Home > Knowledge > Stories > Welcome to the future with personalised transportation


Welcome to the future with personalised transportation

It is a sci-fi fantasy for most of us – thinking of the day when transport services will be available at our beck-and-call. “Imagine that you have to drop off your kids before heading off to work. You open an app on your smartphone, where you put in details like your start and end destinations, pick up time and the money you are willing to pay. With these details, you then get a personalised pick-up with minimal waiting time. This concept of Mobility-as-a-Service (MaaS) will be a reality in the near future,” says Dr Muhamad Azfar Bin Ramli, a Research Scientist from the Complex Systems Group (CxSy) of the Institute of High Performance Computing (IHPC) at A*STAR.

The authorities in Singapore are gearing up to make this a reality. Moving in this direction, the Land Transport Authority (LTA) of Singapore recently awarded nearly half a million dollars in contracts to two companies to carry out on-demand bus service on a trial basis. .

Making this vision a reality is a group of data scientists like Dr Ramli at A*STAR’s IHPC. These scientists use complex computational modelling and simulations to analyse and predict transport demand and supply; and study various research problems relating to future transport modes and challenges.

1. The Psychology of Urban Mobility

Singaporeans rely heavily on public transport. Although private car hire services like Grab have greatly expanded their services here, the volume of public transport continues to increase. Latest data released by LTA shows that local residents are making an astounding daily average of 7.2 million trips on public transport. This includes bus and train ridership which rose by 4.3 per cent last year.

“As more and more residents rely on public transport systems, even small service interruptions can cause a ripple effect. In the case of service breakdowns or delays, people also look for alternative routes that will get them to work very quickly,” explains Dr Ramli.

He explains that to understand the effects of disruptions in areas with high concentrations of commuters, data scientists need to have a clear idea of the travel patterns from data collected daily from fare card systems. This set of big data is then analysed to understand the fluctuations arising from periods of peak and non-peak demand. Modelling how commuters make choices on alternative routes and modes, requires more than data – it requires an understanding of user psychology.

“We have psychologists in our team who help us understand how crowds will react to peak period disruption and how they may plan their alternative routes,” Dr Ramli explains. A clearer understanding forms the basis of the next step – The Stress Test.

2. Stress Test

Dr Ramli’s team has utilised the fare card data obtained from daily commuting patterns to create computer-based models that can provide finer details on demand and traffic flows. “One of the aims of our project team is to understand and quantify the resilience of a rapid transit network by measuring the capability of the system to maintain its efficiency in times of disruptions,” he explains.

The project utilises a mix of data science, analytics, visualisation and complexity science – more specifically complex network analysis – as well as resilience analysis.

“To put it simply, we create computer-based simulations of breakdowns. For example, we can simulate a train breakdown on the East-West line during the peak hours on a typical weekday. This simulation will help us understand questions like – what will be the ripple-effect on other lines, what alternative routes are available to commuters and what will be the resulting impact on the other MRT lines and buses?” says Dr Ramli.

3. Getting Future Ready

So how are these tools ultimately used? “Our models help the local transport authorities understand island-wide commuting patterns and how transportation routes and systems can be optimised,” explains Dr Ramli.

The results from this analysis could also be used to inform commuters of the benefits of additional rail lines especially in terms of increasing options of viable alternative paths during their regular commute.

That is not all – the team is also working on understanding how ‘personalised commuting’ where transport services will run based on demand.

Until then, happy commuting!

Dr Muhamad Azfar Bin Ramli
Dr Muhamad Azfar Ramli received his B.Eng. and Ph.D. from the Department of Mechanical Engineering of the National University of Singapore (NUS). He is currently a research scientist in the Computing Science Department at A*STAR’s Institute of High Performance Computing (IHPC). His current research interests include data science and analytics, modelling and simulation of urban systems, complexity science and networks and random/stochastic processes.