Kutsuplus.fi a Novel, Automated Real-time Demand Responsive. Public Transport Service

Transcription

1 10th ITS European Congress, Helsinki, Finland June 2014 Paper Number Kutsuplus.fi a Novel, Automated Real-time Demand Responsive Public Transport Service Kari Rissanen Helsinki Regional Transport Authority, Finland Abstract Helsinki Regional Transport Authority (HSL) and Ajelo Ltd. are ramping up a novel automated demand responsive public transport service based on real-time bookings. The main target group is motorists. Kutsuplus.fi is the new mode of public transport complementing HSL's offering ranging from buses to metro, trams, trains and the Suomenlinna ferry. Multiple levels of service are available: the fare is lower for the flexible customer willing to compromise his/her route selection, pick-up and arrival times for enabling more fellow passengers along the route. At lowest, the fare is close to a regular public transport journey fare, while at the best service level the customer price is still typically less than a taxi fare. The booking system shows also the matching journey planner connection and may offer an alternative taxi journey. The service operates currently with 15 minibuses in the Helsinki metropolitan area, but the control and service system architecture has been designed to scale to thousands of vehicles enabling sustainable extensions to the current service area, and naturally also to other cities, countries and continents. In the following, the kutsuplus.fi service development, ramp-up, passenger statistics, customer feedback and next steps will be discussed. Keywords: Kutsuplus, demand, responsive, public, transport, DRT, A-DRT, automatic, automated, HSL, Ajelo Background Travel behaviour surveys have been conducted in the Helsinki metropolitan area since During the over 40 years of survey, the share of public transport of the public transport and car journeys in the Helsinki metropolitan area had been continuously decreasing, although the absolute numbers of journeys had increased. According to the 2012 travel behaviour survey it seems that the trend had ceased: for the first time during the survey s history the share of public transport has increased (2008: 42 %, 2012: 43 %). However, in absolute terms, the

2 amount of private cars trips in Helsinki metropolitan area continued to increase. In 2012, some 1,1 million private car trips were made on each working day in Helsinki metropolitan area including Helsinki, Espoo, Kauniainen and Vantaa. Orbital traffic is related to this development. In orbital traffic, traditional public transport has not been at its best, as the origin and destination points of journeys tend to spread over a large area and it is therefore difficult to provide a good level of service with a traditional fixed route service. Since 2007, Aalto University undertook a research project for studying the possibility to develop a new, high-quality and cost-effective mode of public transport service that could provide a competitive alternative for private car trips. The project has been funded by the Finnish Funding Agency for Innovation (TEKES), HSL, the City of Helsinki, and the Ministry of Transport and Communications. As the research results were promising, a demand responsive public transport pilot program was launched in June 2011 for developing and gradually ramping up the novel public service that is scalable for large audiences in the greater Helsinki area by the beginning of Later, also focus-group studies and interviews have been conducted to verify that a demand exists for a public service providing a more rational choice for motorists. Additional mock-up experiments have been made to support the concepting of the novel service. The image conveyed to customers was that the service would be ecological, reliable, easy to use and personal; A convenient route for you, always. Objective The aim of the new mode of public transport is to have an increasing number of motorists switched to public transport. This could be achieved by providing a comfortable and easy-to-use service that is flexible in time and travel route as well as in terms of service and price level. When the objectives were outlined, customer time-efficiency was considered to be important in case of motorists. The strategic outline of the service design was made on this basis. The new mode of transport is to become a powerful tool in tackling congestion. The service is to offer good public transport service also in orbiting traffic and in other new areas where regular public transport has not been competitive so far. The new service is to become an increasingly flexible and personal form of public transport enabling door-to-door guided journeys, whilst competing with private car in terms of time usage, ecology, and economy. Solution The emerging solution was a new kind of real-time intelligent transport system in which the route of the vehicles was optimized in real-time by a central unit on the basis of real-time customer bookings. The passengers can themselves choose the desired level of service, and the fare they pay is determined on the basis of their selections [fig. 1, 2], and real-time 2

3 information including guidance to and from the nearest bus stops will be provided [fig. 3]. The passengers going roughly to the same direction are picked up in the same vehicle as the central unit optimizes the routing of each vehicle within the time-budgets of the passengers. Figures 1 and 2 Find Trips-page on Kutsuplus.fi and select the trip page, HSL and Ajelo Figure 3 Real time travel information on a GPS-enabled smart phone, HSL and Ajelo Reliable estimation of driving time is a critical success factor in a service based on automatically combined trips. It is mandatory for efficient operation and for smooth journeys with co-passengers. Therefore, the central unit should be aware of the traffic situation. Trip order, payment, driver s instructions and navigation information all have to happen in real-time to enable efficient and optimized real-time routing of moving vehicles. These are also pre-requisites for the service scalability in terms of service area and fleet size. Given that the above mentioned are implemented properly and the high quality service is marketed and 3

4 priced reasonably for sufficient demand, the increase in vehicular density will dramatically increase vehicular efficiency [1, 2]. On the basis of this, scale benefit, break even, profitable service, soon followed by the next big bang in transport revolution are the expected returns on the sufficient investments on service expansion. Implementation The control and service system implementation started in August In Summer 2012, the system was able to accept ride bookings and combine the individual trips on the basis of optimized routing within the passengers time budgets, and control the movement of the vehicles via Android tablets which were selected as the flexible and future-proof platform for driver s information services including navigation, passenger registration, break requests, and feedback. Also several independent real-time traffic data sources for travel-time forecast system were established to enable a reliable and smooth service. The payment system was integrated with trip ordering and payment in advance was chosen to eliminate fraud. It was decided that in the beginning, trips will be offered max 60 minutes in advance. Pre-order e.g. several hours or days earlier with confirmation as well as the recursive order with confirmation were not to be implemented in the very beginning, as the number of vehicles in the service area was very limited. Assigning a specific car to a specific customer during confirmation, long time before the pick-up could have reduced the efficiency of the limited size fleet in the very beginning due to the chaotic nature and unpredictability of traffic. However, piloting the pre-order is an important item on the future development task list. Manual clearing in relation to the fare-box revenue and transport operating fees was used in the first phase. Numerous parallel development and preparation efforts took place ranging from quotation of the fleet to equipping the vehicles and from the design of the service details to the visual image and service name selection. From Field Tests with Real Test-passengers to a Public Service The field testing of the Kutsuplus.fi service with real passengers started in October 1 st, 2012 with three vehicles and ordered trips have been taken care of reliably from the beginning. During the next few weeks the fleet was expanded to 10 vehicles. The initial service area was mainly located inside Ring Road I, excluding the easternmost Helsinki. The area included the campuses of Aalto University and Helsinki University from which test passengers were invited to book rides from any existing bus stop to any other within the area with aggressive pricing: 1,5 + 0,15 /km. Lauttasaari was added to the service area in the early 2013, Herttoniemi and Mustikkamaa in summer 2013, and Itäkeskus, Myllypuro and Tapiola in the beginning of The service has been load tested several times with real passengers in heavy demand conditions: Valentine s Day free-ride campaigns in 2013 and 2014, the public launch 4

5 Average Fee ( ) Total Fees ( ) Trips and Orders Registered Users Kutsuplus.fi a Novel, Automated Real-time Demand Responsive Public Transport Service Kutsuplus.fi service April 3 rd, 2013 on two TV channels, newspapers and other media, and the free-ride campaign May 31 st, The central system has worked reliably also in heavy-demand conditions where trip-combining is more intense. Passenger numbers and ticket income have been gradually rising although the trip prices were more than doubled to 3,5 + 0,45 /km at the public launch Kutsuplus.fi service in April 3 rd, 2013 (fig. 4, 5). In November-December 2013 the vehicle number was increased from 10 to 15, and the service time was increased from 7.30 am pm to 6 am- 11 pm and vehicular hours were doubled. By April 9 th 2014, people had registered to Kutsuplus.fi Trips Orders Registered Users Figure 4 Passenger statistics 6,00 Average fee Total Fees ,00 4,00 3,00 2,00 1, ,00 0 Figure 5 Average trip price and fare-box revenue 5

6 Service Development Kutsuplus.fi operates in continuous improvement mode. Numerous improvement and new functionalities have taken place since October These include the enhanced Digiroad map database and bus stop database with increased accuracy and complementary directional information, the introduction of alternative service classes, group discounts, friend invitation, business and family pack, service in four languages, price simulator (kutsuplus.fi/pricing), alternative means of payment including premium-sms, trip wallet loading with a credit card, and voucher. A possibility for happy hour and/or rush hour pricing is included in the system as well as lightened web order (kutsuplus.fi/m) for customers using older phone models with lesser internet browsing capability. Registration free premium sms ordering with automatic trip charging on the phone bill was added and passenger wi-fi was tested and taken into use in December Virtual bus stops have been assigned in the area since summer 2013 in order to make the bus stop network denser. In principle, any safe and feasible place within the road network can be defined as a virtual bus stop. The overall usability, reliability and timeliness have also been improving. Although the service combines trips and is dependent on the chaotic nature of the traffic in Helsinki, pick-ups have been late more than 10 minutes (20 minutes) late only in 0,5% (0,1%) of cases or less depending on the month in March Customer Feedback Feedback of the clientele, Kutsuplus.fi bus drivers and other interest groups is important and has been very encouraging. Passenger studies have shown very good results (fig. 6). Earlier studies for the registered have given otherwise very similar results, except that the busses are now more accurately in time on the departure stops than they were during the first studies. In February 2013, the trips were mostly ordered with a smartphones (33%), followed by laptops (30%), desktop in (26%), and tablets (9%). A more detailed study was made in October, 2013 to ensure that the ongoing Kutsuplus.fi service development efforts would be optimally directed. According to the study (n=442), 56% of the registered had one or more private cars in the household. The price level of offered Kutsuplus.fi trips was found (n=351): too low (1,4%), reasonable (58%), a bit too high (37%), and too high (4%). Kutsuplus.fi was found useful for (n=355) personal matters and shopping (79%), business trips (65%), trips to hobbies (50%), trips to school/work (48%) and also to various other kind of trips (12%). 6

7 Figure 6 Customer survey 2/2014 results (n=249). Currently, the control and service system has been configured to offer trips max 60 minutes in advance. The possibility to place an order 0,5-1 hours before departure and was found very important in 66% of the answers (n=353), and right before the departure (55%), 1-8 hours before (54%), 1 day before (34%), 1-7 days before (12%), and ordering trips over a week before departure (0,6%), correspondingly. However, 12% found repeated automatic weekly order very important (n=441). People who sometimes did not accept the offered Kutsuplus.fi trip claimed that (n=246): they were making trip searches for curiosity e.g. to find out the price level (63%), the pickup time of the offer was too late (24%), the price was too high (22%), the travel time was too long (22%), the arrival was too late (21%), a more suitable arrival time had to be searched as the trip searches could not be made on the basis of arrival time (21%), the walk was too long (16%), no money had been loaded in the trip wallet (14%). Other reason was found in 21% of the answers. A wider service area was found very important in 46% of the answers (n=442), service time extension from 6.30 to 11 pm (43% correspondingly), operation on weekends (41%), denser bus stop network (25%), door to door operation even with a higher price (24%), service time extension from am (17%), and service time extension from 11 pm to 4 am (13%). In October 2013 the only method for placing the trip order was via the kutsuplus.fi pages where an account accompanied with an e-wallet had to be created and into which money had to be loaded. As anticipated by HSL already in 2011, a mobile application for placing the 7

8 order was found very important in 60% of the answers (n=442), an sms-order based on using bus stop numbers (26% correspondingly), and a phone call with a reasonable additional fee included (14%). One must, however, bear in mind that the survey questionnaire was directed to the registred. They are familiar with the kutsuplus.fi service and can familiarize themselves with modern technology, but they do not represent the whole population. Other customer occasions for wider audience imply that there could also be significant need for telephone conversation based trip order where the ticket price could also be added into the telephone bill. Regarding potential future payment methods (n=441), paying in advance with the HSL travel card, which is intended for traditional public transport in the greater Helsinki area, was found very important in 47% of the answers, direct credit card payment without the trip wallet (42% correspondingly), whereas paying afterwards with a monthly invoice was found very important only in 18% of the answers. Loyalty program with discounts for regular customers was found very important in 38% of the answers. Conclusions and Next Steps It has become obvious that automated real-time scalable demand responsive transport services that are algorithm driven and can be optimized for vehicular efficiency, will have a bright future. According to the Kutsuplus.fi survey of October 2013, the majority of the current customers is convinced that they will continue using the service also in the future. Only 1,4% state that they will stick with other means of transportation. This result is very encouraging when taken into account the fact that the evolving service pilot has only been running since late 2012 and is still at its early stage of development. Kutsuplus.fi surpasses private car in efficiency in many ways. It can offer more passenger kilometres per vehicle kilometre. It can be a time-saver: no hunting for parking space, clearing the car from ice and snow. The passenger may also start the workday near the home door instead of focusing on driving - thanks to the Kutsuplus.fi service where each vehicle is equipped with a fast free wi-fi connection or the passenger can just relax and consume media. The new mode of public transport can also be efficiently used in orbiting traffic: door-to-door guided journey between any two bus stops without change of vehicle e.g. in the central bus station. It provides individual car-like service, but with a clearly higher efficiency. Automated demand-responsive service can also be efficiently used in conjunction with other public transport by locating the Kutsuplus.fi bus stops near metro, railway stations, and airports. The novel service can also be developed to increasingly support traditional public transport. Transfers e.g. from Kutsuplus.fi to train and vice versa could also be guided with smartphones and augmented reality, and when coupled with real-time information e.g. about estimated arrival time of a delayed train in a snowstorm, the demand-responsive services could adapt and still offer fluent transfers with short waiting times. 8

9 Kutsuplus.fi passenger numbers are increasing and the level of subsidizing is decreasing. Kutsuplus.fi seems already to be on its way towards an economically feasible service. These positive trends are expected to continue, provided that sufficient marketing accompanied with sufficient vehicular capacity and service area, will be offered to the customers. Naturally the development depends also decisively on the pace of introduction of new key functionalities as well as on continuous improving of the existing service on the basis of customer feedback, needs and behaviour. An easy-to-use smart phone applications enabling one-touch trip order to the favourite destinations from the detected location, confirmed pre-order at a preferred time, and repeated order will increase the demand substantially. The Kutsuplus.fi control and service system has to be scalable geographically and in time as well as in vehicle and passenger number to obtain scale benefits. It must be able to serve thousands of vehicles in the future as that will enable improvement in availability, immediacy and dramatically increased vehicular efficiency leading to increased customer satisfaction, and profitability. In a more widespread use, the environmental impacts of the system can also be significant due to reduced congestion, traffic accidents, waste of time, land usage for roads and parking places and related maintenance, and emissions. The target of HSL is to expand the Kutsuplus.fi-fleet in 2016 to 45 vehicles, in 2017 to 100 vehicles. Furthermore, during the next 10 years, the expansion is expected to continue to of vehicles once the service has become an integral part of the dynamic smartphone-assisted real-time travel chains enabling fluent door-to-door guided travelling between cities, and beyond. Acknowledgements The multi-phase intelligent transport program is a joint venture of HSL, a spin-off company Ajelo, Aalto University, the Finnish Transport Agency. References 1. Jokinen, J. Sihvola, T. Hyytiä, E. Sulonen, R. (2011). Why urban mass demand responsive transport? In Integrated and Sustainable Transportation System (FISTS), 2011 IEEE Forum on, pp Sihvola, T. Häme, L. Sulonen, R. (2010) Passenger-Pooling and Trip-Combining Potential of High-Density Demand Responsive Transport. In Transportation Research Board 89th Annual Meeting, no