Journal of Geographic Information System, 2010, 2, 106-112
doi:10.4236/jgis.2010.22016 Published Online April 2010 (http://www.SciRP.org/journal/jgis)
Design of Data Model for Urban Transport GIS
Wen Zeng1,2, Xiaojie Chang2, Jianjun Lv1,2
1Faculty of Information Engineering, China University of Geosciences, Wuhan, China
2Zondy Cyber T&S Co., Ltd, Wuhan, China
Abstract
Constructing the data model for public transportation by integrating the spatial and the non-spatial informa-
tion, is the basis of reasonable plan and effective management of urban public transport. This paper presents
a transit data model based on geographic information systems (GIS) technology, which utilizes arc-node
networks, and manages the foundational bus data with point, link, polygon and record features. In this model,
a transport network is generated and maintained in a dynamic manner, and hence supports planning, con-
struction, management, operation and optimization functions for transit facilities and routes, as well as
day-to-day transactions. Public transportation GIS established on this model foundation will remarkably up-
grade the construction level and the urban service ability.
Keywords: Geographic Information System (GIS), Transit GIS, Transit Route Network, Data Mode
1. Introduction
With the rapid urban development, the population and
vehicles increase faster and faster which bring more and
more pressure for urban traffic. Improving public trans-
portation is the key to solve this problem. To improve
service quality and efficiency of bus, subway and other
public transportation, it’s urgent to promote pubic trans-
portation informationization. From the national applica-
tion trend, GIS has been the most effective method to
foster public transportation informationization develop-
ment. [1,2].
Transportation GIS which integrates spatial and non-
spatial information, integrates computer network, spatial
database technologies which supports map surveying,
data surveying, real-time acquisition , spatial analysis and
expert knowledge base. It can manage and assess the
urban roads, public transportation network, the station
facilities, basic data and the bus operator information
which realizes vari o us business functi o ns.
Urban public transport system based on GIS will dis-
play the overall distribution of urban public transporta-
tion which achieves basic GIS functions, such as, add,
modify, delete, query and browse related attribute of
basic topographic graphics, road, station grounds, bus
lines and operation data. What’s more, it can conduct
analyze and assess public transport facilities and routes
reasonable layout based on the evaluation system and the
knowledge base. GIS data model organizes data in dif-
ferent basic operation unit (such as point, line, surface,
net, etc.), according to spatial entities and non-spatial
information characteristics and describes entities rela-
tions by defining topology, relationships and constraints
[3,4]. Public transport information is various which
needs to be appropriate organized. Data model design is
the basis for Public transportation research and develop-
ment.
This paper presents public transport data mo del which
can establish and create routes in real time, separate road
layer and route layer, and separately manage route layer.
The generation of route is based on site data and site
routes. Public transport route generation has the several
steps, such as, generating route site by site mapping the
centerline and seeking the shortest path by centerline
adjacent sites. It will improve the route accuracy and
facilitate route analysis.
2. Management Object Analysis
Urban public transport managed objects, includes physi-
cal entities and data entities. Physical entities refer to the
road network, public transport facilities, ro utes, vehicles,
passenger traffic distribution and other objective object.
Fund Project: Research and demonstration of key technology of urban
municipal infrastructure management and operation, Key projects o
f
science and technology subjects, National Eleventh Five-Year Plan,
(No.2006BAJ15B03).

W. ZENG ET AL.107
Data entities are mainly public transport daily operation
and management data, such as travel schedule, IC card
credit card data, traffic statistics, sales data, and person-
nel management data.
Road network is the basis for all traffic information
system and road network modeling is widely discussed.
The actual model includes the arc-node data model (plane
or non-flat), along the driving direction model, along
driveway model, etc. Data entities are generally in the
form of database table and every table are correlated by
keyword. This paper mainly studies the public transpor-
tation facilities, lines, modeling, and operation data such
as, operations and eva luatio n d ata.
2.1. Public Transport Facilities
1) Platform: Bus station is set on both sides in pairs. Ac-
cording to actual passenger traffic and geological loca-
tion, it can be divided into three types, hub, the starting
and ending, and common site. The hub is usually in a
downtown with heavy traffic and large population wh ich
has great impact on urban management and operation.
2) Station: Mainly refers to parking lot and maintenance
place. Parking lot is used to park buses which need to
manage and dispatch buses according to the number of
vehicles and starting-ending time.
2.2. Routes
Urban public transport bus route is the most important
part of urban transportation which refers to the directed
physical path [5,6] of every bus route. Bus lines are di-
vided into one-way line and ring lines. One-way line con -
tains uplink and downlink lines. The round-trip journey
can be regarded as different paths. According to fixed
order and the passed sites, determine the spatial location
and morphology of bus route [5,6].
In data model, one line is alternatively connected by
site and road section.
1) Site: Site is location for passengers getting on and
off the bus. Actually, site is the “projection” of platform
on one specific line and its location is according to the
corresponding site. In fact, different buses lines may ove r-
lap and multi-lines may pass the same site. In other wo rds ,
the platform and site has one-to-many relation.
2) Road sections: Section is a line segment between
adjacent sites which is the basic building blocks unit of
bus routes. Specifically, section is the connection of two
adjacent sites along the road centerline.
The following diagram visually reflects the relation-
ship between buses related objects:
Figure 1. Objects relations of public transportation.
2.3. Operation Information
Public transport operation information is bus GIS infor-
mation which is closely connected with the application
business data including the frequency plan, passenger
traffic and passenger revenue and other data. It organizes
public transportation information.
The actual performance of public transport operations
is reflected in many aspects, such as vehicle storage and
organization, each line transceiver classes, schedules,
operating interval, operation mileage management, lines
traffic, operation income, etc. Related information and
route will change accordingly.
2.4. Evaluation Index
Urban bus routes operation should be standardized by
general regulations and supports. The macro index of
public transportation is main involved in seven factors,
the whole length of operation n etwork, veh icle admission
rates, bus travel proportion, bus site coverage rate, trans-
fer coefficient, line density and line repeat coefficient.
These factors change with time periods which reflect the
overall condition of urban public transportation distribu-
tion density, service area, public travel convenience, etc.
3. Data Model Design
3.1. Overall Model
Public transport Information lies in urban infrastructure
and urban road network data. You can see the organiza-
tion correlations public transportation models in Figure
2. In the model, the site data are abstracted from the ac-
tual platform which is taken as node in lines. Each site
has one unique platform which recorded by its spatial
geological location and expressed by point feature. Hub

W. ZENG ET AL. 108
Figure 2. Data model of urban transportation.
site corresponding station is the platform data which has
relatively complex attribute information. It’s usually ex-
pressed in polygon element. All the sites corresponding
to the platform passed by bus is closely related to site
location. Parking and maintenance field is relatively in-
dependent which can be individually expressed.
3.2. Data Representation of Public Transportation
Management Object
Data contents in public transportation management con-
sists four parts, public transportation station, lines, line
network and subsidiary business data. Station and line
data in bus actual application has two major categories,
the current data an d planning da ta.
3.2.1. Publ ic Transp ortation S tation Data
Bus station data contains site and station two parts. The
subsidiary attributes information of station and bus hub
site are relatively complex which record the number of
vehicles and their location and covered area. It will help
vehicles management and bus operation analysis and rep-
resented in polygon features. The first, last station and
the halfway stops are relatively simple which is the con-
nection of every routes. It only relates to the p assed lines
and represented in points feature. It helps recording the
spatial location of entities.
3.2.2. Public T ransportation Route Data
Every bus route data not only record all the passed sites
and relations, it also records large number of public tra-
nsportation professional data. It will directly affect the
operation analysis and comprehensive evaluation of pub-
lic transportation system. Therefore, this model uses da-
tabase table to store data. Actually, public transportation
has uplink and downlink lines. The passed sites are not
completely the same, and some lines are even single lin e
(loop). This model takes every line as one closed loop,
and records every passed site number of the uplink and
downlink lines.
3.2.3. Public Transportation Line Network Data
Public transportation network does not exist in practice.
It’s only the correlation record of point entity coord in ates
of current site data and current site and lines. Then gen-
erate route network information, which record the topo-
logical relations of site and routes. Therefore, it’s stored
in network element structure.
Single-line network generation diagram is as follows.
The virtual circuit displays the passed platforms of lines
vividly. The uplink and downlink line form a closed cir-
cuit, which doesn’t exist in practice. Actually, the bus
routes are overlapped by the road centerline. Therefore,
in order to clearly d isplay the uplink and downlink lines,
migration processing has been done. In the process of
line generation, the platform is mapped to the centerline
sequentially. Then seeks the sh ortest path of the adjacent
sites, such as add control point at the corner (Shown in
Figure midpoint bc1 and bc2), connect the generated site,
thus finally get the route.
3.2.4. Public Tr ansportation Subsidiar y Data
Subsidiary data main contains operation information and
macro information. Operational data includes public tran-
sportation IC credit card data, distribution data of bus
traffic, distribution data of bus traffic, and the existing

W. ZENG ET AL.109
public transportation network data. Macro data, includes
bus number, the proportion of bus travel, public trans-
portation network density, public transportation non-
linear co ef f ic ien t, bu s lin e s cov er ag e , tr an sf er co ef f ici en t,
etc. A large number of attribute information can be s tored
in operation information and macro information database
table.
3.3. Correlation
As for public bus transportation data, site and route are
closely related, which is also the base for bus network.
Other data are relatively independent. This data model
proposes the following correlations to ensure the ration-
ality and integrity of line network data.
1) Route and site, in one route there are many sites, the
current bus lines and current site form one-to- many rela-
tions.
2) Routs and line network, site route table records all
the site information of every passed bus, and thus gener-
ate a bus route which is the base for the urban bus net-
work. The bus route and line network form many-to-one
relationship.
3) Route and single-line, urban bus routes constitute
bus line network. Each bus route is corresponds to one
line in the network.
4) Site and platform, there are many lines in each plat-
form which corresponds to site of each line. Platform and
site has form one-to-many relationship.
5) Operation information and routes: Operation infor-
mation bases on specific routes. The operation line mile-
age, passenger traffic, passenger data and traffic income
are different for the same destination in different time
periods.
3.4. Line Network Generation Process
From Figure 3 we ca n see the single l ine network ge neration
process. Figure 4 flow chart describes the specific con-
cepts and algorithms of line network generation. (The
following are the description of the chart)
4. Data Model Realization
4.1. Database Organization
Based on data model design, bus related basic data, site
field data of parking, maintenance field and hub station
information, planning site and line data, and independent
macro information need to record the information sepa-
rately. Operation information is separately stor ed while it
connected with lines. Current site and line data are cor-
related. All the above data form a transportatio n network.
1) Current site data: Record site number, name, loca-
tion, type, and their respective sections and site. It also
records the area, parking area and construction area.
2) Current line data: It records route designation, dock
site in turn, first and last bus time and bus intervals. It
also records various attribute information, such as, line
length, total number of line vehicles, ticketing form and
full line rate, average transport distance and non-linear
coefficient.
3) Current site lines information: It records the corre-
lation of each line and its corresponding site. One line
record relates to multi site record. In addition, it con tains
the route length and non-linear coefficient. This table
records and stores the correlation of sites and routes.
4) Public transportation network: It records designa-
tion of each route and all sites number and geometric
information on this route. The network file is formed by
the relations of site and route of site data and site line
table. It records the geospatial location and topological
relations of all the bus routes.
5) Station data: Data of parking and maintenance field
and repair shop is relatively independent. It records spa-
tial and attribute in polygon file.
Figure 3. Single-line networ k ge ne r a tion diagram.

W. ZENG ET AL. 110
Figure 4. Flow chart of arithmetic.
6) Subsidiary data: Bus operational information and
macro information store the bus operation related attrib-
utes. Operational information associates with the route
and time. The file records the lien destin ation and date.
4.2. Correlation Realization
In accordance with the database organization , some of bus
data are independent and some are interrelated. You can
see vividly about the correlation of data from Figure 5. In
Figure 5, current bus sites and current lines are corre-
lated through site number and ID. The current site file
and bus line file are connected by site number field.
Current bus line file and current site file are connected
by line designation field. Bus line net file and site line
file are correlated through destinati on fiel d. Bus operati on
Figure 5. Correlations among the data tables in data model.

W. ZENG ET AL. 111
information and lines are connected by line destination.
The macro-evaluation data are independent from site fie ld
data.
4.3. Database Realization
4.3.1. Dat a En try
Bus station site data and route data are the basis of bus
model and the system. The station site, line and data of
every site will be input in database provided by user.
Passenger traffic distribution data and site information
can be input in table file in establishing lib, or by system.
The historical data of operational information and macro-
evaluation information can be directly manually input to
the database and the new records can be added through
the system functions.
4.3.2. Real Time Generation
Bus lines are the most important data and the lines are-
generated on the site route record. Site ro ute file sequen-
tially records corresponding relationship of one line and
the sites that it passed which is generated by the line se-
quentially passed site information record. One line re-
lates to multi-records. According to the line record, the
bus transport number and passed site number, and search
relative site spatial information according to site number
in site files, record site location, and connect all the re-
cords sequentially to generate a route. At the same time,
every single line is saved in bus line net file and all the
single line constitutes th e whole bus network.
The following diagram shows the final outcome of gen-
eration one single line net. The site name is marked on t he
line in turn, and the uplink and downlink line is marked
in different color.
Figure 6. Result generation of single linear network.

W. ZENG ET AL. 112
5. Conclusions
Urban public transport is the public welfare undertaking s
for urban economic development and people’s life. It also
represents the degree of municipal construction moderni-
zation. This paper introduced a new transportation data
model which can establish and maintain route network in
real time. The main features are embodied as follows,
separation of road layer and route layer separation, and
route display doesn’t rely on road net, route site gener-
ated by the mapping of station site to centerline, the
geological location of route are more accurate, line gen-
eration algorithm is independent of any line, on ly related
to site location and road conditions, so user can add, de-
lete, and modify the route. In addition to basic GIS data
processing function, the system store and manage public
transport site data, operation and macro-information w h ich
can effectively organize urban public transportation plan-
ning and information management. The system has good
expansibility, which will better serve urban public trans-
portation GIS and foster urban development.
6. References
[1] W. Z. Xiao, W. Wang, X. G. Li, et al., “Application of
Urban Public Transportation Planning Information Sys-
tem Based on GIS,” China Geographic Information Sys-
tem Association Essays, 1999, pp. 1001-1005.
[2] T. Huang, “Study on Design and Application of Urban
Communication Geographic Information System,” Sanjin
Surveying, Vol. 9, No. 3-4, 2002, pp. 63-66, 78.
[3] J. Chen and S. W. Zhang, “The Third Generation GIS
Data Model and Its Realization,” Sanjin Surveying, Vol.
10, No. 2, 2003, pp. 7-9.
[4] W. Zen, “GIS Software Project,” China University of
Geosciences Press, Wuhan, 2002, pp. 34-40.
[5] J. T. Li and J. F. Yang, “Public Transport Network Model
and Route Query Application Based on GIS,” Journal of
Dalian Railway Institute, Vol. 25, No. 2, 2004, pp. 30-33,
95.
[6] H. Wen, Y. F. Liu and J. H. Zheng, “Public Transport
Network Data Model Research Based on Time Chain,”
Geology and Geological Information Science, Vol. 21,
No. 3, 2005, pp. 35-38.
[7] Y. Li and Z. D. Huang, “Multi-Layer Public Transporta-
tion Site Model Based on UML,” Communication and
Computer, Vol. 24, No. 133, 2006, pp. 13-17.
[8] Z. Hu, F. Y. Zhang and S. L. Liu, “Research and Algo-
rithm Realization on Urban Bus Transfer Data Model,”
Telecom Network Technology, Vol. 4, 2007, pp. 71-74.
[9] X. C. Wu, “Geological Information System Design and
Realization,” Electronics Industry Press, Beijing, 2002,
pp. 91-112.
[10] E. Geneidy and A. Medhat, “The Use of Advanced Infor-
mation Technology in Urban Public Transportation Sys-
tems: An Evaluation of Bus Stop Consolidati on,” Disserta-
tion Abstracts International, Vol. 66, No. 3, p. 1190.