Expert evaluation of criteria describing the quality of travelling by international passenger train: technological, economic and safety perspectives.
Maskeliunaite, Lijana ; Sivilevicius, Henrikas
Reference to this paper should be made as follows: Maskeliunaite,
L.; Sivilevicius, H. 2012. Expert evaluation of criteria describing the
quality of travelling by international passenger train: technological,
economic and safety perspectives, Technological and Economic Development
of Economy 18(3): 544-566.
JEL Classification: C02, R4, L62, L92.
1. Introduction
In recent years, railway transport competing in the transportation
market against other means of transport has been rapidly improving. The
speed and level of the comfort and safety of the trip have been steadily
growing. The prospects of railway transport give hopes to transport
enterprises because the majority of countries give priority to the
development of passenger and freight rail transport. The stability of
the railway transportation market is based on the fact that the problems
of traffic jams in cities, the need for high-speed transportation and
highways blocked up with heavy trucks etc. still remain unsolved. Such
countries as Algeria, Argentina, Israel, India and the region of the
Persian Gulf make much effort to develop rail transport. The USA
possessing highly developed automobile and air transport have changed
their transportation policy and are now developing high-speed railway
systems (Schwieterman, Scheidt 2007). Russia is planning to renew their
rolling stock and to purchase 20 thousand locomotives within the
following 20 years. In 2005, great technological achievements in the
central Japanese railway company helped with increasing profit by
introducing superconductive magnetic systems in rail transport. The
system allowing for developing a speed of 450-500 km/h was first tested
in May 1998 (Nakagawa, Matsuda 2005). Now, Japanese bullet trains
Shinkansen, presented to the public in 1964, use innovative technologies
based on electronics (Hagiwara et al. 2007). China has also contributed
to the development of high-speed railways by building a gigantic system.
When testing the system on the route Beijing-Shanghai, Chinese-made
locomotive CRH380A, pulling a passenger train reached a speed of 486
km/h on 5 December 2010. The new high-speed railway route
Beijing-Shanghai should be opened in 2012.
The main problem of rail transportation in a new age is to ensure
the safe use of railway infrastructure and to meet the specified safety
requirements (Rao, Tsai 2007). Japan's Railway Technical Research
Institute is developing systems capable of detecting train position,
e.g. in the tunnel or tilt of the railway car body (Sasaki 2005). In
North America, simulation software and methods of statistical analysis
are used for determining the capacity and operational characteristics of
railway infrastructure (White 2005). Great attention is paid to the
development of modern technologies promoting the use of wireless mobile
equipment and networks in rail transport (Fitzmaurice 2005). Accidents
with modern trains demonstrate the necessity to develop more effective
systems for detecting rail defects (Scalea et al. 2005). The models are
designed to identify breakdowns in the rails and to control the risk of
railway accidents (Zhao et al. 2007). The rails of the railway track are
tested under heavy loading (Li, Bilow 2008). The dynamics of the railway
vehicle (Lei, Zhang 2011) of the three-dimensional interaction between
the bridge and a high-speed train using a wheel-rail contact model (Dinh
et al. 2009) of compressive stress induced by passing trains in
permafrost subgrade along Qinghai-Tibet Railway (Zhu et al. 2011) and
structural dynamics have been considered for railway transport systems
(Stribersky et al. 2000). The formula for determining the dynamic
coefficient that may be used as a basis for designing and evaluating
bridges belonging to the urban railway system has been offered by (He et
al. 2011). Works on the development and maintenance of transport system
infrastructure present some risk to workers. The problem of safety in
this field is considered to be of primary importance by the
environmental institutions and universities of the United Kingdom. The
elder (less efficient) and underqualified workers are recommended to
retire or change work. The flow of workers from other countries also
poses some problems due to their poor knowledge of the native language,
training and getting and conveying information. Risks are based on the
level of workers' qualification, knowledge and experience (Campbell
et al. 2007). Roads and railroads in Italy are difficult for travelling
because of a number of tunnels on the routes. Now, the total length of
Italian railway routes makes 16000 km and includes 2000 tunnels with the
total length of 1400 km. Therefore, fire protection should be ensured on
trains. Risk analysis of possible fire on the train passing through the
tunnel allowed researchers (Martinelli et al. 2008) making the following
conclusions: the height of the curve of the fire model presented in
Italian standards is considerably reduced and is lower than that of the
testing curve; when the train is full of passengers, a clear and well
thought-out evacuation plan should be prepared. The rescue of passengers
largely depends on their conscious behaviour and quick reaction time.
Therefore, fire detection and alarm systems in passenger cars play a
critical role. Gasparik and Zitricky (2010) propose a new approach to
evaluating the capacity consumption of a track line (occupation time)
based on the graphic approach.
Now, large networks of high-speed railways operate in Europe and
Japan. However, in Japan, they compete with air transport while in
Europe they supplement each other (Clever, Hansen 2008). The cooperation
of trains in a multimodal international railway transport system is
considered along with the plan of extending an intermodal network to
embrace eleven countries from Scandinavia to Greece via Bulgaria, the
Czech Republic, Hungary, Poland, Romania and Slovakia (Kuo et al. 2008).
The development of European Rail Traffic Management Systems (ERTMS)
requires appropriate methods of modelling (simulation) (Jabri et al.
2010). To attract more passengers, transport services should be
improved: in addition to high-quality rolling stock, highly qualified
staff should be trained. The problems of theoretical knowledge and
practical skills of employees have been in the focus of researchers of
various fields in the last decades. In light of rail transport, training
programmes of managing staff (train crew) are developed and practically
used (Morgan et al. 2007). Maintenance costs of freight locomotives (of
particular series) are analysed and assessed thus recommending the ways
of reducing them (Bureika 2011). Research into the fault rate of railway
trains is carried out (Gelumbickas, Vai?mnas 2011) and physical
characteristics of metal used for the wheelset tyre are modelled
(Bazaras, Somov 2011). Mathematical models of making-up trains allowing
for optimizing them on each route and a type of traction are developed
(Dailydka 2010; Ramunas et al. 2011) determining criteria for evaluating
passenger transportation by rail and their significance (Maskeliunaite
et al. 2009; Sivilevicius, Maskeliunaite 2010; Maskeliunaite,
Sivilevicius 2011; Si et al. 2009; Preston et al. 2009).
The present paper is aimed at ensuring (employing expert methods)
the consistency of judgements made by the respondents (passengers),
experts (service staff of the train 'Vilnius-Moscow' and
administrative staff of the Passenger Transportation Directorate of the
joint-stock company Lietuvos gelezinkeliai) considering the weight
(significance) of criterion groups A, B, C and D describing the quality
of passenger transportation. The obtained data are required for creating
an additional model.
2. Description activities undertaken by the joint-stock company
Lietuvos gelezinkeliai and criteria explaining the quality of passenger
transportation
Passenger transportation is a specific area of social and economic
importance associated with the main problem of the state--the provision
of the freedom of travel (movement) (Belov et al. 2001).
Social expenses on railway transport consist of the maintenance
costs of railways (infrastructure), the costs of renewing and
maintaining rolling stock (repairing old and purchasing new locomotives)
(Fig. 1) and expenses on the passengers of trips. The costs of passenger
transportation services may differ depending on their quality.
An increase in the quality of rolling stock and railways decreases
the costs of travelling. Due to an increase in speed, travel time and a
chance to be involved in a railway traffic accident (avoid medical and
insurance costs) reduces.
The costs of railway maintenance and development are related to its
construction, repair and maintenance. Expenses on stock renewal,
purchasing, repair and maintenance, including the costs of fuel,
electric power, spare parts, oil, workers' payment, etc. make the
major part of these costs.
The expenses of rail transport users are related to fare losses
experienced due to delays to trips, traffic accidents, environment
pollution and a lack of comfort during the trip.
By increasing the costs of railway staff to some extent, the total
expenses of passengers and society can be decreased. When the variation
curves of railway staff and passengers cross each other, social expenses
are the lowest, i.e. optimal quality level is achieved. A lack of
investment in infrastructure and rolling stock results in a low quality
of a railway travel or makes this transport facility unattractive to
passengers.
[FIGURE 1 OMITTTED]
To attract more passengers, the quality of transportation should be
improved: in addition to high-quality rolling stock, highly qualified
staff should be trained. Therefore, for several decades, the problems of
theoretical and practical knowledge of employees have been in the focus
of researchers from various countries. The suggested model of a
potential evaluation of knowledge has been adapted to the transport
sector. Taking into account a specific character of criteria describing
it, the model is based on education, professional experience, position,
dutifulness, the scope of decision making and responsibility as well as
on self-dependence at work and work culture. Some researchers also
emphasize the use of technologies and the level of difficulty at work,
motivation and employee's contribution to achievements in the goals
of the organization (Morkvenas et al. 2008). An important point is
identifying the problems of passenger transportation using rail
transport in a particular country, i.e. a decrease in passenger flows,
the growth of transportation cost and insufficient financing of
unprofitable means of transportation.
At the moment, the joint-stock company Lietuvos gelezinkeliai is
facing positive changes. The company has invested 3.6 billion Lt during
the last 10 years and now has a modern park of freight locomotives.
Modern signalling, telecommunication and locomotive control systems are
introduced into operation on major routes and at railway stations. Many
railway buildings and passenger waiting rooms have been renovated and
new energy-efficient passenger locomotives have been purchased. In May
2010, work on the project Rail Baltica started. The introduced project
is a railway line connecting Warsaw, Kaunas, Riga, Tallinn and Helsinki.
The route will allow the further development of railway service, i.e.
freight and passenger transportation between the Baltic States and
European countries. The Republic of Lithuania considers Rail Baltica to
be an economic project of national importance.
In 2010, the trains of the company Lietuvos gelezinkeliai were
running on 52 local and 2 international routes. 189 trains were
operating on local routes and 4 trains--on international routes.
Eighteen trains from foreign railway companies arrived at Lithuania or
went in transit through Lithuanian territory.
The dynamics of passenger transportation in 2007-2010 is shown in
Fig. 2. In 2010, 4.4 million passengers, i.e. nearly the same number as
in 2009, were carried by trains, including
--3.5 million passengers carried by local trains;
--0.9 million passengers carried by international trains.
Compared to data on 2009, income from passenger transportation has
grown by 15.7% (Lietuvos gelezinkeliai. Annual report 2010).
As shown in Table 1, the flows of passengers hardly changed (Table
1). The company carried about 4.4 million passengers in 2010 and the
same number is found in 2009.
3. Methods for evaluating the consistency of judgements made by
respondents and experts to determine the significance of criterion
groups describing the quality of railway trips
The quality of passenger transportation is described by a number of
criteria the significance (importance) of which differs to various
extent and is expressed in different measurement units or that of
dimensionless criteria may be determined, based on their comparison
performed by experts. The methods of expert evaluation allow for more
effective organization of analytical work done by experts and are the
problem of a quantitative evaluation of opinions and processing of the
results obtained. The generalized estimate of a group of experts is
taken as a decision (problem situation). If the decision was made by
experts, the consistency of expert judgements would be evaluated. This
is particularly important, because multi-criteria evaluation methods are
used (Zavadskas 1987; Podvezko 2005; Saaty 1980; Maskeliunaite et al.
2009; Brauers et al. 2008; Sivilevicius et al. 2008). The consistency of
expert judgements is described by the concordance coefficient.
The concept of the concordance coefficient of Kendall (1970) is
based on the sum of ranks of each criterion [R.sub.j] taking into
account judgements made by all experts:
[R.sub.j] = [n.summation over (i=1)] [R.sub.ij] (j = 1,2,...,m).
(1)
To be more precise, it is related to the deviation of value
[R.sub.j] from the sum of squares S (analogue of variance) of the total
mean [bar.R]:
S = [m.summation over (j=1)] [([R.sub.j] - [bar.R]).sup.2]. (2)
The average rank [bar.R] of any criterion is obtained by dividing
its sum of ranks by the number of the criterion:
[bar.R] = [m.summation over (j=1)][R.sub.j]/m or [bar.R] =
[n.summation over (i=1)][m.summation over (j=1)] [R.sub.j]/m = n(m +
1)/2 (3)
where [R.sub.ij] is the rank assigned by the i-th expert
(respondent) to the j-th criterion; n is the number of experts (i =
1,2,...,n); m is the number of the criterion (j = 1,2,...,m).
The average rank [[bar.R].sub.j] of each criterion is obtained by
dividing the sum of the given ranks by the number of experts:
[[bar.R].sub.j] = [n.summation over (i=1)][R.sub.ij]/n (j =
1,2,...,m), (4)
where [R.sub.ij] is the rank given by expert i to criterion j;
n--the number of experts.
If S is a real sum of squares obtained by formula (2), the
concordance coefficient, in the absence of tied ranks, is defined by the
relationship between the obtained S and respective maximum [S.sub.max]
(Kendall 1970):
W = 12S/[n.sup.2]m([m.sup.2] - 1) = 12S/[n.sup.2]([m.sup.3] - m)
(5)
The sum of squares S of the deviations of each criterion's
ranks [R.sub.ij] from the average rank can be calculated applying the
formula:
S = [m.summation over (j=1)] [[[n.summation over (i=1)][R.sub.ij] -
[1/2] n(m + 1)].sup.2] (6)
where m is the number of criteria (j = 1,2,...,m) ; n is the number
of experts (respondents) (i = 1,2,...,n).
Random value S is calculated by adding all values assigned to all
considered objects.
The concordance coefficient may be used practically if its limit
value (when expert estimates are consistent) is determined. M. Kendall
has proved that if the number of objects (criteria) is m > 7, the
significance of the concordance coefficient may be determined using [chi
square] (chi-square) Pearson criterion.
The random value
[chi square] = n (m - 1) W = 12S/nm(m + 1) (7)
is distributed according to [chi square] distribution with v = m -
1 degree of freedom. According to the specified level of significance
[alpha] (under real conditions, [alpha] is chosen to be equal to 0,05 or
even 0,01), critical value [[chi square].sub.kr] = [[chi
square].sub.v;[alpha]] is taken from the table of [chi square]
distribution with v = m - 1 degree of freedom. If [chi square]
calculated by formula (7) is larger than [[chi square].sub.kr], it means
that expert (respondent) estimates are consistent.
When the number of compared criteria m ranges from 3 to 7, m ranges
from 3 to 7, distribution [chi square] should be used sparingly because
the critical value of [[chi square].sub.kr] may be higher than the
calculated value, though the consistency of expert estimates is still
considered to be sufficient. In this case, the probability tables of the
concordance coefficient or the tables of critical values S (with 3 [less
than or equal to] m [greater than or equal to] 7)) should be used
(Podvezko 2005).
The lowest value of concordance coefficient [W.sub.min] does not
allow us stating that the estimates of all n experts (respondents) of
the quality of the investigated object based on m criteria and having
the specified (required) significance a and the degree of freedom v = m
- 1 are consistent and can be calculated using the formula suggested by
(Sivilevicius 2011b):
[W.sub.min] = [[chi square].sub.v,[alpha]]/n(m - 1) (8)
where [[chi square].sub.v,[alpha]] is the critical value of the
Pearson's statistic, found with reference to the table taking the
degree of freedom and significance a ([[chi square].sub.v,[alpha]] =
[[chi square].sub.kr]).
In practice, it is easier to use the significances with the largest
(the best) values (Zavadskas 1987).
When the quality of an object is determined by an additive
mathematical model used for calculating a complex (integrated) quality
criterion (allowing quality to be defined by a single number and
compared to the quality of other objects), significance indicators
[Z.sub.j] rather than average ranks [[bar.R].sub.j] should be used,
which does not indicate the significance level of one or another rank.
The significance of criteria describing the quality of the object
evaluated by experts can be determined normalizing them (i.e. making
their sum equal to one) and calculating the significance indicator
[Z.sub.j] of each criterion from the formula given in (Sivilevicius
2011b):
[Z.sub.j] = [(m + 1) - [[bar.R].sub.j]]/[m.summation over
(j=1)][[bar.R].sub.j], (9)
where m is the number of criteria describing the quality
(characteristics) of the considered object; [[bar.R].sub.j] is the
average rank of the j-th criterion calculated by formula (4).
4. Criteria describing the quality of a railway trip and
determining their significance by applying the method of comparative
analysis
Criteria describing the quality of passenger transportation by
railway (railway trip) were determined and grouped (Table 2). The
conducted analysis was based on the application of the AHP method.
Criteria describing the quality of the railway trip were collected and
appropriate questionnaires on a survey were prepared and later
distributed among the respondents (passengers) and experts (service
staff and representatives of the Passenger Transportation Directorate of
the joint-stock company Lietuvos gelezinkeliai). The diagram of criteria
for establishing the quality of railway trips is divided into groups A,
B, C and D (structure of the questionnaire) and given in Table 2. The
survey was conducted within the period from 3 September 2007 to 16
January 2008. The questionnaire was translated into English and Russian
languages. Thirty two questionnaires were distributed among the
passengers representing eighteen citizens of Lithuania, nine of Russia,
one of the USA, one of Spain, one of Italy, one of Germany and one of
Great Britain. However, only 10 questionnaires were completed by the
passengers (3 from Lithuania, 4 from Russia, 1 from the USA, 1 from
Germany and 1 from Italy) and actually used in the survey because the
remaining 22 questionnaires were found to be inconsistent and,
therefore, rejected. Moreover, 17 questionnaires were given to experts
(i.e. service staff) and only 11 of those were used in the survey. Four
questionnaires were handed over to the managers of the department of
passenger transportation of the joint-stock company Lietuvos
gelezinkeliai and only three of those were applied in the survey.
Meanwhile, one completed questionnaire was rejected for the reason
described above.
The matrix comparing evaluation criteria ([a.sub.ji] =
1/[a.sub.ij]) is as follows (Saaty 1980; Sivilevicius 201 la; Turskis,
Zavadskas 2010; Vilcekova et al. 2011):
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (10)
All RTQ criteria were divided into criterion groups A, B, C and D
(Table 2) using AHP (Analytic Hierarchy Process) technique. The
respondents belonging to three categories, i.e. train passengers
(category K), service staff (category P) and administrative staff
(category A) as well as experts filled in the questionnaires. The
respondents and experts (Fig. 3) had to compare the criteria in each
group thus determining the weight (significances) of the criteria at a
particular hierarchical level with respect to a higher hierarchical
level or to non-structured criteria. The largest eigenvalue
[[lambda].sub.max], C.I. (consistency index) and C.R. (consistency
ratio) were calculated for each questionnaire. The questionnaires with
inconsistent evaluation data were rejected. The estimates of RTQ
criteria found in the properly completed questionnaires were assigned
particular ranks (Table 3) and checked if there were any matching
estimates provided by the same respondents and experts, i. e. if they
were consistent (Table 4).
5. Comparative analysis of the estimates provided by the
respondents and experts
The significant estimates of the j-th criterion in group A elicited
from the respondents and experts of all three categories (K, P, A) are
expressed by weight coefficient [[bar.Q].sub.Aj] calculated by the
formula:
[[bar.Q].sub.Aj] = [3.summation over (e=1)][Q.sub.Aje]/3 =
[Q.sub.AjK] + [Q.sub.AjP] + [Q.sub.AjA]/3, (11)
where [Q.sub.Aje] is weight coefficient (j = 1,2,...,m) assigned to
the j-th criterion of criterion group A (in the questionnaires) by the
experts (representing respondent category e); [Q.sub.AjK] is weight
coefficient assigned to the j-th criterion of criterion group A by the
passengers; [Q.sub.AjP] is weight coefficient assigned to the j-th
criterion of criterion group A by service staff; [Q.sub.AjA] is weight
coefficient assigned by administrative staff to the j-th criterion of
criterion group A.
Mean weight coefficients [[bar.Q].sub.Bj], [[bar.Q].sub.Cj],
[[bar.Q].sub.Dj], of the criteria in other groups (B, C, D) were
calculated using similar formulas (Table 3).
Thus, the profound analysis of the significance estimates of
quality criteria for passenger transportation (railway trips) has shown
that criteria A7, A2 and A6, BI, B15 and B13, Cl, C5 and C4, D4, Dl and
D7 are the most significant to respondents and experts (Tables 2 and 3)
(Maskeliunaite et al. 2009; Sivilevicius, Maskeliunaite 2010).
Given the ranks assigned by the experts and respondents of a
particular category to RTQ criterion (Table 3), the consistency of their
estimates may be determined based on the sum of differences in ranks.
The smaller is the sum of the module of differences in the ranks
assigned to the criteria by two categories of the respondents and
experts or experts and experts, the more uniform are the estimates. The
average differences between the ranks are calculated by the formula
[[bar.R].sub.1-2] = [m.summation over (j=1)][[absolute value of
[R.sub.1-2]].sub.j]/m, (12)
where [[absolute value of [R.sub.1-2]].sub.j] is the modulus of
differences (absolute value) in the ranks assigned to criterion groups
(A, B, C or D) or to the j-th criterion by the criteria or by the
experts and experts; m is the number of the criteria in a group; indices
R1 and R2 denote the respondents of the 1st and 2nd categories and
experts the criterion ranks of whose are compared. When the ranks given
to all criteria of the group by the respondents and experts match each
other, then, [[bar.R].sub.1-2] = 0, and the estimates are the same. When
value [[bar.R].sub.1-2] is growing, differences in the estimates are
also increasing. The calculated [[bar.R].sub.1-2] values (Table 4) show
that, in criterion group A, the estimates of all criteria elicited from
the passengers and administrative staff are more consistent (close to
each other) ([[bar.R].sub.K-A] = 2.31). Data on criterion group B show
that the estimates elicited from the service and administrative staff of
the train are more consistent ([[bar.R].sub.P-A] = 1.89). Information on
criterion group C indicate that estimates for the significance of
criteria elicited from passengers and administrative staff are
absolutely the same ([[bar.R].sub.K-A] = 0). Data on criterion group D
discloses that more consistent estimates for the significance of
criteria were elicited from service and administrative staff
([[bar.R].sub.P-A] = 1.25).
[FIGURE 3 OMITTED]
6. Results of the survey into respondent and expert opinions on the
significance of criterion groups describing the quality of a railway
trip
When grouping criteria describing passenger transportation by
railway (railway trip) and establishing their weight (significance), the
preference order or ranks of criterion groups A, B, C and D were
determined.
Twenty one passengers of the train 'Vilnius-Moscow' and
29 experts (including 20 service staff members of the train and 9
members of the administrative staff of the company Lietuvos
gelezinkeliai) competent in describing the structure and constituent
parts of the train, the technical state of the railway, the management
and technology of a railway trip, provisions for trip safety and
requirements for the quality of transportation were given questionnaires
(Table 5) and asked to assign ranks according to their importance and
considering criterion groups A, B, C and D (Table 2) describing the
quality of the railway trip. Based on the above described methods, the
consistency of the estimates of each respondent/expert and the
judgements of the whole group were determined.
The bar diagrams of calculated average ranks [[bar.R].sub.k]
assigned by all respondents and experts to criterion groups A, B, C and
D (Table 2) describing the quality of the railway trip are presented in
Figures 4, 5 and 6.
7. Comparative analysis of the significance of criterion groups A,
B, C and D describing the quality of the railway trip and opinions of
the respondents and experts involved into the evaluation process
Table 6 provides general data on questionnaires, including
estimates [R.sub.j] showing the significance of criterion groups A, B, C
and D describing the quality of a railway trip, interpreted by the
surveyed respondents (21 passengers) and experts (20 members of the
train team and 9 staff members of the company Lietuvos gelezinkeliai).
Based on the methods described above, the consistency of opinions
expressed by the respondents and experts representing two different
categories as well as the consistency of the estimates elicited from all
respondents and experts (50 evaluators) (Table 6) was determined.
The significance indicator [Z.sub.j] of any criterion obtained from
the respondents and experts is calculated using formula (8).
The calculated average ranks [[bar.R].sub.k] given to criterion
groups A, B, C and D describing the quality of a railway trip show that
criterion C is more significant than B, D and A, i.e. the following
hierarchical order is obtained: C > B > D > A. However, from
the point of view of the service staff of the train, average ranks
[[bar.R].sub.k] given by them to criterion groups A, B, C and D
describing the quality of a railway trip indicate that criterion D is
more important than criteria A, C and B implying that the hierarchical
order is as follows: D > A > C > B. From the point of view of
the administrative staff of Lietuvos gelezinkeliai--C > D >B >
A. The calculated average ranks [[bar.R].sub.k] do not show the
importance of one or another criterion group. The estimates of the
respondents (passengers) and experts (administration staff) are closer
to each other (Table 7) and match with determining the priority of
criterion groups A and C with respect to the quality of the railway
trip.
Concordance coefficient W, critical value [[chi
square].sub.[alpha],v] (taken from the distribution table with a
respective degree of freedom [alpha] = 0.010) and the lowest value of
concordance coefficient [W.sub.min] obtained for the estimates provided
by the respondents and experts are given in Table 6.
The estimates of the significance (preference) of criterion groups
A, B, C and D describing the quality of a railway trip are elicited from
the respondents (passengers) and remain inconsistent, whereas the
judgements made by the experts (train service and administrative staff
of the company Lietuvos gelezinkeliai) are consistent. A general opinion
of all respondents and experts also lacks consistency. It can be assumed
that the interests and requirements of passengers differ to great
extent: for example, some of them are primarily interested in low ticket
prices actually ignoring the level of comfort, while others, on the
contrary, give preference to comfort not paying much attention to the
ticket price.
The estimates of the significance of criterion group A determined
by the respondents and experts of all three categories (K, P, A) take
part in the survey, are expressed by the mean value of weight
coefficient [[bar.Z].sub.A] (when the number of the respondents and
experts in each category is the same) and are calculated applying the
formula
[[bar.Z].sub.A] = [3.summation over (e=1)][Z.sub.Ae]/3 = [Z.sub.AK]
+ [Z.sub.AP] + [Z.sub.AA]/3, (13)
where [Z.sub.A] is weight coefficient given by the respondents of
the e-th category (experts) to criterion group A; [Z.sub.AK] is weight
assigned to criterion group A by passengers; [Z.sub.AP] is weight
assigned to criterion group A by the service staff of the train;
[Z.sub.AA] is weight assigned to criterion group A by administrative
staff.
The mean values of weight coefficients [[bar.Z].sub.B],
[[bar.Z].sub.C], [[bar.Z].sub.D] of other criterion groups (B, C, D)
were calculated employing similar formulas (Table 8):
[[bar.Z].sub.B] = [3.summation over (e=1)][Z.sub.Be]/3 = [Z.sub.BK]
+ [Z.sub.BP] + [Z.sub.BA]/3; (14)
[[bar.Z].sub.C] = [3.summation over (e=1)][Z.sub.Ce]/3 = [Z.sub.CK]
+ [Z.sub.CP] + [Z.sub.CA]/3; (15)
[[bar.Z].sub.D] = [3.summation over (e=1)][Z.sub.De]/3 = [Z.sub.DK]
+ [Z.sub.DP] + [Z.sub.DA]/3. (16)
The significance of criterion group A determined by the respondents
and experts of all three categories (K, P, A) is expressed by the mean
value of weight coefficient [[bar.Z].sup.*.sub.A] (when the number of
the respondents and experts in each category is not the same). This
coefficient is calculated by the formula
[[bar.Z].sup.*.sub.A] = [n.summation over (i=1)] [Z.sub.i] x
[n.sub.e]/[N.summation over (i=1)] [n.sub.e] = [[Z.sub.AK] x [n.sub.K] +
[Z.sub.AP] x [n.sub.P] + [Z.sub.AA] x [n.sub.A]]/[[n.sub.K] + [n.sub.P]
+ [n.sub.A] (17)
where [Z.sub.AK], [Z.sub.AP], [Z.sub.AA] denote weight coefficient
given by the respondents (experts) of categories K, P and A to criterion
group A; [n.sub.K], [n.sub.P], [n.sub.A] are the numbers of the
respondents (K) and experts (P, A).
In the considered case, the number of the respondents and experts
is not the same, and therefore formula (17) is used, which is more
suitable because the respondents make a larger part--21 passengers. The
performed research accepts the opinion of passengers to be of primary
importance.
Mean weight coefficient [[bar.Z].sup.*.sub.k] showing the
significance of criterion groups B, C and D was determined by the
respondents and experts of categories K, P and A, and therefore is
calculated applying formulas:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (18)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (19)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (20)
A general model for calculating the quality of a passenger train,
when the weight coefficients of criterion groups were obtained by expert
evaluation, is determined by the formula:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (21)
where K is the complex quality evaluation criterion of an
international train (K may be in the range from 0 to 1);
[[bar.Q].sub.Aj], [[bar.Q].sub.Bj], [[bar.Q].sub.Cj], [[bar.Q].sub.Dj]
denote the mean weights of the j-th criteria of the k-th group
determined by the expert evaluation method; [x.sub.Aj], [x.sub.Bj],
[x.sub.Cj], [x.sub.Dj] are the variables of the j-th criteria of the
k-th group used for calculating the real criterion value ranging from 0
to 1.
The calculation results of significance (weight) [Z.sub.k] and the
ranks of criterion groups A, B, C and D describing the quality of the
railroad trip determined by passengers (K), service staff of the train
(P) and administrative staff (A) of the company Lietuvos gelezinkeliai
are given in Table 8. The broken lines of significance (weight)
[Z.sub.k] and mean weight coefficients [[bar.Z].sub.k],
[[bar.Z].sup.*.sub.k] are shown in Fig. 7.
Passengers think that criterion group C describing the quality of
the trip by international train is the most important because [Z.sub.CK]
= 0.2714. Criterion group B seems to be less important because
[Z.sub.BK] = 0.2619 while criterion group A with [Z.sub.AK] = 0.2143 is
assessed as the least important (Fig. 7). For the service staff of the
train, criterion group A ([Z.sub.AP] = 0.2800) followed by criterion
group D ([Z.sub.DP] = 0.3350) is the most significant. The
administrative staff of Lietuvos gelezinkeliai believe that criterion
group C with respective [Z.sub.CA] = 0.3667 is definitely the most
important while criterion group A is the least important due to the fact
that, in this case, [Z.sub.AA] = 0.1444. The values of weight
coefficients often differ or even contradict each other. The mean values
of the weight coefficients of various groups of criteria are calculated
using different formulas. Therefore, they show that criterion group D is
more significant than group C, while C is more important than B and B,
in turn, is more important than A. The considered data allow for a
conclusion that railway and train parameters (criterion group A) are the
least important for all categories of the respondents and experts,
whereas the parameters of traffic safety (criterion group D), on the
contrary, are the most important.
[FIGURE 7 OMITTED]
8. Conclusions
The processed data obtained from the surveyed respondents
(passengers) and experts (service staff of the train and administrative
staff of the joint-stock company Lietuvos gelezinkeliai) show that the
problem of the quality of the railway trip allow the authors determining
the weight (significance) of criterion groups A, B, C and D and
individual criteria describing the quality of the railway trip in
various trains and expressing it in a single number.
The use of methods for expert evaluation and AHP technique in
particular was a difficult task of the conducted research. Differently
from the respondents, the passengers of the train did not show enough
initiative. The motivation and experience of the service staff of the
train helped them with filling in the questionnaires. The
passengers' (K) estimates of the significance of criterion groups
A, B, C and D describing the quality of the railway trip are not
consistent. The needs of passengers vary to great extent: some are
interested in the trip cost ignoring the range and quality of the
provided services while the other part requires various high-quality
services and comfort at any cost.
The opinions of service staff (P) on criterion groups A, B, C and D
describing the quality of the railway trip are in a good agreement
(consistent). These people are qualified evaluators.
The estimates of the administrative staff of Lietuvos gelezinkeliai
provided for criterion groups A, B, C and D describing the quality of
the railway trip are consistent.
The opinions of passengers and administrative staff about the
significance (preference) of criterion groups A, B, C and D describing
the quality of the railway trip are similar.
The present investigation has determined the weight coefficients
[[bar.Z].sup.*.sub.k] of the groups of criteria of the additive model
(20) and the mean weight coefficients [[bar.Q].sub.kj] of the particular
criteria of these groups.
doi: 10.3846/20294913.2012.710178
Acknowledgements
The authors express their gratitude to the administration of the
joint-stock company Lietuvos gelezinkeliai for the provided possibility
of conducting a survey in the train 'Vilnius-Moscow. They also
thank administration, train service staff and passengers for
participating in research.
References
Brauers, W. K. M.; Zavadskas, E. K.; Peldschus, F.; Turskis, Z.
2008. Multi-objective decision-making for road design, Transport 23(3):
183-193. http://dx.doi.org/10.3846/1648-4142.2008.23.183-193
Belov, I. V.; Tereshina, N. P.; Galaburda, V. G., et al. 2001.
Ekonomika zheleznodorozhnogo transporta. Moskva: Marshrut. 600 s. ISBN
5-89035-056-0.
Bureika, G. 2011. Multicriteria evaluation of operational
effectiveness of freight diesel locomotives on Lithuanian railways,
Transport 26(1): 61-68. http://dx.doi.org/10.3846/16484142.2011.561947
Campbell, J. M.; Smith, S. D.; Forde, M. Ch.; Ladd, R. D. 2007.
Identifying hazards in transportation construction and maintenance
tasks. Case-based reasoning approach using railroad data, Transportation
Research Record 1995: 69-75. http://dx.doi.org/10.3141/1995-09
Clever, R.; Hansen, M. M. 2008. Interaction of air and high-speed
rail in Japan, Transportation Research Record 2043: 1-12.
http://dx.doi.org/10.3141/2043-01
Dailydka, S. 2010. Choosing railway vehicles for carrying
passengers, Transport 25(1): 11-16.
http://dx.doi.org/10.3846/transport.2010.02
Dinh, V. N.; Kim, K. D.; Warnitchai, P. 2009. Dynamic analysis of
three-dimensional bridge-high-speed train interactions using a
wheel-rail contact model, Engineering Structures 31: 3090-3106.
http://dx.doi.org/10.1016/j.engstruct.2009.08.015
Fitzmaurice, M. 2005. Use of wireless local area networks in rail
and urban transit environments, Transportation Research Record 1916:
42-46. http://dx.doi.org/10.3141/1916-07
Gasparik, J.; Zitricky, V. 2010. A new approach to estimating the
occupation time of the railway infrastructure, Transport 25(4): 387-393.
http://dx.doi.org/10.3846/transport.2010.48
Gelumbickas, G.; Vaiciunas, G. 2011. Analysis of passenger rolling
stock foults and its statistics in Lithuania, Transport 26(3): 315-319.
http://dx.doi.org/10.3846/16484142.2011.618190
Hagiwara, Y.; Ishikawa, S.; Furuya, M. 2007. Innovative lightweight
technologies using power electronics on shinkansen high-speed electric
multiple units, Transportation Research Record 1995: 43-51.
http://dx.doi.org/10.3141/1995-06
He, X.-H.; Scanlon, A.; Li, P. 2011. Dynamic factor of bridges
subjected to linear induction motor train load, The Baltic Journal of
Road and Bridge Engineering 6(3): 185-192.
http://dx.doi.org/10.3846/bjrbe.2011.24
Jabri, S.; El Koursi, E. M.; Bourdeaud'huy, T.; Lemaire, E.
2010. European railway traffic management system validation using
UML/Petri nets modelling, European Transport Research Review 2(2):
113-128. http://dx.doi.org/10.1007/s12544-010-0030-5
Kendall, M. E. 1970. Rank Correlation Methods. 4th ed. London:
Griffin and Co.
Kuo, A.; Miller-Hooks, E.; Zhang, K.; Mahmassani, H. 2008. Train
slot cooperation in multicarrier, international rail-based intermodal
freight transport, Transportation Research Record 2043: 31-40.
http://dx.doi.org/10.3141/2043-04
Lei, X.; Zhang, B. 2011. Analysis of dynamic behavior for slab
track of high-speed railway based on vehicle and track elements, Journal
of Transportation Engineering 137(4): 227-240.
http://dx.doi.org/10.1061/(ASCE)TE.1943-5436.0000207
Li, D.; Bilow, D. N. 2008. Testing of slab track under heavy axle
loads, Transportation Research Record 2043: 55-64.
http://dx.doi.org/10.3141/2043-07
Martinelli, F.; Cara, S.; Domenichini, L. 2008. Rail tunnel risk
analysis. A tool to improve rail tunnel design, Transportation Research
Record 2043: 41-48. http://dx.doi.org/10.3141/2043-05
Maskeliunaite, L.; Sivilevicius, H.; Podvezko, V. 2009. Research on
the quality of passenger transportation by railway, Transport 24(2):
100-112. http://dx.doi.org/10.3846/1648-4142.2009.24.100-112
Maskeliunaite, L.; Sivilevicius, H. 2011. Applying AHP technique to
the assessment of railway trip quality (RTQ), in The 8th International
Conference "Environment Engineering": Selected papers, vol. 3.
Ed. by Cygas, D.; Froehner, K. D. May 19-20, 2011, Vilnius, Lithuania.
Vilnius: Technika, 1133-1141.
Metine ataskaita [Annual report]. 2010. Joint-Stock Company
'Lithuanian Railways' (AB "Lietuvos gelezinkeliai").
Morgan, C. A.; Olson, L. E.; Kyte, T. B. 2007. Development and
pilot testing of crew resource management training program for railroad
industry, Transportation Research Record 1995: 62-68.
http://dx.doi.org/10.3141/1995-08
Morkvenas, R.; Bivainis, J.; Jarzemskis, A. 2008. Assessment of
employee's knowledge potential in transport sector, Transport
23(3): 258-265. http://dx.doi.org/10.3846/1648-4142.2008.23.258-265
Nakagawa, A.; Matsuda, S. 2005. Status of development and running
tests on Japan's superconducting maglev, Transportation Research
Record 1916: 26-29. http://dx.doi.org/10.3141/1916-04
Preston, J.; Wal, G.; Batley, R.; Ibanez, J. N.; Shires, J. 2009.
Impact of delays on passenger train services, Transportation Research
Record 2117: 14-23. http://dx.doi.org/10.3141/2117-03
Podvezko, V. 2005. Agreement of expert estimates, Technological and
Economic Development of Economy 11(2): 101-107.
Ramunas, V.; Gailiene, I.; Podagelis, I. 2011. Increment of railway
line capacity, in The 8th International Conference "Environment
Engineering": Selected papers, vol. 3. Ed. by Cygas, D.; Froehner,
K. D. May 19-20, 2011, Vilnius, Lithuania. Vilnius: Technika, 1193-1199.
Rao, A.; Tsai, T. 2007. Safety standards for high-speed rail
transportation, Transportation Research Record 1995: 35-42.
http://dx.doi.org/10.3141/1995-05
Saaty, T. L. 1980. The Analytical Hierarchy Process. New York: M.
Graw-Hill.
Sasaki, K. 2005. Position-detecting system based on the global
positioning system. Potential use in improving car body tilt control,
Transportation Research Record 1916: 30-33.
http://dx.doi.org/10.3141/1916-05
Scalea, F. L.; Bartoli, I.; Rizzo, P.; Fateh, M. 2005. High-speed
defect detection in rails by noncontact guided ultrasonic testing,
Transportation Research Record 2005: 66-77.
http://dx.doi.org/10.3141/1916-10
Schwieterman, J. P.; Scheidt, J. 2007. Survey of current high-speed
rail planning efforts in the United States, Transportation Research
Record 1995: 27-34. http://dx.doi.org/10.3141/1995-04
Si, B.; Zhong, M.; Gao, Z. 2009. Bilevel programming for evaluating
revenue strategy of railway passenger transport under multimodal market
competition, Transportation Research Record 2117: 1-6.
http://dx.doi.org/10.3141/2117-01
Sivilevicius, H.; Maskeliunaite, L. 2010. The criteria for
identifying the quality of passengers' transportation by railway
and their ranking using AHP method, Transport 25(4): 368-381.
http://dx.doi.org/10.3846/transport.2010.46
Sivilevicius, H. 2011a. Modelling the interaction of transport
system elements, Transport 26(1): 20-34.
http://dx.doi.org/10.3846/16484142.2011.560366
Sivilevicius, H. 2011b. Application of expert evaluation method to
determine the importance of operating asphalt mixing plant quality
criteria and rank correlation, The Baltic Journal of Road and Bridge
Engineering 6(1): 48-58. http://dx.doi.org/10.3846/bjrbe.2011.07
Sivilevicius, H.; Zavadskas, E. K.; Turskis, Z. 2008. Quality
attributes and complex assessment methodology of the asphalt mixing
plant, The Baltic Journal of Road and Bridge Engineering 3(3): 161-166.
http://dx.doi.org/10.3846/1822-427X.2008.3.161-166
Somov, D.; Bazaras, Z. 2011. The restoration of physical and
mechanical properties of wheel rim metal, Transport 26(3): 240-247.
http://dx.doi.org/10.3846/16484142.2011.622132
Stribersky, A.; Moser, F.; Rulka, W. 2000. Structural dynamics of
rail vehicle systems: a virtual systems approach, Developments in
Engineering Computational Technology, 29-36.
Turskis, Z.; Zavadskas, E. K. 2010. A new fuzzy additive ratio
assessment method (ARAS-F). Case study: the analysis of fuzzy multiple
criteria in order to select the logistic centers location, Transport
25(4): 423-432. http://dx.doi.org/10.3846/transport.2010.52
Vilcekova, S.; Kridlova Burdova, E.; Ondova, M. 2011. Site
selection and project planning resulting in sustainable buildings, in
The 8th International Conference "Environment Engineering":
Selected papers, vol. 3. Ed. by Cygas, D.; Froehner, K. D. May 19-20,
2011, Vilnius, Lithuania. Vilnius: Technika, 1009-1013.
White, T. 2005. Alternatives for railroad traffic simulation
analysis, Transportation Research Record 1916: 34-41.
http://dx.doi.org/10.3141/1916-06
Zavadskas, E. K. 1987. Kompleksnaia otsenka i vybor
resursosberegaiushchikh reshenij v stroitelstve. Vilnius: Mokslas. 212
s.
Zhao, J.; Chan, A. H. C.; Burrow, M. P. N. 2007. Probabilistic
model for predicting rail breaks and controlling risk of derailment,
Transportation Research Record 1995: 76-83.
http://dx.doi.org/10.3141/1995-10
Zhu, Z-y.; Ling, X-z.; Chen, S-j.; Zhang, F.; Wang, Z-y.; Wang,
L-n.; Zou, Z-y. 2011. Analysis of dynamic compressive stress induced by
passing trains in permafrost subgrade along Qinghai-Tibet Railway, Cold
Regions Science and Technology 65: 465-473.
http://dx.doi.org/10.1016/j.coldregions.2010.10.011
Lijana MASKELIUNAITE. Doctoral student at the Department of
Technological Transport Equipment, Faculty of Transport Engineering,
Vilnius Gediminas Technical University. Transport engineer (2006) from
the Department of Technological Transport Equipment, Faculty of
Transport Engineering, Vilnius Gediminas Technical University. MSc in
transport engineering (2008). Co-author of 4 scientific papers. Several
years of experience of passenger transportation by railway, including
the organization of passenger railway travel, provision of services for
passengers and service quality. Research interests: passenger
transportation by railway.
Henrikas SIVILEVICIUS. Dr Habil Prof. at the Department of
Transport Technological Equipment, Vilnius Gediminas Technical
University, Lithuania. PhD in the construction of automobile roads
(1984). Dr Sc (2002) in civil engineering. The author and co-author of
more than 180 research papers. Research interests: hot mix asphalt (HMA)
production quality and development of quality control methods;
statistical and expert methods used in the transport system, methods for
assessing the state and service life of flexible road pavement.
Lijana Maskeliunaite (1), Henrikas Sivilevicius (2)
Department of Transport Technological Equipment, Vilnius Gediminas
Technical University, Plytines g. 27, LT-10105 Vilnius, Lithuania
E-mails: (1)
[email protected]; (2)
[email protected] (corresponding author)
Received 19 December 2011; accepted 02 July 2012
Table 1. Passenger flows in 2006-2010
(Lietuvos gelezinkeliai. Annual report 2010)
Year 2007 2008 2009 2010
Payload quantity, mil passengers 5.2 5.1 4.4 4.4
Revenue passenger kilometres, mil 408.7 397.5 356.9 373.1
Average distance per passenger, km 78.8 78.5 81.6 85.5
Average number of trips per 2 2 1 1
inhabitant of the country
Table 2. Criterion groups A, B, C and D describing
the quality of the railway trip
Criteria for the quality of the railway trip
A. Criteria for train B. Criteria for railway trip
elements and planning and technology
technical state of
rails (railway track)
1. Roughness of a 1. Departure and arrival of
railway track trains at the scheduled time
2. Speed of train 2. Delivery of meals included
travel (trip into the ticket price
duration)
3. State of coach 3. Delivery of bedclothes and
exterior (cleanness, their condition; making up
deformation, damage) the bed and its condition
4. Noise reduction 4. Possibility of ordering
measures (measures meals and beverages from the
of noise insulation) dining to the compartment
(by car attendant)
5. Passenger coach 5. Onboard distribution
interior of popular press
6. Operation of 6. Possibility of access
ventilation, air to the Internet
conditioning,
cooling and
lighting systems
in terms of their
timely switch
on/off
7. Temperature 7. Possibility of buying a
required inside a ticket on the train (from
passenger car the train manager)
8. Type (simple or 8. Possibility of reserving
vacuum) and a seat in the dining
condition of
sanitary units
(lavatories)
9. Construction of 9. Possibility of calling a
plank-beds (safety taxi
belts of upper level
plank-beds), special
facilities for
people with
disabilities
10. Availability of a 10. Possibility of settling
regularly operating for onboard services using
shower payment cards
11. Special 11. Onboard sales of
compartments for souvenirs
transporting
bicycles
12. Smoking places 12. Music broadcast and
information in conformity
with passenger requests
13. Radio broadcasting 13. Safekeeping of passenger
unit and its luggage and personal items
centralized
operation (switching
on/off)
14. Dining-car 14. Possibility of acquiring
(buffet-car) a health insurance card
valid abroad
15. Possibility of 15. Possibility of obtaining
calling an attendant a visa at the cross border
to a passenger station
compartment in
emergency cases
16. Possibility of 16. Exterior appearance
using tools off service staff (uniform,
(hairdryer, iron, footwear, hairstyle,
etc.) identification card)
- 17. Communication culture of
service staff (with
passengers and colleagues)
- 18. Foreign language skills
of service staff
- 19. Competence,
impersonality and
communication culture of
customs and cross border
station officers while
dealing with passengers
Criteria for the quality of the railway trip
A. Criteria for train C. Criteria for the
elements and price of a trip ticket
technical state of
rails (railway track)
1. Roughness of a 1. Ticket price
railway track
2. Speed of train 2. Price of meals
travel (trip served in the
duration) dining-room
3. State of coach 3. Price of newspapers
exterior (cleanness, and magazines
deformation, damage)
4. Noise reduction 4. Price of health
measures (measures insurance card
of noise insulation) valid abroad
5. Passenger coach 5. Price of the visa
interior
6. Operation of 6. Delivery of meals
ventilation, air (included into ticket
conditioning, price) to passengers
cooling and going in the
lighting systems first-class double
in terms of their compartment
timely switch
on/off
7. Temperature -
required inside a
passenger car
8. Type (simple or -
vacuum) and
condition of
sanitary units
(lavatories)
9. Construction of -
plank-beds (safety
belts of upper level
plank-beds), special
facilities for
people with
disabilities
10. Availability of a -
regularly operating
shower
11. Special -
compartments for
transporting
bicycles
12. Smoking places -
13. Radio broadcasting -
unit and its
centralized
operation (switching
on/off)
14. Dining-car -
(buffet-car)
15. Possibility of -
calling an attendant
to a passenger
compartment in
emergency cases
16. Possibility of -
using tools
(hairdryer, iron,
etc.)
- -
- -
- -
Criteria for the quality of the railway trip
A. Criteria for train D. Criteria for
elements and a safe railway
technical state of trip
rails (railway track)
1. Roughness of a 1. Availability
railway track of equipment
for fire safety
2. Speed of train 2. Availability
travel (trip of first medical
duration) aid
3. State of coach 3. Possibility
exterior (cleanness, of calling
deformation, damage) an ambulance
4. Noise reduction 4. Operational
measures (measures state of
of noise insulation) axle-box
overheat and
fire alarm
system
5. Passenger coach 5. Availability
interior of emergency
exits
6. Operation of 6. Condition of
ventilation, air handrails,
conditioning, stairs,
cooling and tambours, doors
lighting systems and locks
in terms of their
timely switch
on/off
7. Temperature 7. Operational
required inside a state of a hand
passenger car brake
8. Type (simple or 8. Possibility
vacuum) and of calling
condition of the police
sanitary units (militia)
(lavatories)
9. Construction of -
plank-beds (safety
belts of upper level
plank-beds), special
facilities for
people with
disabilities
10. Availability of a -
regularly operating
shower
11. Special -
compartments for
transporting
bicycles
12. Smoking places -
13. Radio broadcasting -
unit and its
centralized
operation (switching
on/off)
14. Dining-car -
(buffet-car)
15. Possibility of -
calling an attendant
to a passenger
compartment in
emergency cases
16. Possibility of -
using tools
(hairdryer, iron,
etc.)
-
-
-
Table 3. The mean values of weight coefficients and
average ranks assigned to RTQ criteria by the respondents
and experts
Group of criteria
A B
Criterion
Criterion Mean Rank Catergory Mean Rank
number weight (rank) of weight
in a [Q.sub respondents [Q.sub
questionnaire .Aj] and experts * .Bj]
K P A
1 0.0400 12 15 14 11 0.1072 1
2 0.1035 2 1 7 1 0.0293 14
3 0.0307 16 12 16 13 0.0509 9
4 0.0562 8 9 8 8.5 0.0300 13
5 0.0560 9 8 11 8.5 0.0276 15
6 0.0997 3 3 1 3 0.0401 12
7 0.1042 1 2 3 2 0.0600 6
8 0.0813 5 4 2 6 0.0221 17
9 0.0624 7 10 9 5 0.0219 18
10 0.0709 6 6 6 7 0.0467 11
11 0.0345 15 13 13 15 0.0180 19
12 0.0360 13 7 15 14 0.0262 16
13 0.0450 11 14 4 12 0.0898 3
14 0.0822 4 5 5 4 0.0587 7
15 0.0359 14 11 12 16 0.1030 2
16 0.0452 10 16 10 10 0.0560 8
17 0.0794 5
18 0.0507 10
19 0.0826 4
Group of criteria
C
Criterion
Criterion Category (rank) Mean Rank Category (rank)
number of respondents weight of respondents
in a and experts * [Q.sub and experts *
questionnaire .Bj]
K P A K P A
1 1 2 3 0.3258 1 1 2 1
2 16 14 12 0.0982 4 4 5 4
3 4.5 11 11 0.0551 6 6 6 6
4 10 17 14 0.1517 3 3 3 3
5 12 15 18 0.2844 2 2 1 2
6 7 10 16 0.0849 5 5 4 5
7 8 8 7
8 15 18 17
9 19 16 13
10 13 9 10
11 17 19 19
12 18 13 15
13 2 5 4
14 9 6 8
15 3 1 1
16 11 7 9
17 4.5 4 5
18 14 12 6
19 6 3 2
Group of criteria
D
Criterion
Criterion Mean Rank Category (rank)
number weight of respondents
in a [Q.sub and experts *
questionnaire .Bj]
K P A
1 0.1514 2 2 3 2
2 0.0957 7 3 5 8
3 0.1087 5 7 6 5
4 0.1726 1 5 1 1
5 0.1297 4 6 4 3
6 0.0953 8 4 8 7
7 0.1504 3 1 2 4
8 0.0963 6 8 7 6
9
10
11
12
13
14
15
16
17
18
19
* The categories of respondents and experts: K--passengers,
P--service staff, A--administration staff
Table 4. Differences in the ranks assigned to criteria describing the
quality of a railway trip
The Group of criteria
criterion
in the
questionnaire
A B
Pairs of particular category respondents and
experts compared in the analysis
[R.sub [R.sub [R.sub [R.sub [R.sub [R.sub
.K-P] .K-A] .K-A] .K-P] .K-A] .K-A]
1 1 4 3 -1 -2 -1
2 -6 0 6 4 4 2
3 -4 -1 3 -6.5 -6.5 0
4 1 0.5 -0.5 -7 -4 3
5 -3 -0.5 2.5 -3 -6 -3
6 2 0 -2 -3 -9 -6
7 -1 0 1 0 1 1
8 2 -2 -4 -3 -2 1
9 1 5 4 3 6 3
10 0 -1 -1 4 3 -1
11 0 2 -2 -2 -2 0
12 -8 -7 1 5 3 -2
13 10 2 -8 -3 -2 1
14 0 1 1 3 1 -2
15 -1 -5 -4 2 2 0
16 6 6 0 4 2 -2
17 0.5 -0.5 -1
18 2 8 6
19 3 4 1
[m.summation 46 37 43 59 68 36
over (j=1)]
[[absolute
value of
[R.sub.1-2]]
.sub.j]
[[bar.R 2.87 2.31 2.69 3.11 3.58 1.89
.sub.1-2]]
The Group of criteria
criterion
No in the
questionnaire
Pairs of particular category respondents and
experts compared in the analysis
C D
[R.sub [R.sub [R.sub [R.sub [R.sub [R.sub
.K-P] .K-A] .K-A] .K-P] .K-A] .K-A]
1 -1 0 1 -1 0 1
2 -1 0 1 -2 -5 -3
3 0 0 0 1 2 1
4 0 0 0 4 4 0
5 1 0 -1 2 3 1
6 1 0 -1 -4 -3 1
7 -1 -3 -2
8 1 2 1
9
10
11
12
13
14
15
16
17
18
19
[m.summation 4 0 4 15 22 10
over (j=1)]
[[absolute
value of
[R.sub.1-2]]
.sub.j]
[[bar.R 0.67 0 0.67 1.88 2.75 1.25
.sub.1-2]]
Table 5. A questionnaire for ranking criterion groups A, B, C and D
describing the quality of a railway trip (including ranks [R.sub.1])
No of the A group and brief description of criteria Rank
criteria group
1 Cost of the trip (C) (ticketprice, medical 4
care insurance abroad, visa, etc.)
2 Train elements and technical state of rails 3
(A) (roughness of a track,
train speed, structure and equipment
for a passenger car ensuring
the comfort of passengers)
3 Safe railway trip (D) (availability of 1
fire extinguishing and first medical
care facilities, efficiency of wheelset
control, fire alarm and manual braking
systems, possibility of calling doctors
and policemen to the train)
4 Railway trip planning and technology (B) 2
(timely departure and arrival
of the train, quality of services provided
on train, the appearance
of service staff, their personal contact
with passengers, knowledge of foreign
languages, etc.)
Note: 1. The meaning of numerical ranks: 1 (most important); 2 (more
important); 3 (important); 4 (of medium importance)
Table 6. Concordance coefficient W, critical value [[chi square].sub.
[alpha],v] and the lowest concordance coefficient [W.sub.min]
obtained for the estimates provided by the experts (P, A) and
respondents (K)
Value
The category
of respondents W [W.sub.min] [chi [chi square]
and experts * square] .sub.alpha,v]
K 0.038 0.180 2.37 11.34
P 0.295 0.189 17.70 11.34
A 0.501 0.420 13.53 11.34
B (K, P, A) 0.047 0.0756 6.984 11.34
* The categories of respondents and experts: K--passengers,
P--service staff of the train, A--administrative staff, B--both
respondents (passengers) and experts (train service and
administrative staff)
Table 7. The number of tied ranks obtained
from the respondents and experts
Group of criteria K = P K = A P = A
A 0 1 0
B 0 0 0
C 0 1 0
D 0 0 0
Total 0 2 0
* The notation of respondent and expert
categories: K--passengers, P--service staff
of the train, A--administrative staff
Table 8. Significance (weight) and preference order (rank) of
criterion groups A, B, C and D describing the quality of the
railroad trip determined by passengers (K), service staff of the
train (P) and administrative staff (A) of the company Lietuvos
gelezinkeliai
Weights Criterion group describing the quality
[Z.sub.k], [[bar.Z]. of the railyway trip
sub.k], [[bar.Z].
sup.*.sub.k] A B C D Total
Passengers (n = 21)
[Z.sub.kK] 0.2143 0.2619 0.2714 0.2524 1.000
Rank (preference order) 4 2 1 3 -
Service staff of the
train (n = 20)
[Z.sub.kP] 0.2800 0.1900 0.1950 0.3350 1.000
Rank 2 4 3 1 -
Administrative staff
of Lietuvos
gelezinkeliai (n = 9)
[Z.sub.kA] 0.1444 0.2333 0.3667 0.2556 1.000
Rank 4 3 1 2 -
The average estimate 0.2129 0.2284 0.2777 0.2810 1.000
of all experts and
respondents (n = 50)
[[bar.Z].sub.k]
The average estimate of 0.2280 0.2280 0.2580 0.2860 1.000
all experts and
respondents taking into
account their number in
a category [[bar.Z]
.sup.*.sub.k]
Fig. 2. The dynamics of passenger flows in
2007-2010 (mil) (Lietuvos gelezinkeliai.
Annual report 2010)
Passengers (mil)
2007 2008 2009 2010
Local 4.1 4.1 3.5 3.5
communication
International 1.1 1.0 0.9 0.9
communication
Total 5.2 5.1 4.4 4.4
Note: Table made from bar graph.
Fig. 4. The bar diagram of average ranks given
to criterion groups A, B, C and D describing the
quality of a railway trip considering passengers'
position (n = 21), W = 0.038, [W.sub.min] = 0.180,
[chi square] = 2.37, [[chi square].sub.[alpha],v] = 11.34
Groups of The
quality average
criteria rank
A 2.8571
B 2.381
C 2.2857
D 2.4762
Note: Table made from bar graph.
Fig. 5. The bar diagram of average ranks given
to criterion groups A, B, C and D describing the
quality of a railway trip considering position of
service staff of the train (n = 20), W = 0.295,
[W.sub.min] = 0.189, [chi square] = 17.70,
[[chi square].sub.[alpha],v] = 11.34
Groups of The
quality average
criteria rank
A 2.2
B 3.1
C 3.05
D 1.65
Note: Table made from bar graph.
Fig. 6. The bar diagram of average ranks given to
criterion groups A, B, C and D describing the
quality of a railway trip considering the position
of the administrative staff of the company
Lietuvos gelezinkeliai (n = 9), W = 0.501,
[W.sub.min] = 0.420, [chi square] = 13.53,
[[chi square].sub.[alpha],v] = 11.34
Groups of The
quality average
criteria rank
A 3.556
B 2.667
C 1.333
D 2.444
Note: Table made from bar graph.
