Transportation Research Part E: Anselmo Stelzer, Frank Englert, Stephan Hörold, Cindy Mayas

Transportation Research Part E xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Transportation Research Part E

journal homepage: www.elsevier.com/locate/tre
Improving service quality in public transportation systems

using automated customer feedback
Anselmo Stelzer a,⇑, Frank Englert b, Stephan Hörold c, Cindy Mayas c
a
Railway Engineering, TU Darmstadt, Otto-Berndt-Str. 2, 64287 Darmstadt, Germany
b
Multimedia Communications Lab, TU Darmstadt, Rundeturmstr. 10, 64283 Darmstadt, Germany
c
Dept. of Media Production, TU Ilmenau, Gustav-Kirchhoff-Str. 1, 98693 Ilmenau, Germany
a r t i c l e i n f o a b s t r a c t
Article history: In this paper the necessity for standardised automated information exchange between
Received 24 December 2014 travellers and transportation company is evaluated to improve the service quality of public
Received in revised form 28 March 2015 transport. Therefore the needs and expectations of transportation companies and travellers
Accepted 13 May 2015
are defined and the usage of a novel approach for bidirectional information and communi-
Available online xxxx
cation systems in public transport is proposed. As a result, application scenarios for the
usage of customer information are described and the advantages of this novel approach,
Keywords:
especially for dispatching processes, are highlighted. Furthermore, the benefits for cus-
Public transport
Dispatching
tomers and transportation companies in regard to service quality are pointed out.
Customer feedback Ó 2015 Elsevier Ltd. All rights reserved.
Conflict
Service quality
1. Introduction
Public transport serves the society by providing cheap and fast mobility services. Due to long term experiences over dec-
ades, the provided services are highly reliable and affordable. Traditionally, routes, intervals and vehicles are planned top
down based on a priori knowledge about the traveller flows and desired interchanges. Due to the slow changing nature
of traveller flows and a huge basis of domain knowledge, this planning process works reasonably well when every actor
involved is on time. However, in case of short term interruptions, delays or even cancellations ad hoc dispatching is neces-
sary to counteract the disruption. In those situations, two factors limit the service quality. First, the dispatcher only has a
priori knowledge about the local situation. Thus, it is impossible to make optimal dispatching decisions for all travellers.
Second, the dispatcher’s reactions to interruptions are not disseminated efficiently to the affected travellers. This often cre-
ates uncertainty (Cambridge Systematics, 1999), which limits the travel experience and the perceived service quality (EN
13816, 2002).
In former times, it was impossible to gather and process individual journey information for each traveller in a timely
manner. However, this has changed due to the wide availability of modern communication systems like smartphones,
mobile data plans and high performance servers. Having today’s communication and data processing capabilities, it is pos-
sible to collect, process, and include in situ information from all travellers within the operational decision processes and to
disseminate timetable changes in real time. Informing the traveller about dispatching decisions or timetable changes
⇑ Corresponding author.
E-mail addresses: stelzer@verkehr.tu-darmstadt.de (A. Stelzer), f.englert@kom.tu-darmstadt.de (F. Englert), stephan.hoerold@tu-ilmenau.de (S. Hörold),
cindy.mayas@tu-ilmenau.de (C. Mayas).
http://dx.doi.org/10.1016/j.tre.2015.05.010
1366-5545/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Stelzer, A., et al. Improving service quality in public transportation systems using automated customer
feedback. Transport. Res. Part E (2015), http://dx.doi.org/10.1016/j.tre.2015.05.010
2 A. Stelzer et al. / Transportation Research Part E xxx (2015) xxx–xxx
Automated Traveller
Feedback
Transportaon Enhanced Travel

Traveller
Company Experience
Benefits Realme Informaon Benefits:

- Beer dispatching Disseminaon - Improved reliability
- Enhanced planning - Higher consistency
- Improved quality - Beer comprehensibility
management
Fig. 1. Benefits gained by enhanced communication facilities between traveller and transportation company.
reduces the uncertainty, increases the consistency and causes a better comprehensibility for the traveller. This gives the pas-
sengers the feeling of being well informed and thus increases the quality of service (VDV, 2014) (c.f. Fig. 1).
Unfortunately until today, traveller information is not transmitted to transportation companies and their dispatchers in a
noteworthy amount nor in a way which permits an automatised processing of this information.
Thus, an beneficiary integration of automated customer feedback is missing until today. In extension to Stelzer et al.
(2014), in this paper, we show how to integrate and use customer feedback to improve service quality in public transport.
For this, we first present an overview of related work in Section 2. We then derive the need for automated information
exchange between the transportation company and the traveller in Section 3. In Section 4 we present technical constraints
and a novel framework to build systems which permit a broad exchange of customer information in a standardised way. In
Section 5 benefits of and possible application scenarios for automated standardised information exchange are pointed out.
We hereby not only focus on traveller benefits but also on models for transportation companies to improve their service
quality. The findings are discussed in Section 6 where we will also highlight the difference between the described approach
and the actual situation in public transport.
2. Related work
Currently, the disruption management cares mainly about timetable adjustment, rolling stock and crew rescheduling
(Jespersen-Groth et al., 2009). Customer information sometimes is indirectly integrated by staff reports and is then displayed
towards the dispatcher (Informationssystem Transportleitung Personenverkehr). The notification of customers in a timely
and consistent manner, is currently not a highly prioritized aspect in disruption recovery. Taking direct feedback of cus-
tomers into consideration is not part of the process at all. Suhl et al. (2001) defines a customer oriented dispatching by taking
the delays of travellers into consideration. However, the data is derived from models and simulation and not directly trans-
mitted by the traveller. Currently, the planning phase of disruption management also does not consider data of direct cus-
tomer feedback (Chu and Oetting, 2013).
In addition, research in the area of public transport and social media is rather focused on services to connect travellers
among each other or to provide entertainment for the traveller during the different phases of the trip (Foth et al., 2013).
Actual research concerning traveller feedback directly to the transportation company using social media is rare. Austin
(2011) mentions the problem of reacting to customer input, via social media, which in generally cannot be processed in
an automatised way. Deutsche Bahn, the main railway operator in Germany, uses social media to communicate with its cus-
tomers and also reacts to the feedback of the customers (Deutsche Bahn Facebook page, 2014). However, this feedback gen-
erally cannot be used to influence operations. Social media as well as traditional ways of communication do not allow a
transmission of structured information (Bregman, 2012). Therefore, the information cannot be easily evaluated and thus can-
not be used to enhance operations.
In another approach information derived from travellers is displayed in customer information systems (Gruber et al.,
2012). The information is not transmitted to and not used by traffic operations, though. Another community approach per-
mits passengers to inform other travellers about vehicle occupancy. Here as well, the information is solely provided in pas-
senger information systems (Crawford, 2013). An integration for operational improvement is not documented.
In Dollevoet et al. (2012) an approach to solve connection conflicts is presented, which postulates the knowledge of
origin–destination-pairs of travellers. Even today, this information is generally not available (Mayas et al., in press).
Therefore, a suitable approach is needed which enhances the service quality in public transport enabling customers to pro-
vide this information and the transportation companies to use the information.
The use of customer feedback and information, as part of e.g. public transport information, planning and dispatching sys-
tems, raises challenges on both sides. Based on the definition of service quality in EN 13816 (cf. Fig. 2), an approach to
enhance the service quality has to consider the traveller view and the service provider view as well (EN 13816, 2002).
Traveller aspects and provider aspects have to be equally analyzed and parameters for an increase of service quality have
to be identified. Therefore, we describe the different views on the necessity for automated customer feedback in the
following Section 3.
A. Stelzer et al. / Transportation Research Part E xxx (2015) xxx–xxx 3
Fig. 2. Service quality loop according to EN 13816 (2002).
In addition, service quality in public transport is influenced by several company and customer related factors (Fellesson
and Source, 2008; Morfoulaki et al., 2007). A good overview regarding important influence factors is given by Suhl et al.
(2001) who divided these factors into two categories:
hard influencing factors, e.g. waiting times at stations, waiting times in trains and missed connections,
soft influencing factors, e.g. tidiness of cars and station’s security, good service, friendly staff, comfort of journey and qual-
ity of catering.
While hard influencing factors are measurable by the dispatcher, the soft factors are not. Although several studies and
concepts, e.g. about the information chain in public transport (Daduna et al., 2006), and the satisfaction of public transporta-
tion travellers (Grigoroudis and Siskos, 2004), are known, the overall satisfaction with public transportation in Germany is
low (Grigoroudis and Siskos, 2004), compared to other transportation sectors. Therefore, an approach to improve customer
satisfaction by improving the service quality taking into consideration customer feedback is carved out in this paper.
3. Necessity for automated customer feedback
In this section the needs for automated customer feedback is derived from the transportation companies’ point of view
and from the travellers’ point of view as both actors are relevant for the improvement of service quality (cf. Fig. 2).
3.1. Transportation companies
The public transport sector is heavily influenced by the travellers’ needs. Vice versa, many operational decisions directly
influence the travellers. Regarding the service quality as defined in Section 2, a satisfaction of the customer can be achieved
by targeting and delivering the right level of service quality (cf. Fig. 2). Especially the ad hoc dispatching in case of interrup-
tions generally holds high potential for the improvement of service quality. Traditionally, the real time communication chan-
nel between the transportation company and the traveller is unidirectional, which allows the transportation company to
inform the traveller. Staff can be considered as the only exception here. Thus, direct feedback of the traveller is rarely used
to improve the operations of public transport. Without current knowledge about each traveller, it is difficult to integrate the
travellers’ requirements in operational decisions.
Today, transportation companies work with data derived from models, which use different types of models and input
data and approximate reality more or less. Possible data sources which are currently used are censuses, feedback of staff
in vehicles or stations and data derived from ticket sales. Censuses generally cover just a fractional amount, often less than
two percent, of the considered overall quantity, and – of course – are no real time information (Fornauf and Stelzer, 2011).
Data derived from ticket sales are – depending on the ticket type – more reliable and more up to date (Seaborn et al., 2009;
Sels et al., 2011). Still it is an approximation that works with assumptions based on the ticket type and the behaviour of the
traveller and it is not real time information. The most reliable data source until now is the feedback of staff in the vehicles
(Informationssystem Transportleitung Personenverkehr), as this covers real-time information. The information of the staff is
generally based on the ticket a traveller is using or conversation between the traveller and the staff. Still, necessary infor-
mation might not be exchanged during conversation or staff does not manage to visit all travellers. Also, information which
is not or only partially being forwarded by staff might get lost. Furthermore, many vehicles are not manned with staff which
could responsibly declare customer information to the dispatcher.
The above mentioned data is partially used for planning and dispatching, but possibilities are limited due to low data
quantity and quality. Also, due to the lack of real time data, quality measurements cannot be derived from this data.
A survey (Mayas et al., in press) with dispatchers in German transportation companies analysed the actual status of avail-
able real-time customer information, the attitudes, and the potential relevance of the variety of information from customers.
The results of the study lead to the following conclusions:
1. The majority of dispatchers, especially in regional transportation companies, does not have any actual data about cus-
tomers available. Most customer data is available in nationwide transport companies, for instance about the number
of travellers and their destinations. This information is mainly provided by messages from the staff in the vehicles and
existing anticipation or census data, but real-time changes in passenger flows are not registered.
2. Half of the dispatchers have rather positive attitudes towards the integration of real time information about customers
and direct customer feedback. However, 25% of the surveyed dispatchers associate real-time customer information with
rather negative opinions, which refer to a higher workload due to an expected information overload and more complex
dispatching systems. Therefore, the development of automated customer feedback has to integrate also a user-friendly
presentation of the information.
3. The real-time information about travellers is most relevant as an aggregation of interchangers per connection (cf. Fig. 3).
Furthermore, additional information as destinations and itineraries of travellers as well as special travel behaviour can
provide valuable information about possible travel alternatives. In contrast, direct customer feedback, for instance the
effective success of interchanges, are less relevant for dispatching decisions. Nevertheless, the reasons and consequences
of broken connections could provide valuable data in order to evaluate the quality of dispatching decisions in retrospect.
Finally, damage reports by customers are very relevant, especially for regional transportation companies, because regio-
nal dispatchers often also supervise the according infrastructure and expect an improvement of their decisions.
To improve data quantity as well as quality to such an extent that it can be used for dispatching, an automatised data
transfer from the traveller to the dispatcher needs to be established. The transferred data needs to be real-time, representing
the current travel desires of the customer, including data from which travel behaviour can be derived. This data transfer can
be established such that a broad data exchange becomes possible and the already used approximations become much more
precise. The challenge is to transfer as much data as possible relevant for the dispatcher which might be contradictory to the
travellers’ privacy concerns. The contradiction is discussed in Section 6.
An architecture which meets the above mentioned requirements is needed and is outlined in Section 4.
3.2. Travellers
From the travellers’ point of view, mobility is strongly connected to participation in daily life, viability, quality of living
and flexibility (Goddard, 2013). In addition, mobility behaviour is a result of the traveller’s phase of life and daily routines
(Wirtz and Jakobs, 2013), the traveller’s experiences and possibilities as well as the need and wish for combinations of exist-
ing and new means of transportation, e.g. intermodal and multimodal mobility concepts (Beckmann et al., 2006). But mobil-
ity behaviour is changing slowly and is mainly driven by phases of life (Generali Zukunftsfonds, 2013).
Compared to individual mobility concepts, e.g. car usage, public transport journeys consist of several phases and stages,
which themselves provide the base for uncertainty. Less experienced users often avoid public transport, as they feel the risks
of broken connections and other disruptions. Feeling in control of their own mobility is important for a majority of travellers.
This requirement can be reached, if the uncertainty along the journey is reduced by mobility information and users are
involved in the communication and information system.
Already when planning a journey along the travel chain in public transport, the question arises, if interchange times are
sufficient and connections can be reached. While travelling, even small delays can result in uncertainty as to whether the
journey can be conducted as planned or adjustments are necessary. Often it is hard for travellers to predict whether they
Fig. 3. Relevance of automated customer feedback (according to Mayas et al., in press).
may reach the desired connection in time or if an alternative route has to be found. In this case, travellers can only rely on
their experiences and the predicted information of dynamic information systems.
This example shows a typical situation about uncertainty in public transport. Extending existing information systems into
bidirectional communication systems, where information is provided by transport companies and travellers and feedback,
e.g. interchange wishes, can be communicated, can reduce uncertainty and as a result ease the change in mobility behaviour.
Previous studies and analysis in transportation and other application areas show that travellers want to give feedback for
different reasons (Sims et al., 2013; Thøgersen et al., 2009; Singh and Wilkes, 1996). Besides having the opportunity to com-
plain to someone about lack of quality, giving some sort of relief to the actual situation (Barow and Moller, 1996), travellers
often have a connection to their community, inspiring them to give feedback to enhance the community. FixMyCity
(Fixmycity, 2013) and FixMyTransport (Fixmytransport, 2013) are an example for such involvement. In addition to the trav-
ellers’ needs for participation by giving and getting feedback, the travellers’ requirements on mobility information are essen-
tial to enhance the quality of information.
High quality mobility information is a key success factor for using different kinds of mobility offerings. From providing
basic information about infrastructure, fares, timetables to real time information on delays and alternative routing in case
of disturbances, mobility information includes different kinds of information. Which information is suitable along the jour-
ney and will satisfy the information needs of the passengers is strongly connected to the different kinds of users (Mayas
et al., 2012), tasks, and systems as well as the environmental context (Hörold et al., 2013). The information needs can there-
fore be derived from the combination of these factors and reveal the needed information for different user groups, tasks and
stages along the journey (Hörold et al., 2012). Fulfilling the information needs of the travellers is a prerequisite for high qual-
ity mobility information. In addition, the flow of information along the journey, as well as the usability of interactive sys-
tems, are essential for the quality of mobility information (Brossard et al., 2011).
Along the information process, a study in German public transport revealed three basic travellers’ requirements on infor-
mation (VDV, 2014), allowing for a broader definition of mobility information quality:
Reliability: In a highly dynamic mobility context, the reliability of information describes the degree of accuracy between
the communicated information and the actual mobility situation.
Consistency: While mobility information is communicated through different information systems, the consistency of
mobility information focuses on similar communication, in regard to content and design.
Comprehensibility: The transparency about the origin of information and the ability of travellers to make their own deci-
sions can be defined as comprehensibility of information.
Fulfilling these requirements on information within the mobility context is challenging due to the mentioned character-
istics of mobility. In addition, all these quality factors, from information needs to comprehensibility, are related. In order to
satisfy these travellers’ needs and requirements, customer feedback can provide detailed information for transportation
companies. Technical systems, e.g. to monitor the actual situation, are required to build up an information base.
At the moment, customer feedback in public transport is often received through service hotlines, service personnel,
emails or social networks providing public transport companies with insight into typical questions and problems within
their network. Using this information in a real-time decision process is often not possible, as the data is provided in different
forms and often lacks essential information. In some cases, service personnel are able to forward the information in time and
report back to the travellers.
In order to improve the quality of customer feedback and mobility information according to the described travellers’
needs and requirements, it is worthwhile to open new ways for automated customer feedback with real-time information,
which is systematically structured, more reliable, consistent, and comprehensible.
4. Implementing automated customer feedback
In this section we present a way to transfer data between travellers and transportation companies. For this, we define
constraints in Section 4.1, then we present a novel architecture including interfaces in Section 4.2 and an example imple-
mentation in Section 4.3.
4.1. Technical constraints for automated customer feedback
Automated customer feedback needs to fulfil technical requirements and side constraints to be able to contribute to a
quality enhancement. First of all, a general architecture is needed, which permits an individual system design on the one
hand but system interoperability among the traveller and multiple transportation providers on the other hand.
Furthermore the system needs to be scalable such that transportation companies with a small amount of customers as well
as huge transportation companies can use the same data exchange mechanisms with the user.
The use of smartphones is an obvious choice for transmitting data between the transportation company and the traveller.
These devices are carried along with the traveller on her or his trip and can automatically send relevant information to the
transportation company and receive travel updates to be displayed to the traveller.
Secondly, the data exchange among multiple actors implies the need for standardised interfaces and standardised
exchange technology. The used communication technology can be derived from the current state of the art. The data trans-
mitted with the interfaces however, need to be designed for the public transportation domain. We introduce a standard
which defines relevant data structures for current and future needs in public transport based on state-of-the-art communi-
cation technology in the following subsection.
4.2. Technical framework for automated customer feedback
As more and more travellers use smartphone apps provided by the service operators to plan, book and monitor their jour-
ney, the customers smartphone can be used to collect customer feedback. The already existing IT infrastructure of trans-
portation companies can be extended in order to collect this important information individually from each traveller. We
now briefly present how those functionalities can be integrated in existing traveller information systems using the new
Traveller Realtime Information and Advisory Standard (TRIAS) (VDV, 2013a,b) which provides information services for the
customer together with fine granular monitoring capabilities for traveller flows.
TRIAS describes travel-related communication flows between the traveller and the service operator. As shown in Fig. 4,
this information flow between the transportation company and the traveller is bidirectional. A travel companion application
connects the smartphone with a Realtime Communications and Information Service (RTCS), which forwards relevant infor-
mation to the transportation company’s Intermodal Transport Control System (ITCS). In depth description of each compo-
nent’s responsibilities and functionalities can be found in Stelzer et al. (2013, 2014). However, relevant services for
collecting customer feedback are summarised in Table 1. Only the services of TRIAS which permit the traveller to commu-
nicate bidirectional with the transportation company are presented.
By the usage of systems which are compatible to this standard, collecting information on the destination, time and route
for travellers becomes possible. This information is useful to measure, e.g. traveller flows, breaking probabilities of inter-
changes or waiting times with a high accuracy.
The standard defines proactive services for requesting a connection demand. A traveller can use this service to express his
or her willingness to take a certain interchange in the near future. This so called Connection Demand Request Service is
intended to collect actual interchange wishes and thus helps to create statistics on the importance of different interchange
possibilities in case of service disruptions or delays in real-time such that dispatching decisions can be based on them.
In addition to that, using the Connection Report Service, the travel companion application can report whether an inter-
change was successful. This information is useful for monitoring the service quality. Additionally, it becomes possible for the
transportation company to collect information on the usage frequency of interchanges. Especially frequently used inter-
changes could be scheduled with minimal breaking probabilities to reduce the travel time for the travellers.
Besides that, TRIAS provides a Facility Status Service which can be used to report electronically feedback on damages of
facilities. Using this service, it is possible to reduce the reaction time in case of damages or malfunctions of facilities in sta-
tions or vehicles.
The Trip Monitoring Service is intended to give detailed information about the planned trip, including stopovers and
interchanges. It is possible to use this service to reroute travellers in case of disturbances, individualise travel information
or to improve traffic models.
The information that can be exchanged using the above mentioned services are briefly summarised in Table 1. Moreover
TRIAS contains very fine granular possibilities to control frequency, receiver and transmitted data of the data exchange. The
standard is a freely available specification developed for interfacing travel companion applications. Based on the above men-
tioned services, data fusion techniques can be applied to aggregate traveller flows. Those aggregations can be used to
Data Collecon Service Quality Management
Dispatcher
RTCS ITCS
Inter-
Travellers faces
Resources
Fig. 4. Symbolic architecture of a traveller information system.
Table 1
TRIAS services for exchanging individual travel related information.
Service Time frame Source Information

Connection report Past Customer Planned interchange broke unexpectedly
Trip monitoring Presence Transportation company Real time notifications on scheduled changes or delays
Trip monitoring Presence, future Customer Destination, route and planned time
Facility status Presence Customer Damage reports
Connection demand Future Customer Traveller requests an exchange
determine quality parameters for single exchanges or even whole routes. The aggregated traveller flows then can be used to
measure the travellers’ interest on certain interchanges. By processing routes from the Trip Monitoring Service or the
Connection Demand Service it even becomes possible to obtain real time information on traveller flows. The possibilities
gained by gathering data using TRIAS are outlined in Section 5.
4.3. Proof of concept
The Railway Operations Centre Darmstadt (Eisenbahnbetriebsfeld Darmstadt – EBD) is a research facility where the rail-
way environment is realistically simulated, including interlocking technology and dispatching systems for infrastructure
companies as well as transportation companies (Streitzig et al., 2012). The in Section 4.2 described interfaces have success-
fully been implemented in the EBD. Thus the feasibility of data exchange between the traveller’s smartphone and the trans-
portation company could be shown. Data derived from the smartphone could be integrated into the dispatching systems and
thus displayed towards the dispatcher.
The corresponding protocol was specified and tested for large scale deployments with heterogeneous mobile and
web-based clients. Its key design enables a wide applicability in current and future public transport information systems.
The general feasibility of the approach could be proven. Based on this, a field test needs to be carried out. As the standard
relies on state-of-the-art communication technology, an integration of such interfaces in information platforms of any trans-
portation company can be assumed to work without any difficulties. It can be assumed, that the TRIAS protocol will be sup-
ported by a rising number of vendors and service operators in Germany.
5. Using automated customer feedback
A typical journey with public transport provides several opportunities for customers to provide information. At the
moment, social networks allow for an unstructured way of collecting this information, mainly serving as a replacement
for other communication technology, e.g. service points and call centres. Especially mobile applications can provide a more
structured way for collection feedback information. The Munich linked transport system (MVV) for example uses the mobile
application MVV companion to allow customers to report disturbance information, representing a first step into bidirectional
communication and integration of customer feedback (MVV, 2014; Gruber et al., 2012).
From a customer’s point of view, the following areas of bidirectional communication and customer feedback could prove
to be equally useful for transport companies and customers:
planned routes,
interchange wishes,
disturbance and delay notifications,
damage reports, and
accuracy of routing and other information.
Providing feedback functions for customers may serve as a new and easy way to show that the customers’ well-being
along the journey matters to the transportation company. Feeling integrated as an individual can influence the quality per-
ceived, if the feedback is considered and an effect is recognised.
Transportation companies benefit from an increased service quality through new and more satisfied customers. The pro-
cess from communication to integration of customer feedback requires an adoption of background systems, e.g. dispatching
systems, as well as workflows of the service provider. Beyond the increase of the service quality and satisfaction of the cus-
tomer, this integration of customer feedback may result in better decision bases and performance of the public transport
system as well.
The introduction of customer feedback for dispatchers should be accompanied by special measures, in order to allay the
concerns (Mayas et al., in press) and guarantee a good ease of use for the dispatchers. This implies an effective, efficient, and
satisfying use (DIN, 2010) of the complex amount of information. Therefore the user-oriented automated processing and pre-
sentation of the customer feedback is required in reference to the workflow of the dispatcher.
In the following subsections, possible applications of customer feedback for transportation companies are presented. The
benefit of considering customer data is pointed out. The focus is on the dispatching, but also planning and quality manage-
ment are briefly discussed.
5.1. Customer feedback in dispatching
In case of disturbances, the traveller often has to deviate from her or his original travel plan. This is especially valid for
travellers with knowledge about the system, e.g. commuters, but also for most other travellers who gather information from
other sources. These ad hoc changes of travel plans, in case of disturbances, cannot be discovered by currently used data
sources, such as standard traffic models (e.g. Jovicic and Hansen, 2003) or data from ticket sales and cannot even be effi-
ciently registered by transportation company’s staff. Also, current traffic models based on assumptions are often under-
or overestimating the real demands (Flyvbjerg et al., 2005).
With information directly from the customers, very specific data about the current usage of public transport can be col-
lected. This permits individual care for each customer that – in case of conflicts – improves the personal travel experience (cf.
Section 3.2) but also allows to create or improve cost models to evaluate dispatching actions. The data can also be used to
improve or create models about traffic demands and the dispatcher can reflect her or his work by feedback reports about her
or his dispatching quality.
Related optimisation approaches for traveller flows (Suhl et al., 2001; Dollevoet et al., 2012) assume in situ knowledge of
each traveller’s individual destination. In the past, this knowledge was not available and thus those approaches could often
not be applied in public transportation systems. Using customer data, a transportation company is able to compute possible
costs of a connection breakage. A first input is the amount of interchangers derived from connection demands (cf. Table 1).
The usage of such numbers is not new, nevertheless, the way of retrieving it is (cf. Section 4.2).
Having reliable information about interchangers, a transportation company can put a value on a connection. A simple way
would be: using the amount of interchangers. However, using this approach, connections within highly frequented networks,
such as commuter transportation systems, where a traveller can easily catch the next vehicle, would be of equal value to
those on far distant trains, where a broken connection can lead to several hours of delay. This is why a more complex eval-
uation method is needed.
The travel chain of interchangers in not generally known. Nevertheless, also the travel chain can be transmitted by the
customer (cf. Table 1), yet it can be assumed that never all travellers will use this possibility. Thus, two cases need to be dis-
tinguished: (a) The travel chain is known and (b) the travel chain is unknown. In the first case, for each interchanger an alter-
native route can be calculated (Müller-Hannemann and Schnee, 2009) and the difference of waiting Dtw and travel times Dtt
compared to the planned trip according to the timetable can be detected. In the second case, an equivalent train has to be
found, for which we assume the interchanger will wait Dtw at the conflicted station. Although the perception of waiting time
is not linear (Suhl et al., 2001; Oetting and Rio, 2012), this is not considered in the following equation due to its high com-
plexity and is subject to further research.
n equals the number of interchangers for a specific connection. The variable ki :¼ 0; 1 defines if the travel chain is known
for customer i. The customer related total delay d to the connection can then be calculated as follows:
!
X
n X
n
d¼ ki ðDt wi þ Dt ti Þ þ n ki Dt w
i¼1 i¼1
A transportation company can hence compute the effect of a broken connection by summing up the specific and general
waiting times. A monetary evaluation can be obtained by applying a cost value for a delay. d multiplied with this cost value
then returns a monetary cost value for the connection conflict.
In addition to that, European laws grant a reimbursement for the traveller, if the total travel time is prolonged (European
Parliament and the Council, 2007). With a delay of 60 min 25% of the ticket price can be reimbursed, after 120 min 50%. Be pa
the average price of a ticket for travellers with unknown tickets. Be f the factor for calculating the potential reimbursement
cost.
8
< 0;
> if Dt w þ Dtt < 60 min
f ¼ 0:25; if 60min 6 Dt w þ Dt t < 120 min
>
:
0:5; if Dt w þ Dtt P 120 min
For travellers, whose tickets and their prices pi are known, the cost can be individually calculated, and a sum for the
unknown ticket prices can be added, estimating the delay by using an equivalent train as the reference. The customer related
potential cost regarding reimbursements c can be calculated as follows:
!
X
n X
n
c¼ ki f i pi þ n ki f pa
i¼1 i¼1
The dispatching process is highly complex and most of the time, a completely human decided process. In case of disrup-
tions or malfunctions, dispatching actions need to be undertaken to return to normal operation and to lead the traveller to
her or his destination. Most dispatching actions will generate costs that the dispatcher has to take into account while choos-
ing among her or his dispatching possibilities. These cost estimations have been purely gut instinct for a long time but now
become concrete with the help of customer feedback.
In the formulas above, customer feedback delivers the data on which the delays of travellers can be calculated. The use of
customer feedback gives the opportunity to decide more accurately on appropriate dispatching actions and also enables to
evaluate costs for them. Cost models as presented before can be used to get precise information about expected costs in the
current context of operations. A reduction of dispatching costs can therefore be achieved by choosing among cheaper appro-
priate dispatching measures and not choosing too expensive ones at all.
Further more, being able to compute costs for connection conflicts and conflict solutions is a first step to a (partial)
automation of connection dispatching. Simple standard tasks can be automated or a list of possible solutions can be auto-
matically provided to the dispatcher, from which she/he can choose an appropriate one.
Generally all measures in connection dispatching need information about the interchangers. The collected data can thus
be the basis for more sophisticated models as for example presented in Kurby (2012).
5.2. Customer feedback on malfunctions
The feedback of customers can also be used in areas of ad hoc dispatching that is related indirectly to the customer. The
customer can give feedback about problems and damage of vehicles or the infrastructure (cf. Table 1). Damages that affect
more travellers will be reported more often than those affect a little amount of travellers. On the amount of feedback mes-
sages for a specific component or vehicle, the dispatcher can prioritize the repair of affected parts. Again, the feedback alone
is not a sufficient value to prioritize. The criticality of damage also has to be taken into consideration. This can be done by
assigning factors ad to a damage type. A prioritization for maintenance can be easily calculated, as follows, where i is the
amount of identical reports:
p ¼ i ad
The feedback of customers can thus be used in addition to reports of staff. The customer will create reports using stan-
dardised interfaces which can be automatically processed. With the help of customers, damages and malfunctions can be
detected faster than by staff only. Approaches of customer integration today exist in a way that for example numbers are
provided on ticket machines, which can be called in case of malfunction (MVV, xxxx). Using standardised interfaces
improves the quality of customer feedback and also reduces the barrier of reporting problems, assuming proper and
easy-to-use applications are available.
The reported damages can not only be repaired faster, they also can be used in classical transportation dispatching.
Knowing the traveller’s profile, malfunctions, which affect the traveller’s journey, can be considered. A person who cannot
walk stairs, for example, would be highly affected by a disturbed elevator and could be rerouted on a malfunction report.
5.3. Customer feedback for quality measurement of dispatching actions
So far, the dispatcher has no structured knowledge about the effectiveness of her or his dispatching measures. The dis-
patching decision is mainly based on operational processes, corporate regulations and the dispatcher’s experience from the
past (cf. Section 3.1). With the help of customer feedback, the dispatcher can be informed about the effects of her or his dis-
patching measures. The approach to ask the customer for feedback about a dispatching action that has already been taken is
very new. Of course, the feedback has no effect for the dispatching action already taken, but can be used for the dispatcher, to
reflect her or his action and to measure the quality. Thus, the dispatcher can learn about the effectiveness of her or his
actions and improve them in the future. As the connection dispatching is one of the dispatching fields with the most impact
on travellers, and the public perception in this area is very high, the connection dispatching is chosen as an example to high-
light the benefit of post dispatching customer feedback.
With the help of customer feedback about whether an interchange could be reached and used by a traveller (cf. Table 1),
the transportation company can draw conclusions about the effectiveness of the connection dispatching. For example, a ratio
r can be determined of the amount of travellers s who reach their connection and those who do not ( f ) (for the same
connection):
f
r¼
f þs
A good value of r has to be evaluated in a real environment, taking into account, that more travellers might report a failed
connection out of anger, than those who reach it, as negative feedback is more likely formulated (Thøgersen et al., 2009).
The feedback can apply to planned or unplanned connections. Thus it is worthy for ad hoc dispatched interchanges as
well. Furthermore, the feedback in this context can be used to improve the planning of connections, as it delivers information
about the usage and success for planned connections. Waiting times can be adjusted, unused connections can be withdrawn,
and even new (unplanned) connections can be discovered if travellers give frequent feedback.
Also, many transportation companies or authorities define quality specifications. Until today, evaluation exists in forms of
questionnaires. A usage of structured electronic data directly sent back from the customer is not yet known.
Customer feedback can be used to measure the quality and to evaluate if the operations meet the quality specifications.
Based on the input derived from the customer, the quality specifications can be refined and extended. More sophisticated
analysis is possible. Hence, customer feedback is helpful to evaluate quality standards.
5.4. Customer feedback in planning
Also the planning departments of transportation companies can benefit from customer feedback. Exemplarily, the relia-
bility of travel chains has been chosen to point out the support of customer feedback for planning.
Derived from the information about connection success in the preceding section combined with travel information about
the journey of the customer (cf. Table 1), a transportation company can draw conclusions about the reliability of a travel
chain as well. Assuming a trip starts from a to destination b. Be qbreak;n the probability of a connection breakage for connec-
tion cn The function exðcn Þ will return the probability of a connection breakage:
qbreak;n ¼ exðcn Þ
In a practical environment, many different functions exðcn Þ exist, to consider the different environments a connection can
take place in. For this paper, we simply define the function exðcn Þ using the information about interchange success, reported
by the traveller (cf. Section 5.3):
f
exðcn Þ ¼
sþf
Taking into account that negative feedback is more likely to be reported (Thøgersen et al., 2009), a factor m; 0 < m < 1 can be
introduced to attenuate the negative feedback:
mf
exðcn Þ ¼
s þ mf
m needs to be adjusted empirically. Assuming a travel chain with N interchanges, the probability of a break in the travel
chain qbreak can thus be computed by:
Y
N Y
N
qbreak ¼ qbreak;i ¼ exðci Þ
i¼1 i¼1
In the planning phase the transportation company or authority develops an offer of public transport possibilities, respect-
ing travel demands, based on traffic models. With the help of customer feedback an offer can be adapted to come closer to
the travellers’ needs.
6. Discussion
The idea of interacting electronically with the customer is not new. However, existing approaches are mainly focused on
social media, e.g. (Austin, 2011; Deutsche Bahn Facebook page, 2014; Bregman, 2012). While these approaches provide a
base for further developments, they are mainly serving as another communication channel. Partially, information relevant
for the operation is posted through social media channels. But the information is generally unstructured and cannot be auto-
matically processed to be used for improving operations. Furthermore, the communication partner is generally not the dis-
patcher but a special social media team which is specialised in interacting with the customer through social media channels.
Missing customer feedback makes it very difficult to estimate traveller flows as only general models apply. Only a small
quantity of real-time data is available – if any – on which it is hard to built reliable dispatching support tools. Therefore,
nowadays it is not easy to apply the right dispatching measure and the degree of automation is very low. The applied mea-
sures without the integration of customer feedback are often suboptimal. Apart from this, only general travel information
can be communicated towards the traveller.
This makes it impossible to improve dispatching processes. In this paper we focused on an approach to use customer
feedback to improve the travel experience for customers through better information and improved dispatching quality. In
the following we will discuss the concepts and results provided by using the presented novel approach of customer feedback
(cf. Section 4) and the usage of customer data gathered in this way (cf. Section 5).
The EN 13816 defines quality in regard to the mobility offer, the quality of service and the environmental impact (EN
13816, 2002). In regard to the integration of bidirectional communication, especially the quality criteria information and
comfort are affected (EN 13816, 2002) and can benefit from bidirectional communication and feedback. Introducing a feed-
back function, which e.g. allows customers to communicate their actual or future journey, result in a better information ser-
vice, providing information for the individual journey. In addition, being able to have some sort of control and the
opportunity to communicate feedback, which is e.g. integrated into the dispatching process, will result in earlier and better
information and thereby shorten the time the passenger is worrying about his or her journey, leaving her or him more time
to relax. This results in a better experience and perceived quality as well.
Thompson and Schofield (2007) show that satisfaction essentially depends on the ease of use. Thus, the travel companion
app should be able to recognise the traveller’s situation with minimal user input. Related research (Baumann, 2012) indi-
cates, that this is already possible in many cases. However, as each traveller can act autonomously, in certain situations man-
ual user input might be beneficial. But usability is not only important for the travel companion app. As dispatchers will have
to handle more information, better data-mining algorithms together with elaborated visualisation approaches are necessary
in order to avoid distraction and overload.
The survey among dispatchers (Mayas et al., in press) has shown a general need for customer feedback. Nevertheless, the
dispatcher must not be overwhelmed with additional information, but needs a seamless integration into dispatching sys-
tems as well as proper trainings. Also, differences in the appraisal of customer information between regional and nationwide
transportation companies have been presented. Therefore, the integration of customer information and feedback needs to be
adapted to the context of use.
A general approach to integrate customer information to be used in dispatching is the enrichment of traffic models. Traffic
models are generally on a macroscopic level and used to derive general travel demands. Some of them are exact enough to
map timetables. None of them can represent the change of travel behaviour in case of disruption, though. Especially in local
transport, source data to create and refine these models is rare. With a high quantity of customer feedback data, existing
models can be enhanced by integrating detailed travel information. This is a major benefit in cases where only little infor-
mation, such as censuses, is available, i.e. in local transport.
Usage of customer feedback provides real time knowledge regarding the destination of travellers, which permits the build-
ing of real time traffic models as an alternative to simulation. They could be able to show instantly how disruptions change
traveller flows. Therefore, a critical amount of travellers participating has to be achieved to provide the necessary data. Field
applications have to show, what number is sufficient for different situations. Real time traffic models can provide information
about the number of travellers and their travel destinations including interchange behaviour to the dispatcher. The dispatcher
then can base his or her dispatching actions on this information and will presumably improve her or his decisions.
Together with knowledge about the travel plans, the dispatcher can individually support the traveller on her or his devi-
ation or even propose (better) alternatives. An automatic evaluation of the conflicts and conflict resolutions (Oetting and
Stelzer, 2015) can be based on such customer feedback as well (cf. Section 5.1).
While structured customer feedback is useful for enhancing the quality of public transportation services, it obviously
causes privacy concerns. By collecting individual travel times and destinations, the transportation company becomes capable
of compiling personalised movement profiles, monitoring mobility patterns or even predicting future locations of a traveller
(Baumann, 2014). These concerns have to be addressed according to national data protection regulations.
It has been shown in Section 3.2 that the customer is willing to share personal data if she or he can benefit from it by
means of better dispatching results for his or her personal journey. However, to enhance the service quality, the transporta-
tion company needs to store and process personalised data in a way that respect the travellers’ privacy concerns. In general,
feedback services should collect the bare minimum of data which is required enhance organizational processes. The collected
data should not be used for purposes which provide no benefits to the traveller. Additionally, it is necessary to inform the
traveller on the amount and purpose and information content of collected data as well as to provide means for controlling
the granularity of the collected data. Anonymized, pseudonymous and earmarked data collection would be desirable and a
step into the direction of privacy protection.
Altogether, customer feedback holds enormous potential for improving the dispatching and service quality in public
transport, as long as legitimate privacy concerns are addressed and solved by proper and transparent techniques of data
selection or anonymization.
7. Conclusion
While technological advances, e.g. mobile information systems and communications, become more and more present and
customers tend to share more and more information with each other, transportation companies should evaluate how they
can benefit from these developments in order to enhance their information systems and their mobility offerings. The dis-
cussed approach provides this opportunity by using established communication technologies to enhance the service quality
of public transport in general. In order to keep current public transport customers and acquire new customers, public trans-
port companies have to adapt their information systems to new technologies and new developments in the daily lives of
their customers. Satisfaction with public transport services, the travel experience and information quality will be key success
factors in this regard in the near future.
An integration of the traveller information in the decision process simplifies the collection of long term, fine grained trav-
eller flows and enables novel analysis, e.g. for finding frequently breaking connections or unnecessary waiting times. Due to
these reasons, a deep integration of traveller aspects in the connection planning and dispatching process is a worthwhile
optimisation for the transportation companies as well as for the travellers.
The information exchange presented in this paper does not reduce the probability of negative events, but it
certainly reduces their impact. Augmenting the existing operational dispatching processes with real time information about
the traveller flows enables better dispatching decisions, enhanced planning capabilities and an improved quality
management.
Further work has to cover the implementation of the presented interfaces to gather customer feedback in high quantities
from realistic environments. Based upon that, the proposed application models can be verified. Also, the change of perceived
service quality due to the implementation and application of customer data needs to be evaluated.
In the future, systems need to be developed which use standards as presented, which integrate and process gathered cus-
tomer information such that the travellers can benefit from improved customer oriented operations and more individual tra-
vel information.
References
Austin, J., 2011. Use of Twitter and Facebook by public transport operators and authorities in a time of crisis. In: Presented at IIID Traffic&Transport 2011:
Traffic, Transport and Social Media (09).
Barow, J., Moller, C., 1996. A Complaint is a Gift – Using Customer Feedback as a Strategic Tool, first ed. Berrett-Koehler Publishers Inc., San Francisco.
Baumann, P., 2012. Abstract: user context recognition for navigation systems in public transportation. In: Workshop Proceedings of the 10th Annual IEEE
International Conference on Pervasive Computing and Communications (PERCOM). IEEE Computer Society, pp. 558–559.
Baumann, P., 2014. Adaptive sensor cooperation for predicting human mobility. In: Proceedings of the 2014 ACM International Joint Conference on
Pervasive and Ubiquitous Computing: Adjunct Publication. ACM, pp. 349–354.
Beckmann, K.J., Chlond, B., Kuhnimhof, T., Ruhren, S.V.D., Zumkeller, D., 2006. Mobilität – everyday users of multimodal means of transport, Internationales
Verkehrswesen. Transp. Mobility Manage. 58 (4), 138–145.
Bregman, S., 2012. Uses of social media in public transportation. Transport. Res. Board, 99.
Brossard, A., Abed, M., Kolski, C., Uster, G., 2011. Consideration of the travel time experience in the conceptual models of personalized interactive
applications. In: Kolski, C. (Ed.), Human–Computer Interactions in Transport. ISTE Ltd and John Wiley & Sons Inc., pp. 283–310.
Cambridge Systematics, 1999. Morpace International, National Research Council (U.S.) Transportation Research Board, Transit Cooperative Research
Program, Transit Development Corporation, A Handbook for Measuring Customer Satisfaction and Service Quality, vol. 47. National Academy Press,
Washington, DC <http://onlinepubs.trb.org/onlinepubs/tcrp/tcrp_rpt_47-a.pdf>.
Chu, F., Oetting, A., 2013. Modeling capacity consumption considering disruption program characteristics and the transition phase to steady operations
during disruptions. J. Rail Transp. Plan. Manage. 3 (3), 54–67, Robust Rescheduling and Capacity Use, http://dx.doi.org/10.1016/j.jrtpm.2013.10.006
<http://www.sciencedirect.com/science/article/pii/S221097061300022X>.
Crawford, D., 2013. Making transit more passenger friendly. ITS Int. 19 (3), 28–29, this article was published online with the title: Apps Help Passengers
Avoided Overcrowded Public Transport <http://www.itsinternational.com/sections/transmart/features/apps-help-passengers-avoided-overcrowded-
npublic-transport/>.
Daduna, J.R., Schneidereit, G., Voß, S., 2006. Dynamische Fahrgastinformation im Öffentlichen Personennahverkehr - Grundstrukturen und
kundenorientierte Funktionalitätsanforderungen. In: Mattfeld, D., Suhl, L. (Eds.), Informationssysteme in Transport und Verkehr. DSOR Beiträge zur
Wirtschaftsinformatik, DS&OR Lab, Paderborn, pp. 47–69.
Deutsche Bahn Facebook page, 2014. DB Bahn <https://www.facebook.com/dbbahn> (accessed 29.10.14).
DIN EN ISO 9241-210, 2010. Ergonomics of human system interaction-Part 210: human-centred design for interactive systems.
Dollevoet, T., Huisman, D., Schmidt, M., Schöbel, A., 2012. Delay management with rerouting of passengers. Transport. Sci. 46 (1), 74–89.
EN 13816, 2002. Transportation – Logistics and Services – Public Passenger Transport – Service Quality Definition, Targeting and Measurement (04).
European Parliament and the Council, 2007. Regulation (EC) No 1371/2007 of the European Parliament and of the Council of 23 October 2007 on rail
passengers’ rights and obligations. Off. J. Eur. Union 50 (L 315),14–41 <http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=OJ:L:2007:315:TOC>.
Fellesson, M., Source, M.F., 2008. Perceived satisfaction with public transport service in nine european cities. J. Transport. Res. Forum 47 (3 (Public Transit
Special Issue 2008)), 93–103. http://dx.doi.org/10.5399/osu/jtrf.47.3.2126.
Fixmycity web page, 2013. <http://www.fixmycity.de/> (accessed 10.12.13).
Fixmytransport web page, 2013. <http://www.fixmytransport.com/> (accessed 10.12.13).
Flyvbjerg, B., Skamris Holm, M.K., Buhl, S.L., 2005. How (in) accurate are demand forecasts in public works projects?: The case of transportation. J. Am. Plan.
Assoc. 71 (2), 131–146.
Fornauf, L., Stelzer, A., 2011. Überblick über das Reisendenerfassungssystem (interview with Matthias Hartmann 24.03.11).
Foth, M., Schroeter, R., Ti, J., 2013. Opportunities of public transport experience enhancements with mobile services and urban screens. Int. J. Amb. Comput.
Intell. 5 (1), 1–18.
Generali Zukunftsfonds (Ed.), 2013. Wie ältere Menschen leben, denken und sich engagieren. FISCHER Taschenbuch, Frankfurt a. M.
Goddard, S., 2013. What kind of city do you want? In: Presented at Praxisforum Verkehrsforschung 2013 (11 2013) <http://www.infas.de/themen/
praxisforum>.
Grigoroudis, E., Siskos, Y., 2004. A survey of customer satisfaction barometers: some results from the transportation-communications sector. Euro. J. Operat.
Res. 152 (2), 334–353.
Gruber, G., Mühl, A., Zöllner, R., 2012. Neuer Community-Ansatz: Fahrgäste informieren Fahrgäste-Der MVV-Staumelder für die elektronische
Fahrplanauskunft. Nahverkehr 30 (10), 28–30.
Hörold, S., Mayas, C., Krömker, H., 2012. Identifying the information needs of users in public transport. In: Advances in Human Aspects of Road and Rail
Transportation. Advances in Human Factors and Ergonomics Series. CRC Press, pp. 331–340, http://dx.doi.org/10.1201/b12320-39, http://dx.doi.org/10.
1201/b12320-39.
Hörold, S., Mayas, C., Krömker, H., 2013. User-oriented development of information systems in public transport. In: Anderson, M. (Ed.), Contemporary
Ergonomics and Human Factors 2013. CRC Press, pp. 160–167.
Informationssystem Transportleitung Personenverkehr: Benutzerhandbuch (BHB-ISTP), Frankfurt/Main.
Jespersen-Groth, J., Potthoff, D., Clausen, J., Huisman, D., Kroon, L., Maróti, G., Nielsen, M.N., 2009. Disruption Management in Passenger Railway
Transportation. Springer.
Jovicic, G., Hansen, C.O., 2003. A passenger travel demand model for copenhagen. Transport. Res. Part A: Policy Pract. 37 (4), 333–349.
Kurby, S., 2012. Makroskopisches Echtzeitdispositionsmodell zur Lösung von Anschlusskonflikten im Eisenbahnbetrieb, Ph.D. thesis. Technische Universität
Dresden, Dresden
Mayas, C., Hörold, S., Krömker, H., 2012. Meeting the challenges of individual passenger information with personas. In: Stanton, N.A. (Ed.), Advances in
Human Aspects of Road and Rail Transportation, Advances in Human Factors and Ergonomics Series. CRC Press, pp. 822–831, http://dx.doi.org/10.1201/
b12320-92, http://dx.doi.org/10.1201/b12320-92.
Mayas, C., Hörold, S., Stelzer, A., Englert, F., Krömker, H., 2015. Evaluation of dispatcher requirements on automated customer feedback in public transport.
In: 13th International Conference on Human–Computer Interaction – INTERACT 2015. Bamberg, Germany (in press).
Morfoulaki, M., Tyrinopoulos, Y., Aifadopoulou, G., 2007. Estimation of satisfied customers in public transport systems: a new methodological approach. J.
Transport. Res. Forum 46 (1 (Spring 2007)), 63–72. http://dx.doi.org/10.5399/osu/jtrf.46.1.981.
Müller-Hannemann, M., Schnee, M., 2009. Efficient timetable information in the presence of delays. In: Ahuja, R.K., Möhring, R.H., Zaroliagis, C.D. (Eds.),
Robust and Online Large-Scale Optimization, Lecture Notes in Computer Science. 0302-9743, vol. 5868. Springer-Verlag, Berlin Heidelberg, pp. 249–272.
Munich linked transport system (MVV), 2014. MVV-Companion – the journey planner as an app for smartphones <http://www.mvv-muenchen.de/en/
journey-planner/mobile-services/efa-app-mvv-companion/index.html> (accessed 10.12.14).
Münchner Verkehrs- und Tarifverbund. Ticket machine malfunctions <http://www.mvv-muenchen.de/en/tickets-fares/ticket-machines/ticket-machine-
malfunctions/index.html> (accessed 23.03.15).
Oetting, A., Rio, A., 2012 Evaluation of supply quality in passenger transport as a basis for the assessment of railway infrastructure measures. In: Universität
Stuttgart (Ed.), International Symposium on Networks for Mobility, vol. 6. Universität Stuttgart.
Oetting, A., Stelzer, A., 2015. Conflict resolution in connection dispatching. In: Hansen, I.A., Tomii, N., Hirai, C. (Eds.), 6th International Conference on Railway
Operations Modelling and Analysis, pp. 024-1–024-19.
Seaborn, C., Attanucci, J., Wilson, Nigel H.M., 2009. Analyzing multimodal public transport journeys in London with smart card fare payment data. Transport.
Res. Rec.: J. Transport. Res. Board 2121 (Volume 2121/2009 Data Systems and Travel Survey Methods 2009), 55–62, http://dx.doi.org/10.3141/2121-06
<http://trb.metapress.com/content/j1523k7t70710617/>.
Sels, P., Dewilde, T., Cattrysse, D., Vansteenwegen, P., 2011. Deriving all passenger flows in a railway network from ticket sales data. In: International
Association of Railway Operations Research (Ed.), 4th International Conference on Railway Operations Modelling and Analysis RailRome2011.
International Association of Railway Operations Research.
Sims, R.E., Ross, T., May, A.J., 2013. Crowd sourcing of public transport problems. In: Anderson, M. (Ed.), Contemporary Ergonomics and Human Factors 2013.
CRC Press, pp. 176–188.
Singh, J., Wilkes, R.E., 1996. When consumers complain: a path analysis of the key antecedents of consumer complaint response estimates. J. Acad. Market.
Sci. 24 (4), 350–365.
Stelzer, A., Englert, F., Oetting, A., Steinmetz, R., 2013. Information exchange for connection dispatching, in: Euro – Žel 2013: Symposium Proceedings.
University Žilina, pp. 222–230.
Stelzer, A., Englert, F., Hörold, S., Mayas, C., 2014. Using customer feedback in public transportation systems. In: IEEE (Ed.), International Conference on
Advanced Logistics and Transport (ICALT) 2014, Hammamet, Tunisia, pp. 29–34, http://dx.doi.org/10.1109/ICAdLT.2014.6864077.
Streitzig, C., Stelzer, A., Schön, S., Chu, F., 2012. TU Darmstadt – Research, Training & More Besides, EURAILmag 2012 Second Half Year (Issue 26), 152–159.
Suhl, L., Biederbick, C., Kliewer, N., 2001. Design of customer-oriented dispatching support for railways. In: Voß, S., Daduna, J.R. (Eds.), Computer-Aided
Scheduling of Public Transport. Springer, pp. 365–386.
Thøgersen, J., Juhl, H.J., Poulsen, C.S., 2009. Complaining: a function of attitude, personality, and situation. Psychol. Market. 26 (8), 760–777.
Thompson, K., Schofield, P., 2007. An investigation of the relationship between public transport performance and destination satisfaction. J. Transp. Geogr.
15 (2), 136–144.
VDV 431-1, 2013a. Echtzeitkommunikations- und Auskunftsplattform EKAP Teil 1: Systemarchitektur.
VDV 431-2, 2013b. Echtzeitkommunikations- und Auskunftsplattform EKAP Teil 2: EKAP-Schnittstellenbeschreibung.
VDV 7035, 2014. Nutzerorientierte Gestaltungsprinzipien für mobile Fahrgastinformation.
Wirtz, S., Jakobs, E.-M., 2013. Improving user experience for passenger information systems. Prototypes and reference objects. IEEE Trans. Prof. Commun. 56
(2), 120–137.

Transportation Research Part E: Anselmo Stelzer, Frank Englert, Stephan Hörold, Cindy Mayas

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Transportation Research Part E: Anselmo Stelzer, Frank Englert, Stephan Hörold, Cindy Mayas

Uploaded by

Copyright:

Available Formats

Transportation Research Part E xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Transportation Research Part E

Improving service quality in public transportation systems

Transportaon Enhanced Travel

Beneﬁts Realme Informaon Benefits:

Fig. 2. Service quality loop according to EN 13816 (2002).

3. Necessity for automated customer feedback

3.1. Transportation companies

Fig. 3. Relevance of automated customer feedback (according to Mayas et al., in press).

4. Implementing automated customer feedback

4.1. Technical constraints for automated customer feedback

4.2. Technical framework for automated customer feedback

Data Collecon Service Quality Management

Service Time frame Source Information

4.3. Proof of concept

5. Using automated customer feedback

5.1. Customer feedback in dispatching

5.2. Customer feedback on malfunctions

5.3. Customer feedback for quality measurement of dispatching actions

5.4. Customer feedback in planning

You might also like