

The Kaizen Centre highlights projects across four key areas: transformative initiatives, safety enhancements, sustainability efforts and productivity gains.

Depot 4.0 is one such initiative.

Another example is the Project Overwatch initiative, which uses artificial intelligence to monitor train services.

The centre will primarily serve SMRT’s internal teams and is not open to the public. It will also host invited local and overseas visitors from government agencies, institutions, railway operators, and industry and community partners.

Note: This article was written by Vanessa Paige Chelvan and first appeared in The Straits Times on 21 November 2025.

Source: The Straits Times