Funded by EPSRC and led by Rym M'Hallah, professor of systems engineering at King’s College London, the three-year project will develop a door-to-door ride share service through the development of novel algorithms that optimise routing and scheduling integrated with dynamic pricing.
“This project aims to make DRTS more attractive, flexible, accessible, available - in rural areas - and inclusive,” said Prof M’Hallah. “It will reduce congestion, accident risks, emissions, and pollution-related health issues. It will improve life quality by connecting people at a reasonable fare.”
DRTS aims to overcome the limitations of services such as Dial-a-Ride, which relies on large subsidies from local councils and the Department for Transport, lack route planning flexibility and cannot manage high demand.
To this end, the project will develop a scheduling and routing optimisation algorithm for a fleet of green vehicles that can provide instant accept/reject decisions on journey requests from riders. To do this, the algorithm will anticipate potential future demand and be continuously globally optimising schedules across the fleet in the background.
Financial viability will be ensured with revenue management formulations that allow the prices of journeys to be changed dynamically, with prices dependent on journey length and service quality.
The project will also develop a deep understanding of customer behaviour and preferences, obtained by running surveys and focus groups and using the data collected to build so-called choice models that describe how potential passengers make decisions.
DRTS has a strong dependence on the probability that a rider accepts a suggested journey with given price and time windows, explained Prof M’Hallah. Neither starting nor finishing points are fixed; the bookings are dynamic with pickup and drop-off time windows, while demand is uncertain and based on evolving customer behaviour.
“These new components…provide additional challenges for dynamic pricing,” she said. “While pricing for last mile and food delivery has some similarities, existing pricing techniques are unlikely to be efficient on this more complex large-scale problem. Most interestingly, the interdependency between the constantly evolving routing and scheduling and the committed pricing offer is new.”
Prof M’Hallah continued: “Calculating the opportunity cost of accepting a request is needed for setting optimal prices. It requires generating a new route and schedule and estimating their impact on future purchase behaviour. This is a novel, exciting advance.”
Similar efforts have been tried before, notably Kutsuplus in Helsinki, a defunct project to roll-out user-centred mobility on demand.
“While [DRTS] will take advantage of the lessons learnt, it also aims to overcome some of [Kutsuplus’] limitations: the adopted operating model was not financially sustainable and most riders were already public transport users,” said Prof M’Hallah. “Unlike Kutsuplus, the proposed DRTS will employ a mixed fleet of green vehicles, not only taxi types. The routing adapts vehicles’ capacity to existing and forecasted demand so that it optimises the occupancy and utilisation rates of the fleet. It nudges private car, taxi, and Uber users toward a public transport-like service with the comfort and flexibility of a taxi. Changing private car users’ travel behaviour is among its primary goals.”
Partners in the project titled ‘A scaled and sustainable demand responsive transport service’ include Mott MacDonald, Leeds City Council, Busreinvented.com, CUSP London, and West Yorkshire Combined Authority.