Advanced Engineering Informatics 32 (2017) 6676

Contents lists available at ScienceDirect

Advanced Engineering Informatics

journal homepage: www.elsevier.com/locate/aei

Full length article

Data analysis for metropolitan economic and logistics development

Shulin Lan, Chen Yang , George Q. Huang
HKU-ZIRI Lab for Physical Internet, Dept. of Industrial and Manufacturing Systems Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong, China

a r t i c l e i n f o a b s t r a c t

Article history: Logistics industry is an integral sector encompassing transportation, warehousing, handling, circulation
Received 29 September 2016 and processing, delivery and information technology. With the progress of economic globalization and
Received in revised form 4 December 2016 integration, logistics industry has become a new momentum driving the fast development of national
Accepted 7 January 2017
and regional economy. The close relationship between economic development and logistics advancement
receives wide attention from the academia. However, current research on the coordination between
economy and logistics mostly focuses on concept interpretation, and qualitative discussions. Very rarely
Keywords:
do scholars conduct quantitative analysis on the coordination of metropolitan economy and logistics. To
Data analysis
Metropolitan economic and logistics
fill this gap, we first examine whether there exist interactions between metropolitan logistics and econ-
development omy by building evaluation index systems for metropolitan logistics and economy. Then we introduce the
Coordinated development entropy method and Granger causality test to evaluate and test the level of logistics and economic devel-
Relation analysis opment in five cities: Beijing, Shanghai, Guangzhou, Chongqing, and Tianjin from 2009 to 2013. From the
dimensions of regional economic investment, regional economic capacity and strength, we finally test the
relationship between three economic subsystems and three logistics subsystems to further validate the
relationship between metropolitan economy and logistics.
2017 Elsevier Ltd. All rights reserved.

1. Introduction However, current research efforts mostly focus on the influenc-

With the globalization of the worlds economy, metropolises
have played an increasingly important role in the international
economic competition and cooperation. The booming metropolitan
economy and the rapid development of modern commodity mar-
kets both contribute greatly to the rise of metropolitan logistics,
which refers to the logistics activities both within the metropolis
and across metropolitan borders. Metropolitan logistics, as an
important constituent of metropolitan economy guarantees the
smooth operation of metropolitan economy by managing the flow
of things between the point of origin and the point of consumption.
Efficient metropolitan logistics can act as a strong driving force to
the development of metropolitan economy. However, it can also
lead to a waste of governmental and social resources, if the econ-
omy cannot generate enough demand for logistics capability and
further investment are still flowing into the logistics industry. In
other words, the development of metropolitan economy and logis-
tics should be coordinated. Thus attentions should be paid to the
research on coordinated development of metropolitan economy
and logistics, which can maximize the stakeholders benefit.
Corresponding author.
E-mail address: wzhyoung@gmail.com (C. Yang).
ing effects of metropolitan logistics on metropolitan economy and
fail to consider the reverse influence, let alone the mutual effects
between the two using the quantitative methods. The systematic
study on the Coordinated Development between Metropolitan
Economy and Logistics is beneficial to understanding the mutual
effects between the metropolitan logistics and economy, and is
conducive to clarifying whether metropolitan logistics develops
harmoniously with its economy.
This paper attempts to make quantitative analyses of the
impact the metropolitan economic development has on logistics
development, as well as the impact the latter casts on the former.
The development of logistics industry has a far-reaching influence
on the optimization of metropolitan industrial structure. With the
support of information technology, the modern logistics industry
has widely adopted many new information technologies. Judging
from the various effects that economic growth has produced, it
has exerted a positive and profound impact on logistics. It not only
affects the growth of the logistics demand, but also boosts a qual-
itative leap for its progressing [18].
In order to prove the rationality of the above discussions, the
authors, by using of data from China Statistical Yearbook, quantita-
tively analyzes the coordinated development between metropoli-
tan economy and logistics. Beijing, Shanghai, Guangzhou,
Chongqing and Tianjin are taken as research objects. Each group

http://dx.doi.org/10.1016/j.aei.2017.01.003
1474-0346/ 2017 Elsevier Ltd. All rights reserved.
 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676
of analyses involves an economic development index and a logis-
tics index. Corresponding indexes are selected to conduct the cal-
culation of coordinated development degree. Exploring the
interaction between logistics industry and economic development,
and correctly grasping the direction of logistics development, will
cast a positive influence on logistics industry and national eco-
nomic development.
The paper is organized as follows: Section 2 reviews the related
research literature; Section 3 describes the appraisal model of the
coordinated development of metropolitan logistics and economy
and leverages the Granger causality test to examine the interactive
relationship between logistics and economy of five metropolises;
Section 4 explores the relation between economic subsystems
and logistics subsystems from three economic forms at a micro-
scopic level to achieve a systematic and in-depth investigation;
and Section 5 provides brief concluding remarks.
2. Literature review
Logistics industry, as an important industry in the national
economy, attracts attentions from both the academia and industry.
Extensive research has been conducted in this regard. For example,
Wiengarten et al. [27] explored the impact of a countrys logistical
capabilities on external supply chain integration of a company, to
provide insights for companies to deal with economic globaliza-
tion. Among the literature, much attention has been paid to
metropolitan logistics. At present, the studies concerning logistics
mainly focus on the construction planning, influencing factors
and related action mechanisms.
As for the planning and construction of metropolitan logistics,
Taniguchi et al. [25] has proposed the concept of metropolitan
logistics and elaborated on the importance of setting up urban
logistics nodes with considerations of energy reservation, conges-
tion alleviation, environmental protection, and labor cost reduc-
tion. Visser et al. [26] put forward related topics regarding urban
logistics transportation and policy design, including development
policies for freight forwarding in Netherlands, France, Germany
and Japan. These policies were primarily about setting up freight
transportation center based on cooperation between the authori-
ties and transport enterprises.
Some scholars have illustrated the influencing factors of
metropolitan logistics, such as the interpretation of metropolitan
logistics development method from the perspective of policies.
Dablanc et al. [7] analyzed how the Paris government influenced
regional logistics. The article introduced freight transportation in
Paris and evaluated its effects on policies regulating freight trans-
portation. Loo and Hook [17] analyzed the logistics industry in
Hong Kong ports. They arrived at the conclusion that Hong Kong
ports were under multiple influencing factors, among which mar-
ket factors and political factors are the most profound ones. Fisher
[9] comprehensively analyzed logistics nodes in cities with three
different logistics purposes, namely for production, for consump-
tion, and for transshipment, and studied the location and scale of
the logistics nodes.
Some elaborate on metropolitan logistics development by tak-
ing infrastructure construction into account. Gandlur [11] main-
tained that a scientific and effective transport network will speed
up optimal resource allocation. The transport network for logistics
is made of multiple logistics infrastructure. The author construed a
model which comprises of multiple nodes and transportation
routes. On the basis of the model, the author analyzed trade and
logistics costs under different network status. Carruthers et al.
[5] chose Hong Kong and Singapore as the study subjects and ana-
lyzed how modern logistics has driven economic growths in Hong
Kong and Singapore. The article pointed out that investments in
67
fixed assets in logistics will have direct impact on improving sup-
ply capabilities in the regional logistics, thus increasing logistics
demands and promoting regional economic growth.
However, those works seldom explores the influencing indica-
tors that can comprehensively indicate the development level of
metropolitan logistics, which is important for policy making and
investment planning. Also, the effects of metropolitan economy
are not fully considered to promote the logistics development.
Meanwhile, more and more scholars attach more importance to
the influence of economic development on logistics. Based on the
most recent available data on Chinese logistics and economy, Lean
et al. [15] tested the relationship between logistics development
and economic growth in both the short and long run from a total
output, demand and disaggregate output perspective. The result
shows that economic growth Granger-causes logistics output,
implying that economic development causes more demand for
logistics services and hence leads to logistics development. Lee
and Yang [16] analyzed the development strategy of the Korean
Incheon International Airport. The result showed that economic
prosperity in the East Asian region was the main driver behind
the enormous logistics demands. The article also pointed out the
potentials and influences of the airport and offered suggestions
on its future development. Fujita and Mori [10] proposed an evolu-
tionary model of spatial economic development. From the evolu-
tionary model, it can be determined that in the development of
big cities, there exist interactions between clustering economies
and the hub effects of transportation nodes. Arvin et al. [1] ana-
lyzed causal relationships among transportation intensity, eco-
nomic growth, CO2 emissions, and urbanization using data of the
G-20 countries.
As the study on metropolitan logistics deepens, more and more
scholars conduct special studies on roles that metropolitan logis-
tics plays. Among them, the topic on how metropolitan logistics
promotes economic growth has become a hot research spot.
Skjtt-Larsen et al. [23] took the Oresund bridge, built by the
Danish and Sweden governments, as a model of logistics develop-
ment and fully demonstrated what role logistics infrastructure has
played in economic development. Talley [24] put forward a model
illustrating the relation between regional economic growth and
regional transport infrastructure investment, which served as the
basis to clarify the influences that regional transport infrastructure
investment exerts on regional economic production and service.
Based on the model which showed the relation between regional
transport infrastructure investment and economic production,
the authors proved that the relation between the two is compli-
cated. Different from the previous research, the formula in the
model, through evaluating spatial accessibility and transport ser-
vice quality, probed into how regional transport infrastructure
investment influenced economic production and transport service.
Debbage [8] analyzed the correlation between airfreight and regio-
nal economy, and concluded that airfreight was positively driving
regional economic growth. Ham et al. [12], based on mathematical
model, probed into regional material movements and multi-mode
transport system versus regional economic growth in both direct
and indirect ways.
In summary, the relationship between regional logistics and
economy has become closer and has attracted wide attention from
scholars. However, current research still has shortcomings in the
following aspects: Firstly, the existing research taps more on logis-
tics influence on economy and rarely explores the influence of
economy on logistics. This reverse effects of economy on logistics
make it necessary to re-exam the relationship between metropoli-
tan logistics and economy [14]. Secondly, in the empirical analysis
concerning logistics and economic relations, most of them use sin-
gle indicator to represent logistics or economic development, for
example, using total social logistics costs and GDP to indicate the
68 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676

development level of logistics and economy respectively, making it Table 1

difficult to reveal the practical logistics and economic development Evaluation index system for metropolitan logistics and economy.

level. Thirdly, although some scholars adopt multiple indicators to Dimension Indicator
conduct measurement and evaluation, most of them are subjective Indicators for metropolitan Added value in the tertiary industry
evaluation indicators, and the evaluation methods they use are economic development GDP per capita
subjective or qualitative, which would make the studies less scien- Total Investment in Fixed Assets
tific. Fourthly, most work takes a macro perspective in probing into Total retail sales of consumer goods
Urban per capita disposable income
the relationship between logistics and economic and lacks an in- Household consumption level
depth analysis of the relationship between the subsystems of econ- Public income
omy and logistics using mathematical formula. Those gaps should Indicators for metropolitan Number of logistic employees
be filled to provide useful insights and guides for metropolitan logistics development Total fixed asset investment in logistics
development. Total postal traffic volume
Civic truck ownership
Road freight turnover
3. Relation between metropolitan economy and logistics Total length of postal routes
Highway mileage
In order to verify the mutually influencing relationship between
the development of metropolitan economy and logistics, we take
mined by the influence of each indicators variation on the whole
five Chinese metropolises as research subjects and use their statis-
system. The more it varied, the more weighting it will carry. This
tical data of economic and logistics development for correlation
applies to all fields including Statistics and Valuations [21]. Specific
analysis and regression analysis [3].
steps are as follows:
This section includes: the standardized process of original data
about five metropolises obtained from the China Statistical Year-
3.2.1.1. Decision matrix setting-up. Given the development of a city
book; application of the entropy method to determine the weight-
in m years needs to be evaluated and there are n indicators in the
ings of indicators for development levels of logistics and economy
evaluation index system, then it means the evaluation system con-
in five metropolises; leveraging the Granger causality test to exam-
sists of m samples and n evaluating indicators. The decision matrix
ine the interactive relationship between logistics and economy of
X will be:
five metropolises.
2 3
x11 x12 ... x1n
3.1. Standardized processing of original data 6x x22 ... x2n 7
6 12 7
X6
6 .. .. .. .. 7
7 3
This study, based on scientific, systematic, measurable and 4 . . . . 5
available principles, as well as existing research results, teases xm1 xm2 . . . xmn
out an evaluation index system for the development level of
metropolitan logistics and economy. The indexes for metropolitan
3.2.1.2. Standardized decision matrix. After normalizing decision
economic development include: Value-Added of the Tertiary
matrix X, standardized matrix V v ij mn is established v ij is the
Industry, GDP per capita, Total Retail Sales of Consumer Goods,
normalized value of xij . maxxj and minxj are the maximum
Per Capita Disposable Income of Urban Households, Household
and minimum values of the jth indicator.
Consumption Level and Public Finance Revenue. The indexes for
metropolitan logistics development include: Number of Logistic xij  minxj
v ij 4
Employees, Total Fixed Asset Investment in Logistics, Total Postal maxxj  minxj
Business Volume, Civilian Truck Holdings, Road Freight Turnover,
Total Length of Postal Routes and Highway Mileage. As is shown
in Table 1. 3.2.1.3. Weighting of the ith evaluated object under the jth indica-
The original data is obtained from the China Statistical Year- tor. For a certain indicator j, the bigger the variance of v ij means
book, and it should first be processed by standardized methods. the more influence the indicator has on the evaluated object,
We use formula (1) and (2) to standardize the data. namely more useful information are provided to measure evalu-
( ated object by the indicator. Entropy can be used to measure
xi =kmax xi xi : positiv e index how much information are provided according to the definition
xi 1
kmin xi =xi xi : negativ e index of entropy. Under the jth indicator, weighting of the ith evaluated
object is pij . 0 6 pij 6 1.
(
yi =kmax yi yi : positiv e index v ij
pij Pm 5
yi
kmin yi =yi yi : negativ e index
2
i1 v ij

xi and yi represent the indicator value of metropolitan logistics

3.2.1.4. Calculating the entropy value ej under the jth indicator. The
development and the indicator value of metropolitan economic
entropy value ej under the jth indicator can be worked out. m rep-
development respectively.
resents the number of evaluated objects. When pij 0 or pij 1,
pij  lnpij 0.
3.2. Measuring the development of metropolitan logistics and economy
1 X m

3.2.1. Calculate index weighting by using entropy method ej  p  lnpij 6

Based on the standardized data, the entropy method is adopted
to calculate the weightings of relevant logistics and economic
indexes. As an objective weighting method, it determines weight-
ing of relative indicators through evaluating information offered by
each observed value [2]. Weightings of indicators can be deter-
lnm i1 ij
3.2.1.5. Calculating the difference coefficient dj under the jth indica-
tor. According to the formula (6) of entropy value, for a certain
indicator Dj , the more similar the variances of v ij are, the higher
 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676 69

Table 2
Indicators weightings for metropolitan economic and logistics development.

Dimension Indicator Weighting

2009 2010 2011 2012 2013
Indicators for metropolitan economic development Added value in the tertiary industry 0.153 0.146 0.156 0.159 0.155
GDP per capita 0.087 0.089 0.096 0.097 0.100
Total Investment in Fixed Assets 0.086 0.098 0.122 0.137 0.153
Total retail sales of consumer goods 0.202 0.198 0.209 0.204 0.196
Urban per capita disposable income 0.099 0.100 0.112 0.114 0.114
Household consumption level 0.109 0.110 0.119 0.117 0.112
Public income 0.264 0.260 0.187 0.171 0.169
Indicators for metropolitan logistics development Number of logistic employees 0.147 0.140 0.117 0.148 0.098
Total fixed asset investment in logistics 0.091 0.073 0.111 0.101 0.127
Total postal traffic volume 0.186 0.161 0.197 0.196 0.144
Civic truck ownership 0.122 0.175 0.123 0.115 0.097
Road freight turnover 0.088 0.089 0.097 0.095 0.104
Total length of postal routes 0.113 0.114 0.122 0.097 0.141
Highway mileage 0.253 0.249 0.233 0.249 0.288

the value of ej will be. Difference coefficient dj can be determined
according to the definition of entropy. Higher value of dj means
more information provided by the corresponding indicator. As a
result, more weighting should be given to the indicator.
dj 1  ej
3.2.1.6. Determining the weightings for each indicator. The entropy
weight of each indicator can be calculated using Eq. (8). n repre-
sents the number of indicators. Results are shown as Table 2.
dj
wj Pn j 1; 2; . . . ; n
k1 dk
3.2.2. Calculate the levels of logistics and economic development
Based on the indexes weightings and standardized data, the
scores of logistics and economic development can be obtained
via the following formula. f x represents the logistics capability
of a metropolis, while gy represents the economic capability of
the metropolis. The equation goes as follows:
X
m To sum up, metropolitan logistics and economy are closely
f x ai xi
i1
X
m
gy bi yi
i1
ai and bi represent weightings, while xi and yi are the standardized
value of xi and yi . Using Eqs. (9) and (10), the results of calculation
taking five Chinese metropolises as subjects are shown as Table 3.
3.3. Relation analysis of metropolitan economy and logistics
We conduct the Granger causality test (regression analysis) to
explore the relationship between metropolitan economic and
logistics development. The sample sizes of experiments for Table 4
and 6 are both 25 (five cities from 2009 to 2013), while the sample
sizes of experiments for Table 5 and 7 are both 20 (five cities from
2010 to 2013). The data come from the entropy method calculated
in Section 3.2, which are f(x) and g(y) of five metropolises.
As shown in Table 4, we first take logistics development level as
the dependent variable, and economic development level as the
independent variable. The result of regression analysis shows that
the logistics development is significantly related to economic
development at 0.05 significance level. The regression coefficient
is 0.925 (p = 0.003).
As shown in Table 5, we take logistics development level as
dependent variable, and the economic development level and lag
phase11 economic development level as the independent variables.
The result of regression analysis shows that the development of
logistics is significantly related to lag phase1 economic development
7 at 0.1 significance level, the regression coefficient was 3.42
(p = 0.096).
As shown in Table 6, we take the economic development level
as dependent variable, and logistics development level as indepen-
dent variable. The result of regression analysis shows that the
development of economy is significantly related to the develop-
ment of logistics at 0.01 significance level, and the regression coef-
8
ficient is 0.49 (p = 0.003).
As shown in Table 7, we take economic development level as
the dependent variable, and logistics development level and lag
phase1 logistics development level as independent variables. The
result of regression analysis shows that the development of econ-
omy and the development of logistics and the logistics develop-
ment in lag phase1 are significantly correlated at 0.1 significance
level, the regression coefficients are 0.2664 (p = 0.079), 0.4049
(p = 0.095).
9 related. Metropolitan logistics is conducive to the growth of
metropolitan economy, and economic prosperity can promote the
development of metropolitan logistics.
10
3.3.1. Metropolitan logistics is conducive to metropolitan economy
growth
The development of metropolitan logistics provides transporta-
tion support for metropolitan economic development, speeds up
the convergence and flow of needed resources for economic con-
struction, thus meets the needs for economic development [22].
From Table 6, it can be drawn that logistics has a positive influence
on economic development (Coef. = 0.492761; P = 0.003).
The reason lies in that from the perspective of individual con-
sumption, the development of logistics industry is conducive to
further stimulating the consumer demands, thus promoting the
development of consumer economy. For example, the develop-
ment of cold chain logistics promotes trade flows of the fresh agri-
cultural produce across regions, thus consumers have more choices
for the agricultural produce, which enhances possible residents
consumptions and will boost the prosperity of the fresh agricul-
tural produce market. For the enterprises, the development of
logistics can meet development needs of non-logistics enterprises,
1
lag phase1 refers to the period of previous year (comparing to the current year).
70 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676

such as manufacturing and commercial enterprises, facilitating

0.633
0.409
their resource circulation, which is needed for their business run-

TJ
ning. It can also accelerate the product distribution and expansion,
speed up backflow of capital chain, and improve enterprises profit

0.527
0.710
potential and competitiveness, thus promoting the development of

CQ
the market [13]. Another example, Jingdong self-run logistics mode

0.425
0.698
is designed for Jingdong mall. Compared with the traditional logis-
GZ tics system, this self-run logistics system is better in improving
0.475 transport efficiency thus saving costs, thereby further attracting
0.873
SH

consumers and other companies to follow this pattern. For the

whole economic system, the rapid development of logistics is
0.874
0.490
2013

favorable for effective connectivity of local logistics networks, pro-

BJ

moting increasingly closed relations between the economic sub-

jects, lowering the transaction costs between each other, and
0.342
0.619

effectively avoiding the possibility of conflicts, thus accelerating

TJ

positive interactions among products production, circulation and

0.659
0.517

consumption.
CQ

In addition, the development of logistics can help to expand

0.342
0.690

employment, increase taxes, optimize investment environment,

GZ

and deeply influence the development of social economy. It can

be drawn from Table 7 that considering the lag effect of logistics,
0.534
0.895

logistics development will have a positive impact on economic

SH

development of the following year (Coef. = 0.404904; P = 0.095).

0.624
0.887

This fully demonstrates that the logistics influence on the eco-

2012
BJ

nomic development is continuous. The reason for the lagging is:

under normal circumstances, the influence logistics has on the eco-
0.319
0.600

nomic development does not happen overnight, but accumulates

TJ

progressively.
0.663
0.491

For example, the newly built logistics infrastructure of this year

CQ

may not fully yield economic benefits but in the second year, it
will. From another point of view, the development of logistics is
0.352
0.682

likely to be a barometer for economic development of the follow-

GZ

ing year. To certain extent, it can predict or reflect the economic

0.612
0.927

trends of the second year. In a word, the development of the logis-

SH

tics can stimulate consumption, reduce metropolitan economic

transaction costs, improve efficiency and effectiveness of the eco-
0.643
0.895
2011

nomic activities, and thus promote the economic development.

BJ
0.311
0.549

3.3.2. Metropolitan economic prosperity promotes metropolitan

TJ

logistics development
Metropolitan economic development can provide a more solid
0.684
0.433
CQ

foundation for its logistics development, increase logistics invest-

ment, improve logistics market demand and supply potential,
0.646
0.330

thereby stimulating the development of logistics industry. It is

GZ

shown in Table 4 that the economy exerts a positive influence on

0.562
0.951

logistics development (Coef. = 0.925114; P = 0.003). The reasons

SH

are as follows:
First of all, the development of economy means there exists a
0.585
0.888
2010

huge market for demand. Research shows that commercial activi-

BJ
Development levels of metropolitan logistics and economy.

ties often derive from logistics. The more prosperous its regional
0.279
0.511

economy is, the more developed its manufacturing and business

e
TJ

are, thus it induces huge demands for market for logistics develop-
ment. At the same time, the economic booming drives the
0.621
0.399
d

improvement of residents disposable income, which increases

CQ

the residents consumption. Once it has increased, especially the

0.296
0.618

expansion of cross-regional consumption, which puts forward

c
GZ

higher requirements and standards for logistics, it stimulates logis-

GZ stands for Guangzhou.

tics development and upgrading.

CQ stands for Chongqing.
0.977
0.640

SH stands for Shanghai.

b

Second, economic development means increase of its fiscal rev-

SH

BJ stands for Beijing.

TJ stands for Tianjin.

enue. Therefore, government will invest more on fixed assets of

0.552
0.889
2009

logistics infrastructure, such as highways, railways, and airports.

Year

a
BJ

The perfection of logistics infrastructure is decisive to regional

logistics development [6].
Indicator

Third, metropolitan economic development can promote the

Table 3

gy
f x

development of science and technology. Science and technology

b
a

e
c

are important factors affecting the development of modern logis-

 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676 71

Table 4
Regression analysis of logistics index and economic index.

Logistics Coef. Std. err. t P > |t| [95% Conf. interval]

Economy 0.925114 0.260969 3.54 0.003 0.368872 1.481357
Constant 0.2363 0.225539 1.05 0.311 0.71703 0.244426

Table 5
Regression analysis of logistics index and lag phase1 economic index.

Logistics Coef. Std. err. t P > |t| [95% Conf. interval]

Economy Lag1 3.429349 1.868158 1.84 0.096 0.73317 7.591865
Constant 0.169063 0.421561 0.4 0.697 0.77023 1.108361

Table 6
Regression analysis of economic index and logistics index.

Economy Coef. Std. err. t P > |t| [95% Conf. interval]

Logistics 0.492761 0.139005 3.54 0.003 0.196479 0.789043
Constant 0.58366 0.079794 7.31 0 0.413584 0.753736

Table 7
Regression analysis of economic index and lag phase1 logistics index.

Economy Coef. Std. err. t P > |t| [95% Conf. interval]

Logistics 0.266393 0.136082 1.96 0.079 0.03682 0.569603
Logistics Lag1 0.404904 0.219326 1.85 0.095 0.08378 0.893592
Constant 0.493303 0.140111 3.52 0.006 0.181116 0.805489

tics. Advanced science and technology speed up logistics equip-
ment upgrading and renewal, promote application and utilization
rate of advanced logistics equipment and technology, thus acceler-
ating the transformation and upgrading of logistics industry.
It can be drawn from Table 5 that considering the lag effect of
economic efficiency, the economic prosperity will have a positive
impact on the logistics development of the following year (Coef.
= 3.429349; P = 0.096). It fully shows that logistics construction
affects sustainability of economic development. If the economy
develops well, it means that it may yield larger logistics demands,
further stimulate logistics investment, and thereupon contribute to
next years fast development of the logistics industry.
4. Relation analysis of economic subsystems and logistics
subsystems
Section 3 has conducted the Granger causality test on the rela-
tionship between metropolitan economic and logistics develop-
ment, which shows the mutually influencing relation between
the two. However, it only takes a macro perspective in probing into
the relationship between logistics and economic and lacks an in-
depth analysis of the relationship between the subsystems of econ-
omy and logistics using mathematical formula. Thus this section
explores the relation between economic subsystems and logistics
subsystems from regional economic investment, regional eco-
nomic capacity and strength at a microscopic level to achieve a
systematic and in-depth investigation of relation between
metropolitan logistics and economy. Section 4.1 studies on regio-
nal economic investment, Section 4.2 on the regional economic
capacity, and Section 4.3 on regional economic strength. The three
sections collectively analyze the impacts of metropolitan economy
on metropolitan logistics industry.
4.1. Relationship between regional economic investment and logistics
industry development
From the perspective of industry, investment in fixed assets
includes investment of fixed assets in such logistics industries as
transportation, warehousing and postal service. The investment
in fixed assets is one of the important indicators to measure regio-
nal economic investment. As an important engine of economic
growth, investment in fixed assets is the important means to
increase the social fixed assets, to expand production scale and
to develop the national economy. Along with the increasing of
the total investment volume and the expanding investment
demands, the social total demands will continue to expand,
thereby driving economic growth correspondingly.
Regional economic investment and regional logistics develop-
ment have a relation of mutual influencing and mutual improving.
Therefore, there is a close relation between investment in fixed
assets and the logistics industry development in our country. In
order to verify the improvement of regional economic investment
to the logistics industry development, here this study conducts the
correlation analysis and regression analysis of total investment in
fixed assets and urban road area [19].
Among them, the total investment in fixed assets is the work-
ing volume of activities in the construction and purchase of fixed
assets in monetary terms; it is the comprehensive indicator to
reflect the scale, speed, proportion and use direction of invest-
ment in fixed assets. Urban road area is the important indicator
to measure the development level of the logistics industry.
Assuming that the original hypothesis H0 means there is no cor-
relation between two variables, according to the data in Table 8,
the results of analysis is illustrated in Table 9, the scatter diagram
is shown in Fig. 1.
72 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676

Table 8
Relative data in five metropolises.

Indicator Unit Beijing Shanghai Tianjin

2009 2010 2011 2012 2013 2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Total investment 100 million CNY 4617 5403 5579 6112 6847 5044 5109 4962 5118 5648 4738 6278 7068 7935 9130
in fixed assets
2
Urban road area 10,000 m 9179 9395 9164 13,509 13,884 8609 9299 9481 9717 9932 8357 9159 10,492 11,611 12,440
Indicator Unit Guangzhou Chongqing
2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Total investment 100 million CNY 2660 3264 3412 3758 4447 5214 6689 7473 8736 10,435
in fixed assets
Urban road area 10,000 m2 9502 9731 10,032 10,140 10,241 8953 9931 10,870 11,936 12,723

Table 9 shows that two variables are not zero correlation. So there is a cor-
The results of the correlation coefficients between the total investment in fixed assets relation between regional economic investment and logistics
and urban road.
industry development. Judging from Fig. 1, logistics industry devel-
Total investment in Urban road opment level and regional economic investment development
fixed assets area level among different metropolis are uneven. Economic growth
Total investment in Pearson 1 0.641** depends on investment demands and total investment in fixed
fixed assets Correlation assets is promoted by the development of economy. With the
Sig. (2-tailed) 0.001
advanced technologies and equipment, improvement of infrastruc-
N 25 25
ture, increase of fixed assets investment, basic conditions to
Urban road area Pearson 0.641** 1
improve the present situation of logistics are provided. The pursuit
Correlation
Sig. (2-tailed) 0.001 of modern logistics industry is not a single link, the optimal storage
N 25 25 or transportation, but taking more emphasis on the optimal logis-
** tics system as a whole [4]. The development of modern logistics
Correlation is significant at the 0.01 level (2-tailed).
industry has a positive function and meaning to promote industrial
structure adjustment, transform the development mode, increase
According to Table 9, it can be seen that the correlation coeffi- the investment demands, and raise the level of the national
cient of total investment in fixed assets and urban road area is economy.
0.641, which shows that there is strong correlation, the probability Since there is for most metropolises a conspicuous positive cor-
P value - the coefficient of correlation test is 0.001 approximately. relation relationship between the total investment in fixed assets
Therefore, when the significance level a is 0.05 or 0.01, correlation and urban road area (Pearsons r = 0.641, p < 0.01), this paper
coefficient test of the null hypothesis should be rejected, which makes a linear regression analysis to exam the effects of the

Fig. 1. The simplified scatter diagram of total investment in fixed assets and urban road area in different metropolis.
 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676 73

Table 10 Table 12
The results of the regression analysis between the total investment in fixed assets The results of the correlation coefficients between per capita disposable income of
(Dependent Variable) and urban road area. urban households and total postal business volume.

Model Unstandardized Standardized t Sig. Per capita disposable Total postal

coefficients coefficients income of urban business
households volume
B Std. error beta
Per capita disposable Pearson 1 0.647**
1 (Constant) 2453.018 2088.758 1.174 0.252
income of urban correlation
Urban road area 0.801 0.200 0.641 4.005 0.001
households Sig. (2-tailed) 0.000
N 25 25
Total postal business Pearson 0.647** 1
changes of the urban road area on the total investment in fixed volume correlation
Sig. (2-tailed) 0.000
assets. From Table 10, the regression analysis shows that the total N 25 25
investment in fixed assets is significantly related to the urban road
**
area at 0.01 significance level, the regression coefficient is 0.801. Correlation is significant at the 0.01 level (2-tailed).

Hence the regression equation is as follows:

Total Investment in Fixed Assets 100 million Yuan
0:801  urban road area 10; 000 m2
4.2. Relationship between regional economic capacity and logistics
development
Building up regional economic capacity means noticeable
improvements both in regional living conditions and in the house-
hold consumption levels. It also means that the consumption con-
cept is also changing. In this sense, establishing a consumption
logistics system that is more perfected, convenient, faster with
more safety and efficiency can meet the constantly improving
household consumption level. Besides, the on-going increase in
urban residents disposable income obviously contributes to the
development of the economy, which also plays a promoting role
in the growth of logistics. This disposable income increase also
opens more market room for e-commerce to develop. E-
commerce has been changing modern business models, speeding
up economic reformation. Meanwhile the development of the
Internet technologies provides more technological space for the
application of e-commerce, promoting the further growth of
express logistics demands (see Table 11).
In order to testify the promotive role that regional economy
capacity plays in the development of logistics, this research con-
ducts the correlation analysis between per capita disposable
income of urban households and total postal business volume.
Per capita disposable income of urban households refers to resi-
dents total cash income that covers daily household expenses. This
indicator effectively shows the living conditions of urban residents
and profoundly reflects the capacity of regional economy. Total
postal business volume is also an important indicator of logistic
development and can show the market demand for logistics. Given
the original hypothesis H0 means there is no correlation between
those two variables, then the analysis results are showed in
Table 12, and a scatter diagram is obtained in Fig. 2.
According to Table 12, the correlation coefficient of per capita
disposable income of urban households and postal total business
volume is 0.674, which shows a strong positive correlation
between those two factors and its correlation test probability P-
approximates 0.000. Therefore, when the significance level a is
0.01, it denies the null hypothesis of the correlation test and shows
that the two ensembles are not zero correlation. As a result, there is
a correlation between the per capita disposable income of urban
households and postal total business volume. Fig. 2 shows that
for different cities, the economic capacity is also different. But all
cities show an increase in their economic capacity, which demon-
strates a positive correlation between the economic capacity and
postal business volume and that regional economic capacity grows
as the postal total business volume increases.
All the five cities show strong positive correlation between per
capita disposable income of urban households and total postal
business volume (Pearsons r = 0.647, p < 0.01). This research con-
ducts a further linear regression analysis to exam the effect of
postal total business volume on per capita disposable income of
urban households in the past five years of five cities. From Table 13,
the regression analysis shows that per capita disposable income of
urban households is significantly related to total postal business

Table 11
Relative data in the five metropolises.

Indicator Unit Beijing Shanghai Tianjin

2009 2010 2011 2012 2013 2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Per capita disposable CNY 26,738 29,073 32,903 36,469 40,321 28,838 31,838 36,231 40,188 43,851 21,402 24,293 26,921 29,626 32,294
income of urban
households
Total postal business 100 89 107 130 142 163 195 177 148 191 259 26 34 23 26 29
volume million
CNY
Indicator Unit Guangzhou Chongqing
2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Per capita disposable CNY 27,610 30,658 34,438 38,054 42,049 15,749 17,532 20,250 22,968 25,216
income of urban
households
Total postal business 100 16 20 23 23 126 25 31 26 31 39
volume million
CNY
74 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676

Fig. 2. Scatter diagram on the correlation of per capita disposable income of urban households and total postal business volume in different metropolises.

Table 13
The results of the regression analysis between per capita disposable income of urban households (Dependent Variable) and total postal business volume.

Model Unstandardized coefficients Standardized coefficients t Sig.

B Std. error Beta
1 (Constant) 24551.994 1827.187 13.437 0.000
Total postal business volume 67.513 16.569 0.647 4.075 0.000

volume at 0.01 significance level, the regression coefficient is
67.513.
Hence, the regression equation is as follows:
Per capita disposable income of urban households 100 million Yuan
24551:994 67:513
 postal total business volume 100 million Yuan
4.3. Relationship between regional economic strength and logistic
development
Regional economic strength is the fundamental driving force of
regional logistic development. When the regions economy is
growing, it brings sufficient investments for logistics industry.
The more advanced the regional economy is, the higher standards
it requires for regional logistics. And a strong regional economy
provides a solid economic base and reliable technologies for the
development of logistics industry. The development of regional
economy will certainly be accompanied with greater demands
for logistics, which will in turn stimulate regional logistics to grow
(see Table 14).
In order to testify that the promotive role that regional eco-
nomic strength plays in the development of the logistics industry,
the author conducts correlation analysis between total retail sales
of consumer goods and number of logistic employees. For a given
region, its regional total retail sales of consumer goods reflects to
some degree regional economic strength, and the number of logis-
tic employees involved in logistics can show the prosperity of this
regions logistic market. Given the original hypothesis H0 means
there is no correlation between those two, then the analysis result
is as showed in Table 15, and a scatter diagram is obtained in Fig. 3.
From Table 15, it can be shown that the correlation coefficient
between the total retail sales of consumer goods and the number
of logistic employees is 0.876, which means an obvious positive
correlation between these two factors, with its correlation test
probability P-approximates 0.000. Therefore, when the significance
level a is 0.01, it denies the null hypothesis of the correlation test
and shows that the two indicators are not zero correlation. It is
shown directly in Fig. 3 that there is a correlation between total
retail sales of consumer goods and the number of logistics
employees.
With the economic development, total retail sales of consumer
goods are on the rise, thus exerting an influence on residents con-
sumption concepts and consumption patterns. Consequently,
 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676 75

Table 14
Related data in the five metropolises.

Indicator Unit Beijing Shanghai Tianjin

2009 2010 2011 2012 2013 2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Total retail sales of 100 million 5310 6229 6900 7703 8375 5173 6071 6815 7412 8052 2431 2860 3395 3921 4470
consumer goods CNY
Number of logistic 10 thousand 50.1 51 57.4 57.8 59.2 35.6 36.3 41 38 49.2 12.3 12.5 11.4 14.1 14.3
employees persons
Indicator Unit Guangzhou Chongqing
2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Total retail sales of 100 million 3616 4500 5243 5977 6883 2479 2939 3488 4034 4600
consumer goods CNY
Number of logistic 10 thousand 20.7 20.9 22.9 23 33.5 13.5 14.3 15.1 15.8 26.1
employees s persons

Table 15 the latter including such facilities and equipment as postal routes,
The results of the correlation coefficients between total retail sales of consumer goods
and number of logistic employees correlations.
highways and railway systems. There is no doubt that the rapid
development of logistics will draw more excellent logistic workers
Total retail sales Number of to engage themselves in this industry.
of consumer logistic
goods employees
Since there is a conspicuous positive correlation for the five
cities between total retail sales of consumers goods and number
Total retail sales of Pearson 1 0.876**
of logistic employees (Pearsons r = 0.876, p < 0.01). This research
consumer goods correlation
Sig. (2-tailed) 0.000 conducts a further linear regression analysis to exam the effect of
N 25 25 total number of logistic employees on total retail sales of consumer
Number of logistic Pearson 0.876** 1 goods in the past five years. From Table 16, it can be learned that
employees correlation
total number of logistic employees is significantly related to total
Sig. (2-tailed) 0.000
N 25 25
retail sales of consumer goods at the 0.01 significance level, and
the regression coefficient is 96.119. Hence, the regression equation
**
Correlation is significant at the 0.01 level (2-tailed). is as follows:
Total retail sales of consumer goods 100 million Yuan
higher requirements are put forth on the service quality of logis-
2286:852 96:119
tics, especially the need to perfect logistic platforms. Logistic plat-
forms consist of two parts: software parts and hardware parts,  total number of logistic employees 10 thousand persons

Fig. 3. Scatter diagram on the correlation between total retail sales of consumers goods and number of logistic employees in different cities.
76 S. Lan et al. / Advanced Engineering Informatics 32 (2017) 6676
Table 16
The results of the regression analysis between total retail sales of consumers goods (Dependent Variable) and total number of logistic employees.
Model Unstandardized coefficients
B Std. Error
1 (Constant) 2286.852
Number of logistic employees 96.119
5. Concluding remarks
In order to validate the interactive relationships between the
development of metropolitan economy and logistics, correlation
and regression analysis are employed. Based on the data of the
above five metropolises, and three economic forms, which are
regional economic investment, economic capacity and economic
strength, this research conducts correlation analysis and regression
analysis on urban road, total postal business volume and workforce
in logistics industry respectively so as to explore the in-depth rela-
tion between the development of metropolitan economy and logis-
tics. It demonstrates that the economic growth and the logistics
development in metropolises mutually influence each other. The
exact equations between sub-systems of metropolitan economy
and sub-systems of metropolitan logistics are acquired by conduct-
ing correlation and regression analyses. This can potentially be
explained in the following two aspects:
On the one hand, as urban economy develops, various com-
modities and services and huge volume of information builds up.
Total retail sales of consumer goods also increase, together with
accumulating circulations of commodities. It is certain that all
these factors will generate enormous demands on logistics, practi-
cally and potentially, which will provide a big space for trial logis-
tic services and promote further development of logistics. With full
utilization of the economic strength, the favorable economic devel-
opment in metropolises provides a solid base to develop logistics,
bolstering logistic infrastructure development and industrial
scales, improving logistic service quality, and keeping the rapid
growth of the whole industry.
On the other hand, regional logistics is an important catalyst for
regional economic development [20]. Logistics industry is increas-
ingly important for serving and supporting regional economy. The
development of modern logistics can drive economic growth,
which then generates more investment in economy, particularly
in fixed assets such as logistics infrastructure and industrial tech-
nology platform. As a result, the service of modern logistics will
be improved. Logistics mainly consists of multiple nodes and lines.
Booming in logistics industry will reduce the market transaction
cost and integrate different resources of logistics industry, which
then provides guarantee for the improvement of regional economic
capacity. The postal industry, which has flourished with the devel-
opment of e-commerce, promotes the growth of information
industry and economic restructuring, leading a decrease in human
resources cost and trade cost. This will affect the operation of the
social and economic system and become an important force in eco-
nomic strength that cannot be overlooked.
From this study, we can know that the development of
metropolitan economy and logistics are positively correlated
(through the calculated co-efficient and P-value). Metropolises
should actively develop the logistics industry if there is a gap in
logistics development. In this way, they can enhance coordinated
development and ultimately promote fast growth of metropolitan
economy.
Standardized coefficients t Sig.
Beta
374.429 6.108 0.000
11.030 0.876 8.714 0.000
References
[1] M.B. Arvin, R.P. Pradhan, N.R. Norman, Transportation intensity, urbanization,
economic growth, and CO2 emissions in the G-20 countries, Utilities Policy 35
(2015) 5066.
[2] M. Avellaneda, Minimum-relative-entropy calibration of asset-pricing models,
Int. J. Theoret. Appl. Finan. 1 (04) (1998) 447472.
[3] C.F. Baum, An Introduction to Modern Econometrics Using Stata, Stata Press,
2006.
[4] B.M. Beamon, Measuring supply chain performance, Int. j. Operat. Product.
Manage. 19 (3) (1999) 275292.
[5] R. Carruthers, J.N. Bajpai, D. Hummels, Trade and logistics: an East Asian
perspective, East Asia integrates: a trade policy agenda for shared growth
(2003) 117137.
[6] J.-K.J. Cho, Firm Performance in the E-Commerce Market: The Role of Logistics
Capabilities and Logistics Outsourcing, University of Arkansas, 2001.
[7] L. Dablanc, D. Diziain, H. Levifve, Urban freight consultations in the Paris
region, Euro. Transp. Res. Rev. 3 (1) (2011) 4757.
[8] K.G. Debbage, Air transportation and urban-economic restructuring:
competitive advantage in the US Carolinas, J. Air Transp. Manage. 5 (4)
(1999) 211221.
[9] O. Fisher, The planning of regional logistics, J. Melbourne Univ. 4 (2002) 102
122.
[10] M. Fujita, T. Mori, The role of ports in the making of major cities: self-
agglomeration and hub-effect, J. Dev. Econ. 49 (1) (1996) 93120.
[11] K.S. Gandlur, Implementation of Adaptive Routing in Public Logistics
Networks, 2002.
[12] H. Ham, T.J. Kim, D. Boyce, Assessment of economic impacts from unexpected
events with an interregional commodity flow and multimodal transportation
network model, Transport. Res. Part A: Pol. Pract. 39 (10) (2005) 849860.
[13] D. Kisperska-Moron, Logistics change during the transition period in the Polish
economy, Int. J. Prod. Econ. 35 (13) (1994) 2328.
[14] S. Lan, R.Y. Zhong, Coordinated development between metropolitan economy
and logistics for sustainability, Resour., Conserv. Recycl. (2016), http://dx.doi.
org/10.1016/j.resconrec.2016.08.017, in press.
[15] H.H. Lean, W. Huang, J. Hong, Logistics and economic development: experience
from China, Transp. Policy 32 (2014) 96104.
[16] H. Lee, H.M. Yang, Strategies for a global logistics and economic hub: Incheon
International Airport, J. Air Transp. Manage. 9 (2) (2003) 113121.
[17] B.P. Loo, B. Hook, Interplay of international, national and local factors in
shaping container port development: a case study of Hong Kong, Transp. Rev.
22 (2) (2002) 219245.
[18] M.F. Melndez, The Logistics and Transportation Problems of Latin American
Integration Efforts: The Andean Pact: A Case of Study, 2001.
[19] C.A. Parvin, An introduction to multivariate statistical analysis, T.W. Anderson.
Hoboken, NJ: John Wiley & Sons, 2003, 742 pp., $99.95, hardcover. ISBN 0-471-
36091-0, Clin. Chem. 50 (5) (2004) 981982.
[20] Y. Qiu, H. Lu, H. Wang, Prediction method for regional logistics, Tsinghua Sci.
Technol. 13 (5) (2008) 660668.
[21] R.M. Reesor, Relative Entropy, Distortion, The Bootstrap and Risk, University of
Waterloo, 2001.
[22] R.E. Schuler, W.L. Holahan, Competition vs. vertical integration of
transportation and production in a spatial economy, Pap. Reg. Sci. 41 (1)
(1978) 209225.
[23] T. Skjtt-Larsen, U. Paulsson, S. Wandel, Logistics in the resund region after
the bridge, Eur. J. Oper. Res. 144 (2) (2003) 247256.
[24] W. Talley, Linkages between transportation infrastructure investment and
economic production, Logist. Transport. Rev. 32 (1) (1996) 145.
[25] E. Taniguchi, M. Noritake, T. Yamada, T. Izumitani, Optimal size and location
planning of public logistics terminals, Transport. Res. Part E: Logist. Transport.
Rev. 35 (3) (1999) 207222.
[26] J. Visser, A. Van Binsbergen, T. Nemoto, Urban freight transport policy and
planning, City Logist. I (1999) 3969.
[27] F. Wiengarten, M. Pagell, M.U. Ahmed, C. Gimenez, Do a countrys logistical
capabilities moderate the external integration performance relationship?, J
Operat. Manage. 32 (1) (2014) 5163.