Discrete event simulation can be used to model various processes, using entities
or tokens moving through queues over time. Figure 9.3 describes the project
steps in such an effort, beginning with establishment of the objectives and scope
and proceeding to implementation. Typically, there are two types of queues
involved in discrete event simulation. First, we have those where entities just
wait, what we normally consider a queue. Second, we see those where entities
reside while the �value-adding work� is carried out. These queues are often
considered to be activities. The �event� occurs when the model changes state,
normally instantaneously. This is typically the start or end of an activity or some
form of interruption, such as a breakdown, lunch break, etc. At the end of an
activity, the simulation would check to see what other activities can now begin
as a result and schedule the end of that activity after allocating the required
Figure 9.3. Project Steps
About Lanner Applications in Air Travel Status
Establish
Objectives & Scope
Data & Level of Detail
Structure Model
Build Model
Verify Model
Validate Model
Experimentation
Project Management
Documentation
& Communication
Present Results
Implementation
Project Definition
Model Building and Testing
Experimentation
Project Completion
About Simulation
Saturday, July 3, 2010
WHY SHOULD A FIRM SIMULATE POTENTIAL BUSINESS PROCESSES?
Simulation becomes valuable when there is variability of outcomes in the business
processing or when parameters and rules are changed. Under these conditions,
it is not easy to predict the outcomes, particularly when these factors change
over time. For example, the famous �beer distribution game� represents a simple
supply chain consisting of a retailer, wholesaler, and factory. The retailer orders
cases of beer from the wholesaler, which in turn orders beer from the factory.
There is a delay between placing an order and receiving the cases of beer. The
game demonstrates that a small perturbation in the number of cases of beer sold
by the retailer can cause large shifts in the quantity of cases stored and produced
by the wholesaler and factory, respectively. Such a system is subject to dynamic
complexity � like the majority of contemporary supply chains.
Of course, many operating systems are subject to both variability and dynamic
complexity. Indeed, the variability of one component interacts with the
variability of another to create dynamic complexity.The level of customer
service is simple to predict, since there is no variability
in the system and because there is no interaction between components.
The lack of interaction is a result of the service time being exactly the same
for each customer, which means that there is no queuing or blocking. The
Most activities, however, do not take exactly the same time every time.
Assume that the times given above are averages, so customers arrive on average
every five minutes and it takes on average five minutes to serve a customer.
What is the average time a customer spends in the queue waiting to be served?
This is not an easy question to answer, especially when subsequent steps are
also considered, as there is variability in both customer arrivals and service
times. Queues would develop between the steps and create consequential effects
on performance. Most people, when asked for a specific time, tend to underestimate
the likely queuing time. Of course, the actual queuing time depends
on many things even in the one-step system, such as:
Variability in arrivals
Variability in regular service time
Variability in service time that might be affected by type of customer
or time of day
Discrete event simulation provides the technique for evaluating such systems
effectively, modeling the process from the �bottom up,� starting at the
required level of detail. Changing the process inherent in the model provides
the ability to evaluate the effect of such a change. Operations that involve
multiple processes and how they interact can also be modeled, by effectively
linking multiple models together.
The benefits of simulation include:
Risk reduction
Greater understanding of process conditions and interactions1.
processing or when parameters and rules are changed. Under these conditions,
it is not easy to predict the outcomes, particularly when these factors change
over time. For example, the famous �beer distribution game� represents a simple
supply chain consisting of a retailer, wholesaler, and factory. The retailer orders
cases of beer from the wholesaler, which in turn orders beer from the factory.
There is a delay between placing an order and receiving the cases of beer. The
game demonstrates that a small perturbation in the number of cases of beer sold
by the retailer can cause large shifts in the quantity of cases stored and produced
by the wholesaler and factory, respectively. Such a system is subject to dynamic
complexity � like the majority of contemporary supply chains.
Of course, many operating systems are subject to both variability and dynamic
complexity. Indeed, the variability of one component interacts with the
variability of another to create dynamic complexity.The level of customer
service is simple to predict, since there is no variability
in the system and because there is no interaction between components.
The lack of interaction is a result of the service time being exactly the same
for each customer, which means that there is no queuing or blocking. The
Most activities, however, do not take exactly the same time every time.
Assume that the times given above are averages, so customers arrive on average
every five minutes and it takes on average five minutes to serve a customer.
What is the average time a customer spends in the queue waiting to be served?
This is not an easy question to answer, especially when subsequent steps are
also considered, as there is variability in both customer arrivals and service
times. Queues would develop between the steps and create consequential effects
on performance. Most people, when asked for a specific time, tend to underestimate
the likely queuing time. Of course, the actual queuing time depends
on many things even in the one-step system, such as:
Variability in arrivals
Variability in regular service time
Variability in service time that might be affected by type of customer
or time of day
Discrete event simulation provides the technique for evaluating such systems
effectively, modeling the process from the �bottom up,� starting at the
required level of detail. Changing the process inherent in the model provides
the ability to evaluate the effect of such a change. Operations that involve
multiple processes and how they interact can also be modeled, by effectively
linking multiple models together.
The benefits of simulation include:
Risk reduction
Greater understanding of process conditions and interactions1.
SIMULATION AND KEY PERFORMANCE INDICATORS:
Key performance indicators (KPIs) help an organization define and measure
progress toward organizational goals. Once an organization has analyzed its
mission, identified all its stakeholders, and defined its goals, it needs a way to
measure progress toward those goals. KPIs are typically used for that purpose
as measurements that are quantifiable, agreed to beforehand, and reflect the
critical success factors of an organization. They will differ depending on the
organization, but within a vertical industry sector some should be common and
used to benchmark a company�s performance against competitors.
A business needs to set targets for each KPI. A company goal to be the
employer of choice might include a KPI of �people turnover rate.� After the
KPI has been defined as �the number of voluntary resignations and terminations
for poor performance, divided by the total number of employees at the beginning
of the period� and a way to measure it has been set up by collecting the
information in the human resources system, the target has to be established.
�Reduce people turnover by 5% per year� is a clear target that everyone will
understand and be able to take specific action to accomplish. The question then
becomes: What are the specific actions that will deliver the objective?
Business process management is about managing corporate or business performance
through managing those processes which drive the firm to the desired overall objectives.
KPIs therefore need to be designed at various levels in order
to be relevant at an individual process level. While the KPI defined above
concerning turnover rate might be relevant for the board, it is not appropriate
for the supervisor of the human resources department�s contact center. That
center�s performance may well have an effect on turnover rate, but it requires
its own process-level KPIs, perhaps based on responding successfully to employee
questions, in order for it to be managed effectively.
Simulation and optimization supporting the identification and design of
KPIs have been used extensively to help improve processes and consequently
to ultimately improve business performance. Simulation can be used to both
help define and set targets for a KPI. This feature is in addition to the use of
simulation to identify process changes (resources, rules, and structural needs),
which deliver improved performance. Simulation aids the definition of a KPI
(that is, the equation and data of which it is composed), through examining the
reaction of that KPI under different circumstances. This procedure ensures that
the KPI does in fact properly connect to the organization�s goals and will help
drive the appropriate behavior and reaction to different potential events. Most
KPIs are developed directly from the overall goals or critical success factors
of the organization. Process-based KPIs tend to be focused on results or outputs
from a process, such as customers served per hour or claims processed.
Simulation is also valuable in the setting of targets against KPIs (such as
the number of calls to be answered per hour), because the simulation can
accurately assess what is achievable in theory by the process and resources
employed. Many KPIs are now presented as part of a �digital cockpit,� using
gauge or dashboard-style displays. Depending on the KPI and the targets selected,
they often include some use of zones to highlight acceptable levels (e.g.,
green, yellow, and red). It is important when calibrating these gauges to understand
the effect of natural fluctuations due to inherent randomness within processes
or behavior. Simulation can support this calibration to help ensure that
unnecessary reaction to natural short-term fluctuations does not occur.
Processes have a series of steps, and at each step measurement can be taken
to better understand the behavior of the process under different circumstances.
Identification of key points in the process and the relevance of specific measures
(probes) in predicting failure of the process can provide significant opportunity
for improved process management and business value. Effective process monitoring
through probes at specific points in the process involves watching factors
to give prior warning that the target levels are under threat. Simulation is able
to help identify these probes and the threshold values that indicate a deterioration
of the process which will ultimately end in an unacceptable KPI.
progress toward organizational goals. Once an organization has analyzed its
mission, identified all its stakeholders, and defined its goals, it needs a way to
measure progress toward those goals. KPIs are typically used for that purpose
as measurements that are quantifiable, agreed to beforehand, and reflect the
critical success factors of an organization. They will differ depending on the
organization, but within a vertical industry sector some should be common and
used to benchmark a company�s performance against competitors.
A business needs to set targets for each KPI. A company goal to be the
employer of choice might include a KPI of �people turnover rate.� After the
KPI has been defined as �the number of voluntary resignations and terminations
for poor performance, divided by the total number of employees at the beginning
of the period� and a way to measure it has been set up by collecting the
information in the human resources system, the target has to be established.
�Reduce people turnover by 5% per year� is a clear target that everyone will
understand and be able to take specific action to accomplish. The question then
becomes: What are the specific actions that will deliver the objective?
Business process management is about managing corporate or business performance
through managing those processes which drive the firm to the desired overall objectives.
KPIs therefore need to be designed at various levels in order
to be relevant at an individual process level. While the KPI defined above
concerning turnover rate might be relevant for the board, it is not appropriate
for the supervisor of the human resources department�s contact center. That
center�s performance may well have an effect on turnover rate, but it requires
its own process-level KPIs, perhaps based on responding successfully to employee
questions, in order for it to be managed effectively.
Simulation and optimization supporting the identification and design of
KPIs have been used extensively to help improve processes and consequently
to ultimately improve business performance. Simulation can be used to both
help define and set targets for a KPI. This feature is in addition to the use of
simulation to identify process changes (resources, rules, and structural needs),
which deliver improved performance. Simulation aids the definition of a KPI
(that is, the equation and data of which it is composed), through examining the
reaction of that KPI under different circumstances. This procedure ensures that
the KPI does in fact properly connect to the organization�s goals and will help
drive the appropriate behavior and reaction to different potential events. Most
KPIs are developed directly from the overall goals or critical success factors
of the organization. Process-based KPIs tend to be focused on results or outputs
from a process, such as customers served per hour or claims processed.
Simulation is also valuable in the setting of targets against KPIs (such as
the number of calls to be answered per hour), because the simulation can
accurately assess what is achievable in theory by the process and resources
employed. Many KPIs are now presented as part of a �digital cockpit,� using
gauge or dashboard-style displays. Depending on the KPI and the targets selected,
they often include some use of zones to highlight acceptable levels (e.g.,
green, yellow, and red). It is important when calibrating these gauges to understand
the effect of natural fluctuations due to inherent randomness within processes
or behavior. Simulation can support this calibration to help ensure that
unnecessary reaction to natural short-term fluctuations does not occur.
Processes have a series of steps, and at each step measurement can be taken
to better understand the behavior of the process under different circumstances.
Identification of key points in the process and the relevance of specific measures
(probes) in predicting failure of the process can provide significant opportunity
for improved process management and business value. Effective process monitoring
through probes at specific points in the process involves watching factors
to give prior warning that the target levels are under threat. Simulation is able
to help identify these probes and the threshold values that indicate a deterioration
of the process which will ultimately end in an unacceptable KPI.
CONSIDER THE ADVANTAGES OF EXPERIMENTATION:
Consider a health service example, where the number of patients admitted to
the clinic is a set number arriving according to a variable timing profile. To
establish the weekly variability of utilizations of clinical staff, the model can
be run for either a single run of 50 weeks or for 50 runs of a single week. Each
of these runs will enable the variation of utilizations to be observed and confidence
intervals calculated. Multiple runs are important. You would not judge
the fairness of a coin by just one or two tosses!
It is true that the above model is simplistic, as indeed the profile of patients
in such a model would probably vary in ways that would require more extensive
modeling (e.g., the definition of a likely profile over a longer period to take into
account current observed fluctuations from week to week). However, the experimentation
options hold true: either a short period of experimentation can be
repeated many times (with different random number sampling), or a model can
be run for a much longer time and the range of random effects can be observed.
How many repeats and how long to run a model are questions that can be
answered by a combination of input and output data analysis. It is important
when running a model that it experience all the randomness of the input data
and that the output results �settle down� before conclusions are drawn.
At a second level, a user may wish to alter different parameters within a
model to observe the different effects. Again, with this type of experimentation
the range of results might also be important, once they are more accomplished
through elongated or repeated experiments.
At a third level, a user may wish to compare one model with another. For
example, in manufacturing there may be two investment options: production
layout A and production layout B. These may indeed be the only options,
although within each solution there may be parameter choices (e.g., containing
buffer storage level options). In addition, there is again the optional value of
establishing the potential range of results.
With all these levels of experimentation, there are a variety of experimental
designs that can be used � different numbers of replications and different types
of factorial experimental designs, ranging from full factorial to half factorial to
Latin square-type designs where different parameter combinations are chosen.
For different models, the model itself can simply be considered a different type
of parameter.
the clinic is a set number arriving according to a variable timing profile. To
establish the weekly variability of utilizations of clinical staff, the model can
be run for either a single run of 50 weeks or for 50 runs of a single week. Each
of these runs will enable the variation of utilizations to be observed and confidence
intervals calculated. Multiple runs are important. You would not judge
the fairness of a coin by just one or two tosses!
It is true that the above model is simplistic, as indeed the profile of patients
in such a model would probably vary in ways that would require more extensive
modeling (e.g., the definition of a likely profile over a longer period to take into
account current observed fluctuations from week to week). However, the experimentation
options hold true: either a short period of experimentation can be
repeated many times (with different random number sampling), or a model can
be run for a much longer time and the range of random effects can be observed.
How many repeats and how long to run a model are questions that can be
answered by a combination of input and output data analysis. It is important
when running a model that it experience all the randomness of the input data
and that the output results �settle down� before conclusions are drawn.
At a second level, a user may wish to alter different parameters within a
model to observe the different effects. Again, with this type of experimentation
the range of results might also be important, once they are more accomplished
through elongated or repeated experiments.
At a third level, a user may wish to compare one model with another. For
example, in manufacturing there may be two investment options: production
layout A and production layout B. These may indeed be the only options,
although within each solution there may be parameter choices (e.g., containing
buffer storage level options). In addition, there is again the optional value of
establishing the potential range of results.
With all these levels of experimentation, there are a variety of experimental
designs that can be used � different numbers of replications and different types
of factorial experimental designs, ranging from full factorial to half factorial to
Latin square-type designs where different parameter combinations are chosen.
For different models, the model itself can simply be considered a different type
of parameter.
Fitting.Organization,.Environment,.and.ICT:
In the previous section, we argued that businesses can be involved in three types of organizational
integration. As the business needs to be integrated, ICT systems need to be
integrated, too (as is discussed in contingency theory; see, e.g., Borgatti, 2001). Therefore,
as companies are confronted with three basic types of integration at organizational level
(internal, within the Extended Enterprise, and with the marketplace) we should recognize
three levels of IT-integration as well.
The first type of IT-integration companies should realize is the internal integration of the
diverse systems within company walls generally referred to as �enterprise application integration�
(EAI).
The two other types of IT-integration concern B2Bi, the topic of this chapter. First there is the
extended enterprise integration (EEi). In the context of the extended enterprise, companies
that dispose of capabilities that are useful for each other try to cooperate/collaborate. It is
important to note that partnering organizations have decided to do business with each other
for an extended period of time. They know the other company can deliver to a certain extent
what is needed. A partnership is set up to get more out of the other company than what is
already being delivered, and it is recognized that some form of coordination is necessary to
realize additional benefits. Partnering enterprises need to find out how they can be of more
value to each other. The development of customized software is part of this value adding
effort. It is clear that partner-specific IT investments can be made.
Essentially, this is not the case in the other type of B2Bi. This second form of B2Bi we call
market B2Bi. Companies that do business in the marketplace do not cooperate/collaborate.
Basically for each transaction they try to find out who can deliver what is needed. Every
time again, companies have the free choice to choose the services from a company (present
in the marketplace) that fulfills the needs. Therefore, no thorough coordination among the
companies is needed. Of course, service-providing companies try to pick up signals from the
market to deliver the services that are useful, and they try to minimize costs, but there is no
partnering. This scenario shows the IT integration alternatives. Market Web services have
mainly been developed in isolation and may be found through a market mechanism such
as the global UDDI (universal description, discovery, and integration) registries. Furthermore,
organizations may do business with many other organizations through an electronic
marketplace. Figure 3 shows the ideas presented here.
Currently, the boundary between EEi and market B2Bi is vague. These two types of B2Bi
actually cover a whole continuum of B2Bi practices (as is also clear from organization theory).
With the current state of technology, we believe that Market B2Bi primarily concerns the
indirect integration through electronic marketplaces. In the future new Web services standards
and semantic Web standards may be developed that enable organizations to dynamically
integration. As the business needs to be integrated, ICT systems need to be
integrated, too (as is discussed in contingency theory; see, e.g., Borgatti, 2001). Therefore,
as companies are confronted with three basic types of integration at organizational level
(internal, within the Extended Enterprise, and with the marketplace) we should recognize
three levels of IT-integration as well.
The first type of IT-integration companies should realize is the internal integration of the
diverse systems within company walls generally referred to as �enterprise application integration�
(EAI).
The two other types of IT-integration concern B2Bi, the topic of this chapter. First there is the
extended enterprise integration (EEi). In the context of the extended enterprise, companies
that dispose of capabilities that are useful for each other try to cooperate/collaborate. It is
important to note that partnering organizations have decided to do business with each other
for an extended period of time. They know the other company can deliver to a certain extent
what is needed. A partnership is set up to get more out of the other company than what is
already being delivered, and it is recognized that some form of coordination is necessary to
realize additional benefits. Partnering enterprises need to find out how they can be of more
value to each other. The development of customized software is part of this value adding
effort. It is clear that partner-specific IT investments can be made.
Essentially, this is not the case in the other type of B2Bi. This second form of B2Bi we call
market B2Bi. Companies that do business in the marketplace do not cooperate/collaborate.
Basically for each transaction they try to find out who can deliver what is needed. Every
time again, companies have the free choice to choose the services from a company (present
in the marketplace) that fulfills the needs. Therefore, no thorough coordination among the
companies is needed. Of course, service-providing companies try to pick up signals from the
market to deliver the services that are useful, and they try to minimize costs, but there is no
partnering. This scenario shows the IT integration alternatives. Market Web services have
mainly been developed in isolation and may be found through a market mechanism such
as the global UDDI (universal description, discovery, and integration) registries. Furthermore,
organizations may do business with many other organizations through an electronic
marketplace. Figure 3 shows the ideas presented here.
Currently, the boundary between EEi and market B2Bi is vague. These two types of B2Bi
actually cover a whole continuum of B2Bi practices (as is also clear from organization theory).
With the current state of technology, we believe that Market B2Bi primarily concerns the
indirect integration through electronic marketplaces. In the future new Web services standards
and semantic Web standards may be developed that enable organizations to dynamically
The.Extended. Enterprise. vs.. Other.Forms. of.Doing. Business:
For a long time, two basic forms of economic organization have been recognized: markets
on the one hand and hierarchies (firms) on the other. Powell (1990) refers to
Ronald Coase as the person who first discussed the firm as a governance structure
rather than just as a black box that transforms inputs into outputs. Coase (1937)
asserts that firms and markets are alternative means for organizing similar
kinds of transactions. Only in the 1970s did ct upon Coase�s findings. One of these proponents,
Williamson (1975, 1985), argues that some transactions are more likely to take
place within hierarchically organized firms (Williamson equated firms with hierarchies)
than through a market interface. More specifically, he states that transactions that are to
be executed within hierarchically organized firms are likely to involve uncertainty about
their outcome, recur frequently and require substantial �transaction-specific investments�
(of money, time, or energy) that cannot be easily transferred. On the other hand, exchanges
that are straightforward, non-repetitive, and require no transaction-specific investments can
be expected to take place across a market interface. This dichotomous view of markets and
hierarchies�as discussed by Williamson (1975)�sees firms as separate from
markets and assumes the presence of sharp firm boundaries. These
sharp boundaries, however, do not always seem to be present. This is true especially in the
case of partnering organizations (extended enterprises, see Figure 1). Transactions between
partnering companies can be seen as a hybrid form of economic organization. That is, if
transactions are distributed as points along a continuum with discrete market transactions
located at one end and the highly centralized firm at the other end, partnering companies
fall in between these poles. The definition of Podolny and Page is very much aimed at
identifying the differences between the network form of organization on the one hand,
and markets and hierarchies on the other hand:
In pure markets companies do not aim at enduring relations, and in hierarchies there is a
clearly recognized, legitimate authority that can resolve disputes that arise among actors.
Besides these two characteristics, some scholars (e.g., Dore, 1983; Powell, 1990) have argued
that network forms of organization also posses another characteristic, namely a distinct ethic
or value-orientation on the part of exchange partners. Hirschman (1970) argues that partners
are willing to make relationship-specific investments without contractual guarantees protecting
those investments, and for Powell the norm of reciprocity is key (1990, pp. 303-304):
on the one hand and hierarchies (firms) on the other. Powell (1990) refers to
Ronald Coase as the person who first discussed the firm as a governance structure
rather than just as a black box that transforms inputs into outputs. Coase (1937)
asserts that firms and markets are alternative means for organizing similar
kinds of transactions. Only in the 1970s did ct upon Coase�s findings. One of these proponents,
Williamson (1975, 1985), argues that some transactions are more likely to take
place within hierarchically organized firms (Williamson equated firms with hierarchies)
than through a market interface. More specifically, he states that transactions that are to
be executed within hierarchically organized firms are likely to involve uncertainty about
their outcome, recur frequently and require substantial �transaction-specific investments�
(of money, time, or energy) that cannot be easily transferred. On the other hand, exchanges
that are straightforward, non-repetitive, and require no transaction-specific investments can
be expected to take place across a market interface. This dichotomous view of markets and
hierarchies�as discussed by Williamson (1975)�sees firms as separate from
markets and assumes the presence of sharp firm boundaries. These
sharp boundaries, however, do not always seem to be present. This is true especially in the
case of partnering organizations (extended enterprises, see Figure 1). Transactions between
partnering companies can be seen as a hybrid form of economic organization. That is, if
transactions are distributed as points along a continuum with discrete market transactions
located at one end and the highly centralized firm at the other end, partnering companies
fall in between these poles. The definition of Podolny and Page is very much aimed at
identifying the differences between the network form of organization on the one hand,
and markets and hierarchies on the other hand:
In pure markets companies do not aim at enduring relations, and in hierarchies there is a
clearly recognized, legitimate authority that can resolve disputes that arise among actors.
Besides these two characteristics, some scholars (e.g., Dore, 1983; Powell, 1990) have argued
that network forms of organization also posses another characteristic, namely a distinct ethic
or value-orientation on the part of exchange partners. Hirschman (1970) argues that partners
are willing to make relationship-specific investments without contractual guarantees protecting
those investments, and for Powell the norm of reciprocity is key (1990, pp. 303-304):
Approaches. to. Business.Modeling:
Business modeling is a vast area of research and practice, which is gaining increasing importance
in the rapid development of e-business and globalisation (Holsapple, 2001). The
following are some approaches used in business modeling:
1. Textual description (e.g., abstract use cases)
2. Categories and visualization
3. Simulation approaches
4. Data/workflow-description programs
5. Process notation languages (e.g., Unified Modeling Language (UML))
The common categorization of B2B models seen in the literature is divisions based on target
audience or type of products/services involved in the business processes.
In order to understand this categorization, we have first to observe the two dimensions
through which companies obtain the products and services they need (Kaplan & Sawhney,
2000). The first dimension is related to the kind of products and services purchased, and the
second relates to the frequency with which companies make their purchases. The products
and services can be classified into manufacturing inputs and operating inputs. Manufacturing
inputs are raw materials applied in the manufacturing process. Operating inputs are
normally low-value products and services, usually commodities (e.g., advertising, utilities,
electricity, office supplies), also known as maintenance, repair, and operations (MRO)
inputs. In relation to the frequency inputs are acquired, there are both systematic sourcing
and spot-market sourcing.
Another more general and easier-to-remember categorization can be attained by focusing on
the processes that these e-business models are designed to facilitate. E-businesses can support
the relationship between a company and its customers and suppliers (usually commercial
transactions) and with related partners like collaborators and contractors. Thus e-business
models can be classified according to its focus (IBM, 2003a; Weill & Vitale, 2001):
� Focus on the process of selling goods and services to other companies (e-commerce
sell-side)
� Focus on the process of e-procurement and other processes related to the supply chain
management. Various sub-models in this category are as follows:
o Buyer model (few buyers, many sellers)
o Marketplace model (many buyers and many sellers)
o Longer-term relationship model (few buyers and few sellers)
o Seller model (few sellers, many buyers)
� Focus on collaboration between businesses, partners, and contractors (i.e., e-collaboration,
in the rapid development of e-business and globalisation (Holsapple, 2001). The
following are some approaches used in business modeling:
1. Textual description (e.g., abstract use cases)
2. Categories and visualization
3. Simulation approaches
4. Data/workflow-description programs
5. Process notation languages (e.g., Unified Modeling Language (UML))
The common categorization of B2B models seen in the literature is divisions based on target
audience or type of products/services involved in the business processes.
In order to understand this categorization, we have first to observe the two dimensions
through which companies obtain the products and services they need (Kaplan & Sawhney,
2000). The first dimension is related to the kind of products and services purchased, and the
second relates to the frequency with which companies make their purchases. The products
and services can be classified into manufacturing inputs and operating inputs. Manufacturing
inputs are raw materials applied in the manufacturing process. Operating inputs are
normally low-value products and services, usually commodities (e.g., advertising, utilities,
electricity, office supplies), also known as maintenance, repair, and operations (MRO)
inputs. In relation to the frequency inputs are acquired, there are both systematic sourcing
and spot-market sourcing.
Another more general and easier-to-remember categorization can be attained by focusing on
the processes that these e-business models are designed to facilitate. E-businesses can support
the relationship between a company and its customers and suppliers (usually commercial
transactions) and with related partners like collaborators and contractors. Thus e-business
models can be classified according to its focus (IBM, 2003a; Weill & Vitale, 2001):
� Focus on the process of selling goods and services to other companies (e-commerce
sell-side)
� Focus on the process of e-procurement and other processes related to the supply chain
management. Various sub-models in this category are as follows:
o Buyer model (few buyers, many sellers)
o Marketplace model (many buyers and many sellers)
o Longer-term relationship model (few buyers and few sellers)
o Seller model (few sellers, many buyers)
� Focus on collaboration between businesses, partners, and contractors (i.e., e-collaboration,
Subscribe to:
Posts (Atom)