RESEARCH ON GRID-BASED MOBILE BUSINESS PROCESS
AND SIMULATION
Dan Chang and Li Si
School of Economics and Management, Beijing Jiaotong University, Beijing, P.R. China
Keywords: Grid management, Mobile grid, Mobile business process.
Abstract: Since the emergence of mobile commerce, there have been much research and practice on how to improve
wireless communication technology and safety technology and so on (Hull, 1997), however, the research
which integrated wireless technology and business processes of the original e-commerce is still in its early
stage, lacking of systematic analysis and theoretical support regarding the information sharing, business
collaboration, and effectively access of mobile devices in practice. In this paper, mobile business processes
is the research object. On the basis of combing and analyzing the current mobile business process, utilizing
the grid management theory construct mobile business process based on grid. Furthermore, a quantitative
simulation will be made on non-grid and grid-based mobile business processes in order to prove the
superiority of mobile business processes based on grid.
1 INTRODUCTION
Currently, mobile business is receiving more and
more attention with the features of flexibility,
convenience and people are eager to enjoy the
convenience of mobile commerce. With further
research, it was discovered that the business process
can reflect the integrated features of wireless
information technology and traditional business. In
the research of mobile business theory and method,
mobile business process is the backbone throughout.
Utilizing the modern theory of process reengineering
and internal logic of mobile business to optimize
mobile business process, in order to enhance the
capacity of mobile commerce services, and resource
utilization levels has become the focal point of
current mobile business research.
The research of mobile business research is still
scattered on this stage, lacking an overall research
and modeling approach. The lack of theory will
delay the development process of mobile business
processes and increase the cost of development
(Taudes et al., 2000). Based on recent research,
grid-based management is an effective method to
solve the problems of resource sharing, collaborative
work in the whole system, meanwhile, the latest
finding-mobile grid is the expansion of traditional
grid in the wireless computing environment, it can
also effectively solve problems of mobile devices
access, offering a solution for overall optimization,
resource sharing, efficient management, wireless
access and other issues in the mobile business
process (Varshney et al., 2000) .As it has a
quantitative mathematical description, so the
logistics flow, information flow and operation flow
in the grid management system can be quantitatively
calculated, achieving a transmission of qualitative
analysis to quantitative analysis. Besides, this mode
can also use simulation for quantitative analysis,
which reflects the superiority of grid management
methods.
In this paper, with mobile business process as the
research object and grid management as the research
theory, combing the computer simulation, conduct a
research of mobile business process based on grid
management. Following is the research ideas. Firstly,
analyze the existing mobile business process, finding
the problems. Then construct a grid-based mobile
business process in the support of grid management
theory. Finally, conduct a simulation comparison
between grid-based and non-grid based mobile
business process.
86
Chang D. and Si L..
RESEARCH ON GRID-BASED MOBILE BUSINESS PROCESS AND SIMULATION.
DOI: 10.5220/0003550200860093
In Proceedings of the 13th International Conference on Enterprise Information Systems (ICEIS-2011), pages 86-93
ISBN: 978-989-8425-56-0
Copyright
c
2011 SCITEPRESS (Science and Technology Publications, Lda.)
2 THEORY
2.1 Grid Management
1Content of Grid Management
Grid management is the integration of a variety of
distributed resources of the community to achieve
the goal of efficient, economical, fast and convenient
management for the purpose of efficient
transmission of information to save materials and
energy and enhance the efficiency of resources.
Huanchen Wang proposed grid management as
an emerging management mode to deal with the
current complex management issues; this idea is
based on the grid boundaries in the chosen system to
achieve information integration, operational
synergies, and combined slice of modern style of a
management network system. In the view of system
structure, grid management should have the features
of grid layout. In the view of resource association, it
should have the feature of resource sharing. In the
view of the operation of the system, it should have
the feature of operate cooperatively and presses
orderly (Wan et al., 2007). In the structure and
operation grid management, it involves at least five
essential factors, namely material, energy,
information, space and time, referred to as MEIST.
2Basic Process of Grid Management
Fig. 1 illustrates the basic process of grid
management that can be abstracted into the
following aspects.
(1) Business acceptance: Accept all classified
business within the network.
(2).Business dispatch: Standardized assessment
and operational coordination, after the approval of
the command center, the segmentation of tasks will
go to the corresponding functional department.
(3) Business process: Functional department
accept business task, organizing work tasks to meet
demand.
(4). Service delivery: The completed task will
turned back to the reception center, by passing to the
users.
(5).supervision and inspection: Monitoring center
compared the completion of verification of service
and user demand.
(6) Disclosure of Information: During the entire
business process, users can query the real-time
business process (Zhen et al., 2005).
Figure 1: Basic process of grid management.
2.2 Mobile Grid
With the rapid development of mobile wireless
network system, the user can access to global
network resources at any place and any time, which
means grid system should also take mobile nodes
into consideration, and this combination leads to
mobile grid computing. It supports mobile users and
resources in a seamless, transparent, safe and
effective way; it is an integration of wireless
technology and girds computing. Mobile grid
integrates mobile devices into grid, becoming a new
type of resource sharing network.
Currently, most grid structures and algorithms do
not take mobile devices into account as mobile
devices have not been considered as a legitimate grid
computing resources or interfaces. There two
methods to locate mobile devices in the grid
computing environment. The first one: as the as the
interface to interact with the grid system, the user
can require service via mobile devices from grid
system to complete tasks, you can perform remote
monitoring and obtain required results from the grid
system. The other one: as the computing resources of
grid system, mobile devices involves in grid
computing task, not just the recipient of grid service.
Therefore, effectively embedded into the grid, the
mobile device can not only work as a recipient of the
grid service but also can be used as grid service
providers. Fig. 2 is the designed mobile grid
computing system structure.
The grid system is divided into three parts: static
grid sites, mobile devices group and the gateway
which connects static and mobile resources. Mobile
devices can be connected to Internet via WLAN or
remote cellular network. On the side of mobile
devices and gateways, mobile agents and service
agents are required. When a mobile user submits a
task to obtain the service of grid computing service,
RESEARCH ON GRID-BASED MOBILE BUSINESS PROCESS AND SIMULATION
87
it interacted by the mobile agents in the mobile
device and gateway first and then submitted to grid
system by mobile agent in the gateway (Wang and
Wang, 2003).
Figure 2: Mobile grid system structure.
2.3 Arena Simulation Software
Arena was visualized interactive integration
simulation software developed by System Modeling
Company; it provides a good organic integration of
computer simulation and visualization. The
theoretical basis of Arena is discrete event
simulation theory, using the event scheduling
method. The basic idea is to use the view of events
to analyze the real system, by defining the events
and the change to the system when events occur. The
mobile business process is a need arrived, processed
and finished process, in line with the principle f
discrete event simulation; accordingly, Arena can be
used for quantitative simulation research.
3 EXISTING MOBILE BUSINESS
PROCESS ANALYSIS
3.1 The Main Process of Existing
Mobile Business
Mobile business works in the Internet open network
environment, based on the application of mobile
communication devices, realizing the B2B and B2C
transactions, a new business operation mode of
mobile payment. As a new economic form, mobile
business is co-created by network economy and
modern logistics. Fig. 3 illustrates the main dynamic
transaction flow.
Figure 3: The main process of mobile business.
1Once the user has need, he can use mobile
devices to login on mobile business information
platform to search for information of targeted
products. However, as each platform assigns
different access rights to different users, so resource
and information cannot be shared in a certain format.
2After obtaining the required information of
targeted products, users still need to login on
different platforms to compare the product
information, in order to find the satisfied product and
fill orders.
3Business confirms the order with customer,
informing the payment, preparing the product and
organizing the logistic delivery.
4Customer receives the product, after accept-
ance, paying the payment via mobile devices,
transaction finished.
3.2 The Shortcomings of Existing
Mobile Business Process
The application of information technology greatly
contributed to the development of mobile business;
however, information technology alone cannot
achieve a holistic approach to mobile business
reorganization of the participants. The shortcomings
of existing mobile business process are as follows:
1Mobile operators do not fully play his role as a
bond. Compared with other market participants, the
mobile operator takes a control role and dominant
position by holding the network resource and
customer base; it can affect the development trend in
depth. However, its advantages do not play an
effective role in a series of cooperation (Ma, 2008).
2Inadequate cooperation among mobile busi-
ness participants. As a new business operating model,
the greatest strength of mobile business lies in its
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
88
integration of a range of social resources using latest
mobile communication technology and tools, serving
for end users and realizing the maximization of
social benefits. However, the lack of cooperation
within mobile business participants inevitably leads
to information asymmetry and low efficiency in
transaction.
3Within the mobile business domain resources
distributed unevenly. In some small enterprises, the
lack of resources hindered the development of
mobile business, whereas, some large enterprises
have vacancy resources which results in a great
waste.
4Internal and external information Island phe-
nomenon exists. Due to the unsuccessful integration
of enterprise e-commerce systems, a great sum of
resources on the network cannot be fully shared.
5The whole mobile business field is not well
integrated, so that members cannot update
information dynamically and efficiently.
3.3 Feasibility Analysis
Grid management provides effective solutions for
the problems in mobile business. Grid broke the
previous technical limitations, so that people can use
a new, freer and more convenient way to utilize
resources, which brings following benefits:
1The emergence of mobile grid technology
makes the application of grid management into the
research of mobile business process possible. It
supports mobile users and resources in a seamless,
transparent, safe and effective way, an integration of
wireless technology and grid computing.
2Grid management integrates varied application
systems together into a unified platform, enables
enterprises to fully share resources and business
process, breaking the bottleneck of information
exchange between enterprises.
3Grid management breaks the restriction that
sharing and collaboration is only limited to data
transmission, it enables resources sharing can reach a
remote operate and control level.
4. Grid messaging service mechanism can
facilitate information exchange between mobile
business systems in a timely manner, making the
communication between businesses and customers
more flexible.
4 CONSTRUCTION OF
GRID-BASED MOBILE
BUSINESS PROCESS MODEL
4.1 Basic Ideas of Construction of
Grid-based Mobile Business
Process Model
The general concept based on the problem analysis
of mobile business process, fully using the existing
grid management and mobile grid research results to
construct a grid-based mobile business process,
achieving the goals of good service, efficient
management and effective regulatory of mobile
business process.
This paper will utilize the research result of grid
management, redefining the job roles of participants,
responsibility and right norms, and business process
within the current mobile business system, in order
to build a basic mobile business process based on
grid management, to form a unified operation of
business process, and to provide grid end-user an
integrated service platform.
4.2 Basic Model of Grid-based Mobile
Business Process
Grid-based mobile business process is based on the
principles of “unified service, resources sharing, and
business collaboration” of grid management to
determine process links, basic data and service port
sharing mode. Accepting the demand in a unified
way for task decomposition, and then distribute it
into various department foe processing. The
grid-based mobile business process can be abstracted
into a unified mode: demand starts order
accepted goods delivery mobile payment
evaluation and feedback see as the Fig. 4
demonstrate.
Further refinement of the steps:
1. Login on the unified platform. End-users login on
the unified platform, sending requirements through
the business platform, the requirements will be sent
to the grid computing service end through the mobile
agents in the mobile device and gateway. After the
unified processing in the grid computing site,
end-users can get useful information through the
interaction between mobile agents in the mobile
device and gateway.
2. Analyze product information. By comparing the
goods
information offered and communicating with
RESEARCH ON GRID-BASED MOBILE BUSINESS PROCESS AND SIMULATION
89
Figure 4: Grid-Based Mobile Business Process.
with businessman online, customer gets a good
understanding of goods.
3. Place an order. End-user fills the order bill,
including price, quantity, delivery address and
contact information.
4. End-user pays for the goods to a third party
payment platform. End-user should pay for the
goods within a certain period of time, or the order
will be cancelled. The third party payment system
ensures the security of transaction by transferring the
payment to businessman only after the customer has
received the goods.
5. Receive payment and notify businessman. The
third party payment system will automatically send a
notification to businessman after receiving the
payment from customer.
6. Dispatch. Businessman prepares the goods and
chooses a logistics company for delivery.
7. Delivery. The third party logistics company
delivers the goods to customer.
8. Receive the goods and confirm the payment.
Customer checks and receives the goods, confirming
to transfer the payment to businessman.
9. Transfer. The third party payment system transfers
the payment to businessman after receiving the
confirmation from customer.
10. Businessman receives the payment.
11. Evaluation and feedback. Customer evaluates the
goods, service quality and logistics quality or gives
suggestion for improvement.
5 COMPARISON OF
GRID-BASED AND NON-GRID
MOBILE BUSINESS PROCESS
SIMULATION
5.1 Non-grid Process Simulation
Model I
1Simulation Model Description of Model I
Fig. 3 illustrates the existing mobile business process,
namely the non-grid mobile business process. The
specific description of the process: Suppose a mobile
business platform provides
X
x
=1,…,
X
categories of service for the public, each kind of
resource is available 8 hours a day, 5 days a week
and users can access and submit requirement to the
24-hour system. When the mobile user needs to get
the No. x business service, it is required to login on n
(
i
=1, …,
n
) business websites to search for useful
information. As to users, it means they have to login
on many distinct platforms until they get the wanted
information. For simplicity, in this paper, we assume
that users give up searching for information if they
cannot find the required information after three times
of attempt.
2Build the Logical Simulation Model
For the convenience of description, in this paper
non-grid mobile business process will be abbreviated
as MBP (Mobile Business Process). In process
model I, users need to search for distinct mobile
business platforms, login on and submit requirement
for several time to get the required information. Fig.
5 demonstrates the model.
Figure 5: Model I Simulation Logic Model.
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
90
Table 1 (Zhao, 2007) illustrates the model unit,
function and distillation function of Model I.
Table 1: Model unit of Simulation Logic Model.
Model
unit
Function of
model unit
Distillation
function
Arrive
User arrives
or service
needs occur
Generation time
is subject to a
random
distribution
Route
Transfer path
between sites
Transfer time is
subject to a
random
distribution
Process
Specific
service
provided by
institutions.
Service time is
subject to a
random
distribution
Decide
Service
judgment
/
Record
Record
related events
and indicators
/
Leave
Service
finished;
leave
/
Dispose Leave system /
5.2 Grid-based Process Simulation
Model
1Simulation Model Description of Model
Fig. 4 illustrates the grid-based mobile business
process model. Each time user login on the platform,
the requirement submitted and the service
department will transact the service by rules and
regulations. Users can access and submit
requirement to the 24-hour system. For simplicity, in
this paper, one resource servers one requirement, not
considering a need having two or more resources.
2Build the Simulation Logic Model
Grid-based mobile business process focuses on
requirement service. Users can submit their
requirement through unified processing window
directly and the order will be sent to various
providers for implement. The response cycle of
grid-based process includes landing process, unified
information service and service waiting time. For
simplicity, the mobile business center will be
abbreviated as MBC (Mobile Business Center), as
Fig. 6 illustrated.
Figure 6: Model Simulation Logic Model.
5.3 Simulation Results Analysis of
Grid-based Process and Non-grid
Process
1Simulation Evaluation Index
According to the grid-based mobile business
process’s goal of serving for the public, in this paper,
some common service quality evaluation
indicators
(Chi, 2008)
are selected to evaluate the service
of grid-based mobile business process.
(1). Average queue length
q
L
q
L
——Average waiting queue length,
expectation value of the number of customers
waiting for service in a steady-state system.
=
T
Lq
T
q
T
dtt
L
0
)(
lim
(1)
)(t
Lq
is the waiting queue length at the time of
t
,
T
is the operation time. Average waiting queue
length means the number of customers waiting for
service in the mobile business process.
(2). Average queue length
L
——Average queue length, expectation value
of the number of all customers.
+
=
T
Lq
T
T
dttSt
L
0
))()((
lim
(2)
()St
means the customers receiving the service,
queue length at
t
means
)(t
Lq
plus
()St
.
The
sum of average customer number marked
as
WIP
.Average queue length in a real mobile
business processes means the total number of
RESEARCH ON GRID-BASED MOBILE BUSINESS PROCESS AND SIMULATION
91
customers in the system (the number of customers
waiting in line plus the number of customers who are
receiving service and).
(3). Average waiting time
q
W
——Average waiting time, the expectation
value of waiting time
=
=
K
i
Zi
K
Wq
K
1
lim
(3)
i
Z
indicates the time customer
i spends waiting
for,
K
indicates the total customer service
number .Waiting time in the real business processes
indicates the period of time from the requirement is
submitted to the time the requirement is processed.
(4). Resource utilization
In terms of each resource, Arena will report two
kinds of utilization statistics, namely Instantaneous
Utilization and Scheduled Utilization.
Instantaneous Utilization is the resource
utilization at a certain point of time. If
()
B
t
is the
number of resource at a
t
in a busy state,
()
t
is
the number of available resource at
t
.
Suppose
() ()/ ()Ut Bt Mt=
,
() 0Mt >
. If
() 0Mt=
, and define
() 0Ut
=
. If
the simulation starts from 0 to
T
,
then the
Instantaneous Utilization is:
T
TdttU
0
/)(
(4)
That is the average time of function
()Ut
.
Scheduled Utilization is the ratio of the
average number of resources in the busy state with
the average number of available resources. The
Scheduled Utilization is:
=
T
T
T
T
dttM
dttB
TdttM
TdttB
0
0
0
0
)(
)(
/)(
/)(
(5)
In summary, the first three indicators are
commonly used to evaluate the steady-state
capability, the services responsiveness and service
congestion can be evaluated from the view of user
and process service.
2Simulation Hypothesis
It is assumed that non-grid and grid-based service
model have the same business, namely user submit
requirement in the same way, arrival time has the
same distribution and the arrival time follows TRIA
1 3 5 triangle distribution, making the
simulation comparable.
Assuming the service cycle is 8 hours a day and
grid-base model users can submit requirement in 24
hour. Besides, we also assume the arrival of system
events and the processing time of each node follow
exponential distribution; the delay of internal
operation process follows triangular distribution. The
basic statistical unit of simulation is minutes; the
length of simulation is one week.
3Analysis of Simulation Results
According to the simulation model and inputted data,
using the Arena software for simulation operation.
The system works 8 hours a day and 7 days each
time, a .txt file will be generated after the simulation
with all the statistic results in a report form. Finally,
a serious of simulation results is obtained, as Fig. 7
stated.
4.4003
0.2494
0
2
4
6
non-grid grid
t
i
m
e
/
m
i
n
user'sdemand
comparisonchartoftheaveragewaiting
time
(a)
17.3451
6.5549
0
10
20
non-grid grid
user'sdemand
comparison chart of WIP
(b)
37%
81%
70%
94%
0%
50%
100%
resourse 1 resourse 2
comparison chart of Scheduled
Utilization
non-grid
grid
(c)
Figure 7: Simulation results.
ICEIS 2011 - 13th International Conference on Enterprise Information Systems
92
(1) Comparison of average waiting time
Waiting time indicates the period of time from
the requirement is submitted to the time the
requirement is processed. The services
responsiveness and service congestion can be
evaluated from the point of view of user and process
service. In Figure (a), the average waiting time in
non-grid mobile business process is more than 4
minutes, and there is almost no waiting time in the
grid-based mobile business process model. It is
obvious that grid-based mobile business process has
faster response capability.
(2) Comparison of WIP
Average queue length means the total number of
customers in the system (the number of customers
waiting in line plus the number of customers who are
receiving service and). The figure (b) demonstrates
that the average queue of non-grid process is about
three time of that of gird-based process.
(3) Comparison of resource utilization
The resource utilization includes Instantaneous
Utilization and Scheduled Utilization. In the
simulation result, they have the same indicators
which referred to here as resource utilization. It can
be seen from the figure (c) that grid-based process
has higher resource utilization than non-grid based
process.
In conclusion, grid-based model has great
advantage over the non-gird model.
6 CONCLUSIONS
With the development of information technology,
mobile business has entered into a service
–oriented and process optimization focused stage.
The main difficulties are information sharing,
business collaboration and business process
reengineering. There is no in-depth systematic
analysis on this issue but only some ideas in
theory. In this paper, combing the research result
of grid management theory, mobile grid and Arena
simulation technology, in the view point of system
engineering, a grid-based mobile business process
is built, paving the way for the research on mobile
business process.
In this paper a grid-based mobile business
process model is built on the basis of grid
management theory. By comparing the service
difference through constructing corresponding
simulation models with related evaluation
indicators, it is
clear to see that grid-base process
has incomparable advantage over that of non-grid
process.
REFERENCES
Hull J. Options Futures and other Derivatives. Prentice-
Hall. 1997
A. Taudes, M. Feurstein, A. Mild. Options analysis of
software platform decisions. A case study. MIS
Quarterly. 2000.24(2):227-243.
U. Varshney, R. J. Vetter, R. Kalakota
Mobile commerce:
a frontier, computer.2000.33:32-38
Tianwei Wan, Huanchen Wang, Dexiu Ma
Principles of
Grid Management and research on grid model
structure
Information Science. 2007. 25 (03):456-461
Shiyuan Zhen, Hui Xv, Huanchen Wang. Summary of grid
and grid management System Engineering.
2005(3):1-7
Dexiu Wang, Huanchen Wang. Thinking the terms on the
Grid Management..Grid Management Seminar Papers.
2003. 24:109-115
Chunlin DengmKai Zou. Analysis of E-Commerce based
on grid technology. Information Science. 2006(11)
Chunyan Ma
The application research of OGSA based
grid service. Hehai University. 2008
Kun HemYong Zhao
Grid Resource Management and
Scheduling Review.Wuhan University of Technology
(Information & Management Engineering). 2007.
27(4):1-5
Wenjin Zhao, Zhaoxiang Shi. Shuguang Huang. Summary
of Mobile Grid Computing. Application of electronic
technology. 2005 (5)
Zhongren Chi. Research of Urban E-Government based on
grid management and Information Distance Theory.
PhD thesis, Shanghai Jiao Tong University. 2008 (2)
RESEARCH ON GRID-BASED MOBILE BUSINESS PROCESS AND SIMULATION
93