Document Sample
 Successful Use of Automated Booking and Call-back for
  Demand Responsive Transport Services in Gothenburg

                                    Yngve Westerlund
                                 LogistikCentrum Väst AB,
                    Taljegårdsgatan 11A – SE 431 53 Mölndal, Sweden
       Tel: +46 31 27 00 24 – Fax: +46 31 27 94 42 - e-mail:

                                    Dr Agneta Ståhl
                  Department for Technology and Society, Lund University
                            Box 118 – SE 22100 Lund, Sweden
        Tel: +46 46 222 9132 – Fax: +46 46 12 32 72 - e-mail:

                        Dr John Nelson and Dr Jenny Mageean
        Transport Operations Research Group, University of Newcastle upon Tyne,
                 Claremont Tower, Newcastle upon Tyne, NE1 7RU, UK
       Tel: +44 191 222 6547; Fax: +44 191 222 8352 - e-mail:


The city of Gothenburg has pioneered the development of Flexlinjen (FlexRoute in English),
a new form of flexible public transport specially tailored to the needs of elderly and disabled
persons. This version of Demand Responsive Transport (DRT) was first successfully
demonstrated in the EU project SAMPO in the Telematics Applications Programme in 1996-
97. In the SAMPLUS follow up project 1998-99 the objective was to modify and improve the
concept and to validate a new function for automated booking of return trips (back to the
home). The conclusion of the extended evaluation (over three years) is that user acceptance for
the concept is extremely high and that also the telematics functions are well received, even
though the average user is 77 years old. One quarter of all trips are booked automatically.
There are now plans to extend the DRT service to the whole city, which will require a fleet of
some 30-40 minibuses. The traffic concept and supporting technology have a large potential to
increase the efficiency of DRT. It can be extended to a new generation of flexible transport,
including new forms of affordable taxi services using low floor maxi-taxi vehicles for 7-8
passengers that are accessible for all potential users.


Demand Responsive Transport (DRT) services represent an intermediate form of public
transport, somewhere between regular bus services and exclusive ride taxi services. DRT
services are routed according to the needs of the customers, generally only stopping where
passengers request collection or dropping off. Over the last decade DRT has grown in
popularity for several reasons which include the shortcomings of conventional regular bus and

taxi services and new developments in community transport. Traditional Dial-a-Ride services
and Special Transport Services (STS) for users with special needs have often been criticised
because of their relatively high cost of provision and their lack of flexibility in route planning
with long pre-booking times. The potential for overcoming these limitations may be realised
through the introduction of telematics-based DRT and this has been widely demonstrated in
the DGXIII funded SAMPO and SAMPLUS projects.

This paper presents selected results of the evaluation of the successful introduction of DRT in
the city of Gothenburg, which has pioneered the development of a new form of flexible public
transport (Flexlinjen). This service is currently specially tailored to the needs of elderly and
disabled persons but could in the future also be opened up for other user groups.


This section briefly presents new concepts of (DRT) service provision emphasising the
advantages of route flexibility. Several new service concepts have been developed as part of
the SAMPO and SAMPLUS projects (1) (2) and these are illustrated in Figure 1.

Two main types of service operate, namely:
• A corridor service which typically operates between two end stop points; and
• An area service beginning and ending at the same end stop point or with no end points.

Awaiting an international standard for terminology in this domain, the SAMPLUS consortium
adopted the following definitions:

(a) Stopping points. These are comprised of the following. “End stop points” (terminals)
    which are locations where routes begin and end. “Fixed intermediate stop points” are
    conventional bus stops. “Predefined stop points” are recognised meeting places, which
    may be marked by a signpost. “Non-predefined stop points” are not fixed and are
    generally the doorstep of the user. The vehicle only stops at these predefined (stop-to-
    stop) and non-predefined (door-to-door) stop points if requested in advance by pre-

(b) Route flexibility. “Semi-fixed routes” are characterised by a DRT service departing from an
    end stop point (terminal) at prescribed times, e.g. every two hours. The vehicle will stop
    at conventional fixed intermediate stop points. In addition there will be deviations to other
    stop points on request. “Flexible routes” have a service departing from an end stop point
    at prescribed times. The vehicle will only call at stop points on request and there are no
    fixed intermediate stop points. “Virtual routes” have a journey where there are no end or
    fixed intermediate stop points. This is more like a conventional taxi service (but with
    shared rides), as it operates for a set number of hours, crossing the DRT service
    operational area with no scheduled departure times from any stop point. The vehicle will
    only call at stop points on request.

The Gothenburg test site (which is described in detail below) has implemented and evaluated a
corridor-based fully flexible route with both predefined and non-predefined stop points.

                               THE GOTHENBURG TEST SITE

Background and Objectives

Göteborg (Gothenburg) is the second largest city in Sweden with a population of about
500,000. In 1992 the city invested in PLANET, a very advanced DRT system for Special
Transport Services (STS) with shared-ride taxi and specially equipped vans. PLANET handles
the booking, planning and dispatching of over 5,000 demand responsive trips per day, mostly
by elderly and disabled persons who are unable to use regular public transport. As can be
expected such a service requires a high degree of municipal subsidy (about 15 million

Provision of door-to-door STS with only a 15 minute pre-booking requirement is very
expensive and budget restrictions are forcing a change. Even though 5% of the population has
STS permits, there still remains a large portion of elderly with mobility problems. This is
related particularly to long walking distances and the hilly terrain in many parts of the city.
The objectives of the demonstration in the SAMPO project were to develop, plan and validate
a new attractive and efficient form of local transport to:
•   Reduce the city’s cost for local STS taxi provision with a minimum of service reduction;

•   Increase mobility (at a reasonable cost) for elderly who are not eligible for STS but still
    have difficulty using regular public transport; and

•   Reduce administrative costs for DRT by using advanced transport telematics functions.

Telematics for a Flexible Transport Solution

In order to meet the above objectives an intermediate form of flexible transportation (between
taxi and scheduled public transport) was invented with the following characteristics:
•   Fully accessible minibuses (12-14 passengers) with low-floors (Figure 2).

•   Two fixed end points with scheduled departures (every 30 or 60 minutes in each

•   A fully flexible service between end points routed according to requests for collection and

•   A door-to-door service for STS eligible persons and frequently posted “meeting points”
    that assure a walking distance of <150 m for other users.

•   Friendly drivers for a very personal service.

This service is characterised as a demand responsive “service route” and has been branded and
marketed as Flexlinjen, which translates in English to FlexRoute.

The supporting technology for the new traffic concept is an enhanced FlexRoute version of
the PLANET system for booking, planning and dispatch at the existing STS call centre.
The following functions were added during the SAMPO project in order to increase efficiency:

•   Automated call-back with a confirmation or a delay notification 15 minutes before the
    previously given pick-up time (Figure 3). This enables a larger optimisation window and
    thus higher productivity. It is also an assurance to the customer about the bus arrival time.

•   Automated booking of travel, from both touch-tone phones at home and GSM.
    In particular the outbound trip (as soon as possible from home) can be booked with a very
    simple dialogue using Interactive Voice Response System (IVRS) technology.

Since a new driver booking function with PCs in the vehicles failed the validation in SAMPO,
a new return booking function was demonstrated in the SAMPLUS project:
•   A card reader booking function has been installed at four major destinations. Users need
    only sweep their STS or Flexcards (customer cards similar to credit cards) in the machine
    and press a button. This means, “I want to travel home as soon as possible.”
    The request is processed in half a minute and the booking is verified with a small receipt
    from the connected printer.

In a co-operative effort with the TELSCAN project during 1999, the accessibility for disabled
persons to the various telematics functions in the FlexRoute service was assessed.
An ergonomic study of the magnetic card booking function resulted in a conceptual design
that allows both sitting users (in wheelchairs) and standing users to swipe their customer cards
in the card reader. The physical design of this piece of furniture was done by L O Design in
Tibro with support from Västtrafik AB (Figure 4).

Extended Validation

The SAMPO demonstration with four minibuses started in October 1996 in the Högsbo
district where almost a third of the population of 16,000 is elderly (65+). The service area
extends over an area of about 8 square kilometres. The route is fully flexible and runs between
Frölunda Torg, a large shopping centre, and Sahlgrenska, a major regional hospital. One bus
depart in each direction every half hour and must reach the other end stop within 55 minutes
in order to be ready to depart in the other direction. Passengers need to pre-book their travel at
least 15 minutes before the bus leaves the end point so that other passengers down the “route”
can be notified about the possible delay caused by additional bookings.

After the initial one year SAMPO demonstration FlexRoute became a permanent service with
some expansion of the traffic area so that it now includes some 75 meeting points. Service is
provided weekdays from 09:00 –17:00 and the fare is the same as for regular public transport.

The positive evaluation of the FlexRoute traffic concept from the SAMPO project was
confirmed by the extended validation in SAMPLUS after three years of operation. The
evaluation of user acceptance was conducted by Lund University and LogistikCentrum (3).


The key findings from the evaluation may be summarised as follows:

•   Ridership is still increasing after three years and the total volume approaches 5,000
    passengers/month in Högsbo, or an average of over seven passengers/vehicle hour.

•   Two out of three STS eligible users prefer FlexRoute to the optional STS shared ride taxi.
    The costs for local STS taxi provision in Högsbo have been reduced by about 70%.
    The resulting savings in the STS authority cover about half of the operating costs for
    FlexRoute, fare revenues a quarter and the rest is the city’s marginal cost for providing
    senior mobility and enabling participation. It is likely to be offset by several (difficult to
    measure) cross-sector benefits, e.g. in health, home care and through a more active life.

•   Users are extremely satisfied with the traffic concept (9.7 on a 10-point scale). The drivers
    are the most important success factor, contributing to the “mobile social club” atmosphere
    of this travel mode.

•   User acceptance for the new telematics is remarkably high, considering that the average age
    of the users is 77 years. The overall attitude toward the telematics functions is 8.8 on the
    10-point scale, based on 178 responses. Table 1 provides a more detailed rating.

                                                    Rating (10-point scale)
         Function                             Concept/idea      How it really works
         Automated call-back                      9.1                    8.4
         IVR booking (from home)                  8.4                    8.7
         Card booking (back to home)              7.5                    5.9

                          Table 1: Users’ rating of telematics functions

•   After the introduction of monetary incentives in May 1999, the two automated booking
    modes have reached about 25% penetration. This is a very good result, despite the fact
    that the card booking function still needs some refinement, e.g. using a smart card with a
    chip instead of magnetic strip as the media.

Based on the positive assessments of FlexRoute a method has also been developed for
planning the future of flexible transport in Gothenburg. This has been used to create a “master
plan” for extension of Flexlinjen from the districts of Biskopsgården and Högsbo to the whole
city, which will require a fleet of some 30-40 minibuses as compared to the 6 vehicles used
today and an additional 10 routes (4). Several other cities in Sweden, including Stockholm,
Alingsås and Uddevalla, have already introduced or have plans for similar services.

As a follow up to SAMPLUS, development of a new generation of fully accessible vehicles
has been initiated in Sweden. These include a minibus with low/adjustable floor height and a
flexible interior layout for quick change between different transport applications and a maxi-
taxi with the same general functionality (5).

                                       C ONCLUSIONS

The results presented in this paper indicate that there is a great market potential in using this
enhanced DRT concept, the telematics based supporting technology and advanced vehicles to
provide attractive and efficient transport in a society with an increasing number of elderly
people. However, it can also be expanded to a new generation of flexible transport services
(e.g. new taxi or limousine schemes) for the general public or for other special user categories
such as safe women transport schemes. By adopting ride-sharing schemes and using the
passengers willingness to be somewhat flexible in their requested collection time it seems
possible to approach unit costs that are about half of that for conventional exclusive ride taxis.
In this case such schemes can be interesting alternatives or complementary to regular public
transport in low demand areas and in low demand periods.


The authors wish to acknowledge the City of Gothenburg, through the Special Transport
Services Administration and the Traffic and Transport Authority for providing the test site
for the above demonstrations and studies, and also the telematic technology providers in the
Gothenburg project, PLANit Sweden AB and Telia Publicom AB, for their efforts to make the
project a success. We also wish to acknowledge the numerous other partners in the SAMPO
and SAMPLUS projects for their input to the general discussions on DRT concepts, user
needs etc and particularly the project coordinators (Ministry of Communications in Finland
during SAMPO and Transport and Travel Research Ltd in the UK during SAMPLUS).
Financial support has been received from the EU Fourth Framework research programme and
from KFB, the national Swedish communication research funding programme.


1. Westerlund, Y. et al, 1997. Demonstration of Demand Responsive Transport in four
   SAMPO sites. Deliverable no. D7 of the SAMPO project. European Commission, DG
   XIII, Telematics Application Programme (TAP), Transport Sector.

2. Mageean, J.F. et al, 1999. Results of the Evaluation and Market Assessment of
   SAMPLUS Technologies. Deliverable no. D7.2 of the SAMPLUS project. European
   Commission, DG XIII, TAP.

3. Stahl, A. and Westerlund, Y. 1999. Traveller Evaluation and User Acceptance for the
   FlexRoute DRT Service in Gothenburg. Internal Report of the SAMPLUS project.

4. Westerlund, Y. et al, 1999. Flexible Public Transport in Gothenburg. Experience of
   FlexRoute Demonstrations and Study of Future Possibilities. (in Swedish only).
   KFB Report 1999:26, Stockholm.

5. Stahl, A. and Westerlund, Y., 1999. Development of Accessible Multipurpose Vehicles.
   Presentation at the TRB 16th National Conference on Accessible Transportation &
   Mobility, Phoenix Arizona, March 1999. Available through the authors.

Figure 1 Examples of DRT Service Route Concepts

End stop point                         Predefined stop points
(always stop here)                     (only stop here if pre-booked)
Fixed intermediate stop point          Non-predefined stop points (door-to-door)
(always stop here)                     (only stop here if pre-booked)

Fixed Route

 Corridor Service:    No deviation
                      2 end stop points
                      Fixed intermediate stop points

An area-wide service will typically have a circular route with 1 end stop point.

Semi-fixed Route

 Corridor Service:    Deviation to predefined stop points
                      2 end stop points
                      Fixed intermediate stop points

Other corridor services may deviate only to predefined and non-predefined stop points
or to non-predefined stop points only. Area-wide services may have the same
combinations of stop points but only 1 end stop point.

Flexible Route

 Corridor Service:    Deviation to predefined and non-predefined stop points
                      2 end stop points
                      No fixed intermediate stop points

This is the Gothenburg type of service. Other corridor services may deviate to either
predefined or non-predefined stop points (door-to-door).

 Area-wide Service:    Deviation to non-predefined stop points (door-to-door)
                       1 end stop point
                       No fixed intermediate stop points

Other area-wide services deviate to predefined and non-predefined stop points or to
predefined stop points only.

Virtual Flexible Route

 Area-wide Service:    Stop at non-predefined stop points
                       0 end stop points
                       No fixed intermediate stop points


Alternatively, there may be predefined stop points.

Figure 2 Low floor mini-buses are used in FlexRoute for comfortable boarding and alighting

Figure 3 The automated call-back function confirms the pick-up time 15 min in advance or
advises of any planned delays in good time for the user to get dressed and go to the nearest
meeting-point or in some cases to the door step.

                                      “This is from
                                      FlexRoute to confi
                                      that we will pick
                                      up at Bank street
                                      9.30 a.m. “

Figure 4 Passenger use a magnetic swipe card to order a return booking from a shopping
centre. The terminal is adjustable for wheelchair and standing users.