Expressways stretch across the island. Train lines run the length and breadth of Singapore, with dense housing on either side. Heavy industrial areas to the far west keep pollution away.

These were the broad elements of Singapore’s first long-term urban plan in 1971, charting out land use for the next 50 years. 45 years later, its essential features are still very much held in place.

However, there are a few “obvious challenges” this plan faces today. “Land constraint is one of the key concerns”, says Peter Quek, Chief Information Officer of the Urban Redevelopment Authority, Singapore’s city planning agency.

This is aggravated by Singapore’s system of dual government. As a city, it must have space for housing and commercial areas; and as a country, it must make space for infrastructure typically located outside a city – seaports, airports, utilities and military sites.

Make room for more

Singapore_skyline_at_sunset_viewed_from_Gardens_by_the_Bay_East_-_20120426

A newer concern is the island’s ageing population. Facilities have to be accessible to the elderly and should be located within their reach. ”Land use planning must be flexible to be able to adjust to that,” Quek says. For instance, elderly and child care centres are being co-located to improve interactions between the young and the old.

Singapore plans its urban landscape on multiple timelines. The 50-year concept plans, starting in 1971, put the broad elements of land use in place and are renewed every 10 years. These are translated into detailed masterplans, guiding property over the next 10 to 15 years. Finally, it has localised plans for precincts.

“We can’t predict the future, but we can look at scenarios”, Quek says. It looks at how population growth, changing economy and housing demand will create different scenarios 50 years ahead. “Then we can look at whether there is a policy intervention we can do to adjust this scenario.”

For instance, Singapore is encouraging people to move away from private cars. URA found that it cannot continue expanding space for roads, Quek says. “The only way is for more people to take public transport, encourage alternative transport modes like cycling, and reduce the number of cars.” The government’s trial of shared driverless cars will also lead to more efficient road use.

The planner’s toolbox

Singapore_Bishan_Park

Quek’s job has ensured that today’s planners have more sophisticated tools than the ones in 1971. Two years ago, URA set up a Digital Planning Lab that pools data across agencies and builds analytics apps to help planners make decisions.

The lab’s work is making it easier to get insights from data. Traditional geographic analytic tools require specialised skills, but these are becoming harder to come by, Quek says: “There’s just a limited pool of people specialised in GIS and data science”. And so everyone from the chief executive to the planner should be able to use these apps. “We are looking for analytics to be very pervasive in the organisation”, he says.

The multidisciplinary digital planning team of 12, overseen by Quek, includes planners, IT experts, UX designers and mathematics experts. They use the agile development method, continuously testing and improving the analytics and apps. “This is a product, not a system. It never finishes,” he says.

Here’s a look at some of the tools built by the team:

The E-Planner app was one of the first tools it built. It pulls insights from 90 datasets – the kind of work that earlier required specialised IT training. But now with two to four hours training, URA’s planner can use the app. It has made it available to other agencies too, with over 25 agencies, including housing, health and public safety, using it.

Another, more advanced tool they built is called GEMMA – GIS-enabled mapping, modelling and analysis. This is used to create future scenarios and aides in long-term planning. It can test different land use strategies, studying how a new road would affect noise levels in the neighbourhood. Or project what kind of new facilities would be needed given the demographics of a housing estate.

This kind of scenario planning used to take months to develop. GEMMA has cut the time down to days, Quek says, and creates many more possible scenarios for how Singapore will look 50 years later.

The team is also building a 3D model of Singapore, allowing URA to understand the physical impact of potential developments. “Even if a building is not there, you can bung it in and analyse the site,” Quek says. For instance, officials can look at how it affects shadows in the area and plan to build more covered walkways and plant trees.

A separate tool, built outside the digital planning lab, is a simulation of real-time commuting patterns in Singapore. The virtual model is used to test the impact of potential train lines, bus routes, housing or commercial developments.

It uses a “synthetic population” that statistically represents the demographics of Singapore, Quek says. This is combined with transport fare card data from EZ Link including location, duration and mode of travel, showing patterns of real commuter behaviour. The tool was built by URA with Singapore’s ETH-Future Cities Lab based on a Swiss project, and was in development for 10 years.

In the pipeline

sg scenery bldg

URA is also experimenting with emerging technologies that can be used by agency. Augmented reality and virtual reality are on Quek’s radar. Augmented reality could be used by planners to visualise new developments, he says. They could also be used by communications officers to explain future plans to citizens. It would create a more immersive experience for users, complementing the 3D model, he says.

Machine learning is another area “we are interested in and started to look at”, he adds. His team is testing “smart search” prototypes, which sift through huge amounts of data and dig out the right information. It would also allow planners to search for data in natural human language, like asking, “what is the population within 400 metres of a certain MRT”, Quek says.

A third area is behavioural analytics, which would help officials understand why citizens behave in a certain way. Understanding certain social norms would help URA decide where is the best place to locate facilities. It would also tell officials what kind of policy changes can be made to “shape things the way you want it to be”, Quek says.

Singapore’s landscape has seen a dramatic but carefully planned shift in the last 50 years. Densely packed slums have given way to towering skyscrapers in the central area. Rural pastures have become parks and reserves.

URA’s job is not so much to predict, but rather, to map the future of Singapore.

South Bridge Road by Nicolas Lannuzel, licensed under CC BY-SA 2.0