In a live webinar organised by LIT Transit, our CEO Jarrold Ong talks about the role real-time passenger information play in on-demand services and gives some insights on the prediction of journey times and future mobility, based on his experience at SWAT Mobility.
Watch the full webinar in the recording below. The introduction of SWAT Mobility by our CEO Jarrold Ong starts at 21:00 minutes.
Q: Jarrold, you co-founded an on-demand mobility start-up, SWAT Mobility, almost 5 years ago. Can you share how you pivoted your start-up – what was your initial product and how it evolved?
When we started out in 2015, on-demand was pretty new, especially in Asia. We started with launching and operating a B2C consumer service. The space was very competitive with heavyweights like Uber and Grab but we gathered the important data and operational knowledge to develop the algorithms we have today.
As the popularity of ride-hailing increase over the years, many governments started to explore how newer demand-responsive technologies could be used in public transport. We had the opportunity to work with the Land Transport Authority of Singapore and also right now, Transport for New South Wales in Australia. Partnering with such public transport operators, we deployed on-demand public transport services to tackle issues like the first/last-mile problem and low ridership.
We also saw the opportunity to work directly with businesses and adapted our solutions for large corporates and industrial park developers to improve their employee's commuting experience.
Today, we have built a mobility platform to better match demand and supply in the employee transport space. We are slowly getting closer to our vision to extend this to other verticals like non-emergency medical transport, schools and even logistics. Ultimately, we want to be able to aggregate demand of entire cities and organise their movement efficiently.
Q: Many agencies are introducing ways to assure measures for maximum occupancy of public transit vehicles. Some are introducing booking apps, the others are installing automatic passenger counters and displaying real-time occupancy information to displays and other channels. How do you think such challenges should be approached? What role will passenger information have in gaining back trust?
We have been monitoring the situation closely in Singapore. The government has said that it is not possible for public transport to operate at half capacity. Firstly, the system will not be able handle the volume efficiently, and secondly, lower ridership means lower revenue and it is not sustainable in the long run. So the guidance is for safe management instead of safe distancing, which means mandatory use of masks and discouraging passenger to passenger interaction.
In the meantime, while ridership has not returned to pre-COVID levels, we have adjusted our services to stay within half capacity. This is to ensure passengers are able to leave a seat between each other for safe distancing.
I think we will see more emphasis on preventing overcrowding in public transport as people will find it very uncomfortable and stressful to be in a packed bus or train. Being able to tell how packed a vehicle would be using real-time passenger information will be very important.
Q: On-demand services were on the rise before the COVID era. Did COVID boost the deployment of new on-demand services? What are the challenges you’re facing in the on-demand market currently?
During the lockdowns, people were moving less and transport was not high on the priority list. But now as cities are reopening, we are seeing a surge in activity. With telecommuting and split team arrangements, everyone is re-looking their transport arrangements because the demand has changed.
While many of our projects have been delayed, we are seeing many new opportunities. For example, we launched an on-demand transport service with Toyota Mobility Foundation for healthcare workers in Thailand and the Philippines. It is to provide a safe and convenient transport service for frontline workers.
Many corporates are starting to look at how they can safely reopen and a large piece of that puzzle is getting their staff to the office safely.
Governments are also starting to relook at some of their past assumptions because travel patterns have changed. And if this is more permanent, they need to figure out how to adapt public transport systems to the new norm with potentially lower ridership and lower fares.
Q: What role does real-time passenger information play in on-demand services? How do you manage to timely link your services to public transit?
On-demand services usually don't operate on a fixed schedule, so real-time passenger information is critical. We have developed much technology around routing and traffic models to be able to do this very accurately. This is one of the hardest aspects to do well in as well.
We are seeing more interest in journey planning, especially for multi-modal trips. However, the industry is still figuring out how to share this data between the different services especially those that do not operate on fixed schedules.
We have not done deep integration at this point but one thing we did do in Singapore is that if there was an on-demand request, we would check if the existing fixed route services would be able to get them there faster and if so, recommend them to use that instead.
I think we will definitely start seeing more integration happen as this real-time passenger information becomes more available.
Q: Accurate and reliable journey time information serves for higher passenger satisfaction and also majorly contributes to the efficiency and thus lower costs of transport. Do you pay special attention to accurate predictions of journey times? Does journey time accuracy affect your business in any way? How shall transit agencies assure high prediction accuracy of their journey times?
We have spent considerable effort to achieve maximum accuracy wherever possible. As mentioned earlier, we had to build technology in-house to get the accuracy that we needed. There are 2 parts, the prediction and the real-time situation. Both are very hard to get right.
For on-demand services, there is no fixed schedule and we rely on accurate arrival times to pick up our passengers. We see that for many passengers, they are less concerned with their wait times when you can tell them more accurately when the bus is arriving.
As we try to do multi-modal planning, the accuracy of journey times will become increasingly important.
Q: Are on demand-services sustainable? What have you learned from the on-demand bus project that you developed with LTA? Could you share some outcomes?
When we talk about sustainability, it's not that simple. Most public transport is subsidised and very few actually make a profit. The question today is more of whether on-demand services are more cost-effective than other forms of public transport (such as fixed routes and rail). It’s still very early days in terms of the adoption and awareness but we are seeing some deployments getting very close to the cost-effectiveness of fixed routes services. They definitely have the potential to surpass fixed routes and I think we will see that happening in the next few years as we improve our operational efficiencies.
From the LTA project, we learned that awareness and service planning is extremely important for such new services and they were a challenge to get right. Nonetheless, we were able to achieve some of the project's goals like shorter journey times for passengers and also lower mileage and running costs for the operators.
Q: Jarrold, with your vast experience in start-ups – Jarrold has been in the start-up world for the last 13 years; mostly as a founder. Where do you see potential for new opportunities using real-time passenger information?
There are going to be lots of new opportunities as real-time passenger information improves. When the data gets more accurate and the industry gets better at sharing this, there is so much potential for better multi-modal planning. Imagine being able to find that perfect price-to-convenience point for every commuter.
The efficiency gains will also be immense. Imagine being able to aggregate demand of entire cities and using this information to optimise route planning. I think it will be a step in the right direction in solving some of the congestion we see in the megacities in Asia brought about by urbanisation.
I'm excited to see where we can go with this.