APPRAISAL OF URBAN HARMONIOUS TRANSPORT
Hongmei Wang, Lingyu Jia and Juanjuan Fan
School of Economics and Management, Beijing Jiaotong Universtiy, Haidian District, Beijing 100044, China
Keywords: Urban transport system, Harmony, Entropy weight method, Grey relational analysis.
Abstract: Beijing has experienced abundant development and a mass of problems in urban transportation. Both
transportation facilities and transit capacity have been improving continuously; the rate of public transport
mode has also been increased steadily. But, the rate of private car transit increased and the rate of green
transport mode, i.e. walking and bicycling declined steadily at the same time. Besides, severe congestion
has aroused a lot of criticism. Thus, a comprehensive view of Beijing urban transport is important. In this
paper, the authors develop a municipal harmonious transport evaluation system. Then, the authors propose
tools in evaluating municipal harmonious transport system. Finally, the paper analyzes Beijing urban
transport system harmonious degree empirically and makes a temporal comparison from year of 2001 to
2009.
1 INTRODUCTION
Beijing is in the process of rapid urbanization and
motorization. Both transportation facilities and
transit capacity, such as number of motor vehicles,
length and areas of roads have been improving
continuously or even significantly in recent years.
According to the data in Beijing statistical yearbook,
number of motor vehicles in Beijing has increased
from 2.124 million in 2003 to 4.019 million in 2009.
Length and areas of the eight districts of Beijing
City also climbed to 6247 kilometres and 91.79
million square meters from 3727 kilometres and
73.44 million square meters respectively.
But Downs Law indicates that the increase of
transportation supply will boost the traffic demand.
The new roads will be occupied quickly by the
traffic flow (Downs, 1962).It is also verified by the
practice of Beijing. From 2003 to 2009, the number
of motor vehicles number, road length in Beijing
length and areas in the eight districts of Beijing City
increased by 89.3%, 44.8%, 67.6% and 25.0%
respectively. It is obvious that the road construction
can not meet the traffic demand caused by motor
vehicles growth.
Thus, a series of policies and measures have
been put forward for transportation demand
management. Among the policies, devoting major
efforts to develop public transport is the first choice.
Public transport priority strategy was brought
forward by Beijing government in 2003. At the same
time, the status of public transportation as public
welfare and service was confirmed. Concessionary
fares have been offered to pubic transport users
since 2007. Besides, the government has being
strived to optimize bus network and speed up the
construction of urban railway system. The length of
operating urban railway system extended to 199km
in 2008, 228 km in 2009 and 336 km in 2010 from
that of 114 km in 2003. Public transport in Beijing
developed rapidly as a result. The number of
passengers by buses and trolleybuses increased from
3.794 billion in 2003 to 5.165 billion in 2009; the
number of passengers by urban rail increased from
472 million to 1.423 billion during the same period.
Accordingly, the rate of public transport users
enlarged from 28.2% in 2003 to38.9% in 2009 (Qiu,
2010).
In spite of the improvement in public transport
sector, the green travel mode declined substantially:
rate of cycling declined from 38.4% in 2000 to
18.1% in 2009 (Qiu, 2010). Furthermore, travels by
private cars boost sharply. Private cars drive 15
thousand kilometer a year on average which is 1.5
times of London and 2 times of Tokyo. In addition,
40% of the trip is less than 5 km in distance, which
indicates that commuters are weak consciousness of
green commute. The above reasons result in the
serious traffic congestion in Beijing.
During the fast urbanization and motorization
process, Beijing has taken a series of measures to
548
Wang H., Jia L. and Fan J..
APPRAISAL OF URBAN HARMONIOUS TRANSPORT.
DOI: 10.5220/0003584505480552
In Proceedings of the 13th International Conference on Enterprise Information Systems (PMSS-2011), pages 548-552
ISBN: 978-989-8425-56-0
Copyright
c
2011 SCITEPRESS (Science and Technology Publications, Lda.)
improve the level of transportation management and
service. In spite of the great progress made, there are
still some crucial problems need to be solved. This
paper focuses on whether the urban transport
realized harmonious development.
2 MUNICIPAL HARMONIOUS
TRANSPORT EVALUATION
SYSTEM
The function of urban transport is providing
necessary conditions for residents travel. Urban
transportation system can be considered as a
complex system which is comprised of three
subsystems: transportation modes, transportation
organization and transportation facilities.
Transportation modes include walking, cycling, bus,
metro, taxi and private car. Transportation facilities
mainly consist of urban road, urban rail and service
facilities such as bus stations, parking facilities,
transportation hubs and illumination facilities.
Transportation organization is the aggregation of
soft measures for traffic problems, including
transportation management measures and policies.
The coordinated development of the three
subsystems is the essential feature of Urban
Harmony Transport. What is more, the three
subsystems interact with economic development
level, environment, resources, technique and
policies. As a consequence, in an urban harmonious
transport system, the simultaneous improvement of
internal subsystems and the external factors is
expected to meet the travel demand as much as
possible.
According to the above analysis, this paper
argues that municipal harmonious transport level
should be evaluated by three aspects: transportation
service level, scientificity of resources allocation,
coordination between transportation and
environment. Considering of the availability of the
data, five indicators include ‘accidents per 10000
motor vehicles (A
1
)’, ‘deaths per 10000 motor
vehicles (A
2
)’, ‘direct economic losses in traffic
accident (A
3
)’, ‘rate of lighting lines (A
4
)’ and
‘number of intersection monitors (A
5
)’ will be used
in evaluating urban transport service level. The
following indicators are set to evaluate the
scientificity of resources allocation: ‘ratio of
transport investment (B
1
)’, ‘ratio of public transport
investment (B
2
)’, ‘ratio of suburban transport
investment (B
3
)’, ‘ratio of suburban public transport
investment (B
4
)’, ‘road density (B
5
)’, ‘area of road
per person (B
6
)’, ‘ratio of urban roads areas (B
7
)’,
‘number of public vehicles per 10000 persons (B
8
)’
and ‘ratio of passengers carried by taxis to that of
public traffic (B
9
)’. In reflecting the coordination
between transportation and environment, we adopt
‘ratio of days have 1
st
or 2
nd
class air quality(C
1
)’,
‘average inhalable particles (C
2
)’, ‘average NO
2
concentration (C
3
)’ and ‘average noises db along
arterial roads (C
4
)’.
3 EVALUATION METHOD AND
MODEL
Based on the evaluation index system stated above,
the tendency of the transport harmony degree of
Beijing from the year of 2001 to 2009 is studied in
this part of the paper.
3.1 Method in Calculating Indicators’
Weight
Entropy method is adopted here to calculate the
weights of evaluation indicators. Entropy is the
measurement of the disorder degree of a system and
it can measure the amount of useful information
with the data provided (Zou, 2006). Calculating the
weights of indicators with entropy method is
scientific and objective. The steps of calculating can
be expressed as follows.
(1) Formation of the evaluation matrix. Suppose
there are m indicators and n evaluation objects, the
evaluation matrix
X
can be defined as
mnmm
n
n
xxx
xxx
xxx
X
...
............
...
...
21
22221
11211
where x
ij
is the value of the jth evaluation object on
the ith evaluation indicator.
(2) Standardization of the original evaluation matrix.
Because of the difference in the meanings,
dimensions and criterions of evaluation indicators,
the data need to be pre-processed to transfer into
comparable sequence, which is called
standardization. The origin evaluation matrix
ij
mn
Xx
can be translated into

ij
mn
Rr
, where
01
ij
r
.
If the value of the indicator is the larger the better,
then the data should be standardized by the formula
APPRAISAL OF URBAN HARMONIOUS TRANSPORT
549
as follows:
min
max min
ij ij
ij
ij ij
xx
r
x
x
(1)
If the value of the indicator is the smaller the better,
then the data should be standardized by the formula
as follows:
max
max min
ij ij
ij
ij ij
x
x
r
x
(2)
If the value of the indicator is optimal at a certain
value, then the data should be standardized by the
formula as follows:
1
max
ij i
ij
ij i
x
r
r
x
r

(3)
(3) Calculation of entropy and the weight. The
entropy of the indicator can be represented as
ln
1
n
j
ijiji
ffKH
(4)
where
1
,
n
ij ij ij
j
f
rr
nK ln1
.
The weight of the ith indicator can be calculated
by
1
1
m
ii i
i
wHmH

(5)
3.2 Evaluation Model
There are many influencing factors of the transport
harmony degree. The evaluation index system we
use is quite incomplete due to the limit in data.
Synthetic evaluation method based on grey
relational analysis is adopted in this paper to
evaluate the transport harmony degree. The grey
system theory was built and extended by Deng
(1982, 1989, 1991, 1992 and 1995) and Liu (2004)
et.al. Grey relational analysis is a branch of grey
system theory. The essence of the method is
analyzing and comparing the geometric proximity
between the factors curves and the result curve. The
more similar, the larger the relational degree is. The
synthetic evaluation model based on grey relational
analysis can be represented as
P
EW
, where
P
is the vector of the evaluation results (while each
element of the vector is a relational degree),
W
is
the weights vector of the indicators which the sum of
the elements is 1.
)(...)2()1(
............
)(...)2()1(
)(...)2()1(
222
111
m
m
m
E
nnn
Where
j
i
is the grey relational coefficient
between the jth object on the ith indicator and the
optimum value of the ith indicator. When finishing
the calculation of the grey relational degree, the
sorting of objects could be presented based on the
value of
P
.
The calculation steps are presented as follows:
(1)
The sequence of each indicator should be
presented as
),...,,(
112111 n
xxxX
),...,,(
222212 n
xxxX
… … … … … …
),...,,(
21 mnmmm
xxxX
where
ij
x
is the value of jth evaluation object on the
ith indicator, m is the number of indicators and n is
the number of evaluation objects.
(2)
The optimum sequence of the indicators.
Suppose that
12
, ,...,
n
X
xx x

is the optimum
sequence of the indicators, where
i
x
the optimum
value of evaluation objects on the ith indicator. The
principle of the optimum value is that if the value of
a indicator is the larger the better, then the maximal
value is optimum; if a indicator is the smaller the
better, then the minimal value is optimum; if a
indicator is optimum at a certain value, then the
value is optimum.
With the optimum sequence confirmed, matrix D
can be constructed as
mmnm
n
n
xxx
xxx
xxx
D
...
............
...
...
1
2221
1111
(3)
Standardization of the data sequence. Because of
the difference in the range and dimension of
evaluation indicators, the data need to be pre-
processed to transfer into comparable sequence.
Formula (1), (2), (3) is generally used for
standardization. The standardized matrix can be
represented as
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
550
11 1 1
21 2 2
1
...
...
... ... ... ...
...
n
n
mmnm
rrr
rrr
R
rrr






(4)
Calculation of grey relational coefficient-. Grey
relational coefficient
j
i
is the differentials
between the indicator sequences and comparison
sequence on the ith indicator. The larger the grey
relational coefficient is, the bigger the differential is.
Grey relational coefficient can be calculated by
the formula as follows

min min max max
max max
iij iij
ji
ji
j
iij iij
ji
rr rr
i
rr rr




(6)
where
is the distinguishing coefficient,
0,1
,the distinguishing coefficient is introduced
to reduce the influence of the extremum value
during the procedure of calculating. The smaller the
distinguishing coefficient is, the bigger the
distinguishability is. 0.5 is generally used in
practical application.
(5)
Calculation of grey relational degree. The
distinguishing coefficient is used to analyze the
relationship between the indicator sequence and the
comparison sequence on each indicator. It only
reflects one-side information. In order to obtain the
relationship between the sequences, grey relational
degree is needed. The grey relational degree can be
represented as

1
m
jij
i
pwi
(7)
where
i
w is the weight of the ith evaluation
indicator.
(6)
Sorting. If the grey relational degree
j
P
is
maximal, then
j
R
is most proximate with the
optimum value
R
, which means the jth evaluation
object is better than others. The evaluation objects
could be sorted according to this principle.
4 DYNAMIC ANALYSIS OF THE
TRANSPORT HARMONY
DEGREE OF BEIJNG
4.1 Calculation of the Indicator
Weights
There are 18 evaluation indicators (A1-C4) and 9
objects (from 2001 to 2009) in this case. Values of
the indicators which collected from Beijing
Statistical Yearbook and Beijing Transport Statiscal
Yearbook are presented in Table1.
Entropy can be calculated based on the
standardized data. Then weights of indicators are
obtained by entropy method.
Table 1: Indicators’ data of Beijing harmonious transport system (2001-2009).
Indicators 2001 2002 2003 2004 2005 2006 2007 2008 2009
A
1
132.66 71.34 53.24 37.46 23.88 18.91 14.72 9.30 7.98
A
2
8.8 7.9 7.7 7.6 6.0 4.8 3.8 2.8 2.4
A
3
62670.7 4112.0 4361.1 4058.0 2609.49 2772.0 2285.1 2038.9 2043.4
A
4
31.31 33.71 34.38 41.25 44.25 30.19 28.6 26.2 30.3
A
5
135 212 390 390 440 439 449 537 752
B
1
27.87 37.14 27.15 21.57 25.86 39.21 38.93 39.83 46.27
B
2
45.35 29.16 43.03 52.96 71.68 54.79 44.22 46.69 63.09
B
3
1.52 0.89 8.17 0.00 0.82 0.00 0.03 0.19 0.26
B
4
0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.55
B
5
2.68 2.67 2.72 2.97 2.98 3.23 3.26 4.52 4.56
B
6
5.85 5.71 7.94 7.73 7.81 6.18 6.20 8.56 8.50
B
7
3.7114 3.9026 5.361 5.3255 5.4395 5.1793 5.5776 6.5343 6.7082
B
8
13.74 12.35 13.29 14.54 13.55 12.96 12.57 13.7 13.5
B
9
13.31 12.16 12.13 11.43 12.37 13.69 13.13 11.65 10.32
C
1
50.68 55.62 61.37 62.7 64.1 66 67.4 75.1 78.1
C
2
0.165 0.166 0.141 0.149 0.142 0.161 0.148 0.122 0.121
C
3
0.071 0.076 0.072 0.071 0.066 0.066 0.066 0.049 0.053
C
4
69.6 69.5 69.7 69.6 69.5 69.7 69.9 69.6 69.7
Source: Beijing Statistical Yearbook (2001-2009), Beijing Transportation Yearbook(2001-2009).
APPRAISAL OF URBAN HARMONIOUS TRANSPORT
551
w
i
=0.0180.0420.0150.0410.0290.034
0.025 0.148 0.276 0.081 0.047
0.0320.0320.0360.0280.0490.043
0.020
4.2 Synthetic Evaluation of Beijing
Transport Harmonious Level
According to the data in Table 1 and indicators’
weight, optimum sequence is constructed as
(7.98, 2.4,2038.9, 44.25, 752, 46.27, 71.68,
8.17, 6.55, 4.56,8.56,6.7082,14.54,10.32,
78.1,0.121,0.049, 69.5)
X 
Calculation with the standardized data, the grey
relational degrees of 2001-2009 is 0.368
0.386
0.513 0.450 0.479 0.417 0.424 0.598
0.832 respectively.
Thus, we conclude that the order of the transport
harmony degree of Beijing from 2001 to 2009 is
2009>2008>2003>2005>2004>2007>2006>2002
>2001
5 CONCLUSIONS
In this paper, we evaluate Beijing urban transport
harmonious level dynamically. The research
indicates that Beijing urban transport harmonious
level improved obviously in 2009 and 2008. The
main reasons for the improvement lie in enlarged
ratio of investment in transport (including public
transport), improvement in transport infrastructure
and decrease in occurrence and loss of traffic
accidents. Efforts by the municipal government play
important role in these. Development of public
transportation, e.g. rail, bus and trolley bus
contributes a lot to transport harmony. But for a
metropolitan city, attention and cooperation from the
public will be more effective.
ACKOWLEDGEMENTS
Supported by “the Fundamental Research Funds for the
Central Universities (Appraisal of TDM Measures in
Occurring Urban Congestion)” and “National Nature
Science Fund of China (Research of residents’ Selective
Mechanism in Public Transport)”.
REFERENCES
Deng Julong. Control problems of grey systems, Systems
& Control Letters
[J], 1982 (5), 288-294
Deng Julong. Grey information space,
The Journal of
Grey System
[J], 1989(2), 103-118
Deng Julong. Extent information cover in grey system
theory, The Journal of Grey System, 1995(2), 131-138.
Deng Julong. Negative power AGO in grey theory.
The
Journal of Grey System
, 2001 (3).
Downs, A. The Law of Peak Hour Expressway
Congestion.
Traffic Quarterly, 1962 (3), 393-409.
Liu Sifeng, Dang Yaoguo, Fang Zhigeng (2004), Grey
System Theory and Application (third edition) [M],
Beijing, Science Press.
Baoxing Qiu. Measures to alleviate congestion in Beijing.
http://news.xinhuanet.com/legal/201005/04/c_127225
0.htm
Zou Zhihong, Yun Yi, Sun Jingnan. Entropy method for
determination of weight of evaluating in fuzzy
synthetic evaluation for water quality assessment
indicators [J],
Journal of Environmental Science,
2006, 5(18): 1020-1023.
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
552