Exploring the Role of Entrepreneurship and Business Models in Technology
Diffusion: A Case Study of the Market Diffusion of an Enabling Technology
for Sustainable C onsumption and Production
Weijia Ran*!, Luis F. Luna-Reyes”, Jing Zhang’, Djoko Sayogo’, Francois Duhamel”, Sergio
Picazo-V ela”, David Andersen*
“University at Albany, Suny, Albany, NY, USA
Universidad de las Américas-Puebla, Escuela de Negocios, Santa Catarina Martir, Cholula,
Puebla, Mexico
“Clark University, Worcester, MA, USA
“University of Muhammadiyah at Malang, Indonesia
Abstract
Entrepreneurial activities and business models describe ways to start and maintain a business.
Empirical data show that they play an important role in bringing technology-based products or
services to market. However, the role of entrepreneurial activities and business models in the
diffusion process has not been specifically and systematically explored and discussed in the
adoption and diffusion literature, and there is a scarcity of simulation models that have examined
technology adoption and diffusion phenomena from an entrepreneurship and business-model
perspective. The purpose of our study is to contribute to this area by exploring the role of
entrepreneurship and business models in the diffusion process through a System Dynamics
modeling and simulation approach. We built a simulation model based on technology-diffusion-
related literature and empirical data collected through the process of implementing a sustainable
consumption and production initiative called I-Choose over three years. Our analysis of
simulation experiment results shows different entrepreneurial activities and business models
leads to different diffusion paths and associated market behaviors.
1. Introduction
Technology adoption and diffusion is an extensively studied topic in economics, marketing,
sociology and information systems. Researchers are interested in factors that would influence
technology adoption, and how products based on a technological innovation would spread in the
market and change the market structure. Entrepreneuring is a purposeful activity to seek the
opportunity, initiate, maintain and enlarge a profit business, and make impacts on the market by
innovatively mobilizing resources (Cole, 1949; Wiklund, 1998). Business models generally
£ Corresponding author: Weijia Ran, University at Albany; E-mail: wran@ albany.edu.
emphasize systemic explanations about how firms are doing business. Empirical data show
entrepreneurial activities and business models play an important role in bringing technology-
based products or services to market. Using different business models to take products or
services based on the same technology to market yields different economic and market outcomes
(Chesbrough, 2010; Chesbrough & Rosenbloom, 2002; Schumpeter; 1947). However, the role of
entrepreneurial activities and business models has not been specifically and systematically
explored and discussed in the adoption and diffusion literature. In addition, there is a scarcity of
diffusion simulation models that have studied technology adoption and diffusion from an
entrepreneurship and business model perspective. Therefore, the purpose of our study is to
contribute to this area by exploring the role of entrepreneurship and business models in the
diffusion process.
The paper is divided into six sections, including this introduction. The second section provides a
review of adoption and diffusion theories and models, definitions of entrepreneurship and
business models, and the role of entrepreneurship and business models in technology diffusion
and related research. The third section of the paper describes our research methods, and
introduces the case (I-Choose Project) that we used to build our simulation model and design
simulation experiments. The fourth section describes our simulation model. The fifth section
presents the main results from the simulation. Finally, the concluding part of the paper
summarizes our findings and proposes future study steps.
2. The Role of Entrepreneurship and Business Models in Technology Adoption
and Diffusion
Technology adoption and diffusion is an extensively studied topic in economics, marketing,
sociology and information systems. Researchers are interested in factors that would influence
technology adoption, and how products based on a technological innovation would spread in the
market and change the market structure. In the current era characterized by the rapid evolution of
information and web technologies, there are many studies that have examined how products
based on information and web technologies were adopted and spread in the market, e.g., products
based on mobile technologies (Bruner, I] & Kumar, 2005; Chen, Y en & Chen, 2009; Haro,
2010; Hung, Ku, & Chang, 2003), e-commerce (Eastin, 2002; Knutsen & Lyytinen, 2008; Liao,
Chen & Yen, 2007; Pavlou & Fygenson, 2006; Ruyter, Wetzels, & Kleijnen, 2001;
Vijayasarathy, 2004), information systems, and social media (Gruhl et al., 2004; Yi & Hwang,
2003). Empirical data show entrepreneurial activities and business models play an important role
in bringing technology-based products or services to market. However, our review of the existing
adoption and diffusion literature has found few studies that have investigated technology
adoption and diffusion from an entrepreneurship and business model perspective. Therefore, the
purpose of our study is to contribute to this literature by exploring the role of entrepreneurship
and business models in the diffusion process.
2.1 Adoption and Diffusion Theories and Models
As a well-explored area, adoption and diffusion research provides a variety of theories and
models to explain adoption and diffusion phenomena. A popular typology by Brown has divided
these theories and models into four categories (Brown, 1981; Miller & Garnsey, 2000). Theories
and models in the first category explain adoption and diffusion phenomena from a
communication perspective. From this perspective, the adoption and diffusion process can be
explained as an imitation behavior triggered by social interactions and communications. The
majority of potential adopters are risk-averse human beings, and they make adoption decisions
under uncertainty through the evaluation of risks and benefits of adoption. The left small portion
of potential adopters who are adventurous and innovative becomes pioneers and early adopters.
Through communications and social interactions, the early adopters’ behavior influences the
risk-benefit evaluations of other potential adopters. As a result, more and more potential adopters
start accepting the technology under influences such as the word of mouth or marketing.
The second category of adoption and diffusion theories and models explain the adoption and
diffusion phenomena as the result of potential adopters’ rational economic considerations. From
this perspective, potential adopters are rational economic agents, who make their adoption
decisions based on the cost-benefit evaluation. If the consumption benefit exceeds the price, the
product will be adopted, while the consumption benefit is determined by the product’s utility and
performance. Many popular theories and models in adoption and diffusion research can be
classified under both the first and second categories, for example, the diffusion of innovation
theory (Rogers, 1995), the technology acceptance model (Davis, Bagozzi & Warshaw, 1989), the
theory of planned behavior (Ajzen, 1991), and the expectation-confirmation theory (Oliver 1993).
These theories and models suggest the adoption decision is influenced by the consumer’s
intention of use. The intention of use is influenced by the consumer’s attitude towards the
product, which, in turn, is influenced by a range of internal and external factors. Intemal factors
refer to the consumer’s and product’s characteristics, such as the consumer’s self-efficacy and
the product’s usefulness. External factors refer to environmental factors such as social influence
or communications.
The third and the fourth categories have respectively addressed the roles of affordability and
availability in the diffusion process (Brown, 1981; Miller & Garnsey, 2000). Theories and
models in the third category explain the diffusion process as a consequence of the unequal
distribution of resources in society and the variation in adopters’ affordabilities. Studies based on
these theories and models see the diffusion process through the lens of development economics.
In contrast, theories and models in the fourth category suggest the diffusion process is s a
consequence of unequal opportunities to adopt. Factors such as geographical locations would
influence the market infrastructure, and the market infrastructure would influence the product’s
availability.
2.2 Entrepreneurship and Business Models
Entrepreneuring is a purposeful activity to seek the opportunity, initiate, maintain and enlarge a
profit business, and make impacts on the market by innovatively mobilizing resources (Cole,
1949; Wiklund, 1998). Business models generally emphasize systemic explanations about how
firms, including startup companies, fruits of entrepreneurial activities, are doing business. The
interest of scholars for the notion of business models has risen a lot since the advent of internet
and the development of e-businesses (A mit and Zott, 2001; Mahadevan, 2000; Timmers, 1998,
1999; Zott, Amit & Massa, 2011). Amit and Zott (2001: 494-495) define business model as “the
design of transaction content, structure, and governance so as to create value through the
exploitation of business opportunities”.
One of the main issues in the literature referring to e-business models concerns the components
of a business model. In spite of the diversity of approaches to identify the essential components
of business models, there is a certain convergence in the literature pointing to the following
elements: Infrastructure (comprising key activities, key resources and a partner network), a value
proposition, customer segments and type of relationships, distribution channels, a cost structure,
methods to generate revenue streams, as well as key performance indicators and explicit linkages
among the variables of the business system to assess its consistency (Dubosson-Torbay,
Osterwalder, & Pigneur, 2002; Osterwalder & Pigneur, 2002).
2.3 Technology Adoption and Diffusion from an Entrepreneurship and Business
Model Perspective
Empirical data show entrepreneurial activities and business models play an important role in
bringing technology-based products or services to market. Using different business models to
take products or services based on the same technology to market yields different economic and
market outcomes (Chesbrough, 2010; Chesbrough & Rosenbloom, 2002; Schumpeter; 1947).
However, the role of entrepreneurial activities and business models has not been specifically and
systematically explored and discussed in the adoption and diffusion literature, although the
importance of some activities related to business-model design have been touched upon in
previous adoption and diffusion research, for instance, marketing and investment activities.
Miller & Garnsey (2000) point out the significance of entrepreneurship is overlooked in existing
diffusion literature. The first category of adoption and diffusion theories and models, the
communication view of adoption and diffusion phenomena, is overly demand-side oriented, and
neglects supply-side processes and actors. The second category, the economic view of adoption
and diffusion phenomena focuses on entrepreneurs’ or organizations’ efforts in improving
technology and product performance and ignores non-technical problems such as marketing,
distribution, pricing, finances, etc. The third and fourth categories of adoption and diffusion
theories and models, which address the affordability and availability issues, have overlooked the
possibility that entrepreneurial activities can change the resource distribution and market
infrastructure by mobilizing resources and propagating technologies and products.
The overlooked importance of entrepreneurship and business models in diffusion research might
be able to explain the scarcity of simulation models that have studied technology adoption and
diffusion from an entrepreneurship and business model perspective. A fair amount of diffusion
simulation models were developed based on the famous Bass Model (1969), which reflects a
communication view, probably an overly demand-side oriented view as pointed out by Miller &
Gamsey (2000), of adoption and diffusion phenomena. The key hypothesis of this model is that
the probability that an initial purchase will be made at a certain time is a linear function of the
previous buyers. The bass model addresses the importance of early adopters and other adopters’
imitation behavior. The pressure operating on imitators increases as the number of previous
buyers increase. A nother set of diffusion simulation models follows the economic view of
adoption and diffusion phenomena and addresses efforts in technical development, but has a
limited discussion on non-technical issues. For example, Jack Homer’s model (1987), describes
how medical technologies evolves because producers are continuously improving the technology
based on consumers’ feedbacks, and consumers keeps adjusting their attitudes towards the
technology as these improvements are made. More up-to-date diffusion simulation models have
been trying to capture a systematic view of adoption and diffusion phenomena and include
various factors such as heterogeneity among potential adopters, market competition, regulations
and policies, etc. (Dattée, 2007; Dattée & Weil, 2005; Weil & Utterback, 2005). However, our
review has found few diffusion simulation models that have studied technology adoption and
diffusion from an entrepreneurship and business model perspective. The purpose of our study is
to contribute to this literature by exploring the role of entrepreneurship and business models in
the diffusion process.
3. Method
In this study we use system dynamics simulation experiments to explore the relationship between
business models and diffusion related market behaviors. The basic structure of our system
dynamics model is built based on theories and models in existing adoption and diffusion
literature. We adapted and developed this basic structure to characterize different scenarios in
which products developed based on the same enabling technology are taken to market by various
(corporate) entrepreneurs using different business models. These scenarios are designed based on
empirical data collected from a case study. The following two sections are an elaborate
description of this process.
3.1 Case Study
Besides the literature, the basis for our model building is an initiative called I-Choose. I-Choose
Project Team has been trying to increase supply chain transparency and product data disclosure
of coffee grown and sold in the North A merican Free Trade Area (NAFTA). An essential goal of
I-Choose project is to create a set of standards to facilitate product certification and inspection
data disclosure, sharing and utilization. The set of standards include a series of ontologies that
defines the certification and inspection knowledge domain, semantically organizes certification
and inspection data, and enables transparent, complete, and reliable certification and inspection
data retrieval. This series of ontologies are called Certification and Inspection Big Data
Infrastructure Building Block (CIDIBB). CIDIBB can be used to produce a variety of
applications or services. These applications or services can make information about how, where,
by whom, and under what conditions a particular product was produced available and usable to
consumers, and helps them make more informed and ethical purchase decisions with regards of
the social and environmental impacts of the coffee they drink (Luna-Reyes, Andersen, Andersen
et al., 2012; Luna-Reyes, Sayogo, Zhang et al., 2012). In the context of this paper, CIDIBB is the
enabling technology under discussion. The purpose of our simulation experiments is to observe
how the diffusion paths of CIDIBB-based applications or services would differ if they are taken
to market by (corporate) entrepreneurs using different business model designs.
3.2 Simulation Experiment Design
Through analyzing I-Choose interview data collected from coffee producers, retailers,
certification and inspection practitioners, and consumers, we designed five scenarios to represent
different ways to startup a CIDIBB-based business. The business models characterized in these
scenarios vary in terms of different designs of certain business-model elements, as shown in
Table 1.
Table 1. Scenarios Represent Different Business-model Element Designs
Business Model Elements
Scenarios: Value Proposition Customer Revenue Cost Structure Partner
(Corporate) Segment(s) | Generation Network
Entrepreneurs M i
#1: Virtual Help prodi attract Producers Certifying Fee Marketing to produc none
Certifier valued consumers CIDIBB construction &
maintenance
#2: Consumer | Enrich consumer Consumers | Information Package | Marketing to consumers; | none
Advocate purchasing experience (Product Ratings) CIDIBB construction &
Sale i
#3: TallMart Help producers Consumers; | Premium from Marketing to producers | Individual
distributing products; Producers consumers; and consumers; producers
Attract valued consumers Distribution fee CIDIBB construction &
by enriching consumer from producers maintenance;
i experience Dividend to producers
#4: iGuide Help retailers attract Retailers Distribution fee Marketing to retailers; Individual
valued consumers; from retailers CIDIBB construction & | producers
Motivate producers to maintenance;
disclose product data Subsidy to producers
#: Producer Help producers Consumers | Premium from Marketing to consumers; | Producers
Association distributing products; consumers; CIDIBB construction & | form an
Attract valued consumers Association fee maintenance; alliance
by enriching consumer (registration fee in
purchasing experience the model)from
producers
In scenario 1, a certifying organization called Virtual Certifier uses CIDIBB to create a virtual
certificate. The virtual certificate allows consumers to trace and obtain detailed certification
information online or scanning the product barcode. The virtual certifier charges producers
certifying fees. Producers benefit by charging consumers an extra premium for the virtual
certificate attached to their products.
In scenario 2, the Consumer Advocate is a product rating firm. The Consumer A dvocate
produces product ratings based on data retrieved through CIDIBB infrastructure and publishes
these ratings as information packages. Their business model is to sell a low cost-subscription to
their information packages to individual consumers.
In scenario 3, a corporation called TallMart recognizes the potential of the CIDIBB to bring
trusted information into the consumer marketplace as well as the commercial potential of
creating a platform wherein retail consumers pay a premium for products that can be sold with
CIDIBB-certified virtual certificates while at the same time producers of sustainable products are
willing to pay a fee to have information about their products distributed on TallMart’s platform
using the CIDIBB standard. TallMart also shares the profit with the producers to strengthen their
partnership.
In scenario 4, the corporation iGuide create a platform similar to TallMart’s , but they have
designed a different way of doing business. They help retailers to distribute products. Also they
subsidize producers and motivate producers to disclose product information. The premium paid
by consumers goes directly to retailers.
In scenario 5, producers form an alliance called Producer Association. Producers create a
platform of their own to distribute products. They share costs and receive premiums from
consumers.
We built system dynamics models that simulate these five scenarios, which are described in the
following section.
4. Model Description
Figures 1 to 5 are abstracted views of our model structures for scenarios introduced in the
previous section, which illustrates the main causal loops that are operating in the simulated
system. The final running model (please see the supporting material) is more complicated than
the high level view provided in Figures 1 to 5. In each scenario, there is a sector to describe each
agent. For example, agents in the Producer Association scenario include consumers and
producers, so for this scenario there is a sector dedicated to consumers, and another to producers.
In the TallMart scenario, there are three sectors, one to describe producers, another sector to
describe consumers, and the third sector for TallMart. Except for corporate entrepreneurs, e.g.,
Virtual Certififer, TallMart, etc., the model assumptions of agent behaviors are the same across
these five scenarios. The sector for the corporate entrepreneur depicts how information and cash
flow among various agents as a result of different business-model-element designs.
To avoid an overly complicated model structure and focus on studying the influence of business-
model-element design on the diffusion process, we have made simplified assumptions of agents
at the current stage of model development. Consumers in our current model make purchasing
decisions by evaluating information trustworthiness (‘information credibility” in the model). The
information trustworthiness will drive whether or not consumers keep buying the product in the
long run. The basic consumer sector structure is an adaption of the classic diffusion model. There
are two paths by which non-adopters would become adopters. One is through the influence of
word of mouth, and the other is through marketing. The degree of the influence of word of
mouth depends on information trustworthiness and the number of adopters, while the degree of
the influence of marketing is determined by available marketing budgets and the number of
adopters. Producers and retailers are rational economic agents, and their behaviors are governed
by evaluating benefits and costs. A high benefit to cost ratio will drive producers and retailers to
join the CIDIBB-based system, while a low benefit to cost ratio will make them leave the system.
Corporate Application Aviles
Entrepreneur: 1: Virtual Maing Cost Helen
Certifer (VC) Bulding Cost
Marketing E ee ie Combines ¢ EY
crevehre Ge ft D)
New VC Produces
rom Merveier Certifying Fees
+
+ Unit Certifying Fee Extra Sales
fom VC Unit VC
Premium + Piet
: y ~ Premimm
4.” ee Btra Profit fora
Producers @ VC Producer
New VC Producers (— "
fromCompetition \7
Figure 1 Causal L oop Diagram of the Virtual C ertifier Scenario
Corporate Entrepreneur 2: et
Consumer Advocate (CA) Gobsumer
i, CA Consumers
yo 8 froin Matheny,
CA Marketing vam aoe
Budget /> itt +
as Profit forthe h)
niece es A ae
+
CA Product as
p* Fees
+
Application Ca Cggt vase
Building Cost - subseeph on
Mak lic os WS *
Soneaon
‘ost CA Producer's Effort in
Information Disclosure
Figure 2 Causal L oop Diagram of the C onsumer A dvocate Scenario
A
Imitation
ASSET STOCK COORDINATION AND LONG-TERM
DYNAMICS IN LOVE AND COMMERCE
From Romeo and Juliet to People Express
Synopsis of a Talk
John DW Morecroft
London Business School
Regent's Park, London NW1 4SA United Kingdom.
+44 (0)20 7000 7000
jmorecroft@london.edu
Abstract
A fundamental idea in system dynamics is that interdependencies in business and society pre-
determine the performance over time of firms and industries. This idea is often summarised in
the phrase ‘feedback structure gives rise to dynamic behaviour’. To illustrate I use a selection of
well-known system dynamics models, spanning a range from Romeo and Juliet to People
Express. I review the feedback structure of the models and the dynamics that arise from different
ways of coordinating ‘operations’ and asset stocks. Using a variety of learning support materials
I analyse asset stock coordination dynamics. The style of model analysis combines visualisation,
and simulation with non-technical narrative interpretation of simulations to yield intuitively
appealing insight into performance paradoxes. This approach, which belongs in the tradition of
case-based system dynamics, helps modellers and executives to identify influential policy levers
and to implement practical policy changes that improve functional coordination and firm
performance.
Keywords: feedback structure, performance over time, behavioural decisionmaking, asset stock
coordination, style of model analysis, policy design.
Predicate: Small is Often Beautiful (in Models of Asset Stock C ordination)
It is a paradox of complexity that puzzling performance through time in business and.
society is often observed in tiny models containing only a handful of dynamical concepts. Even
the most basic dynamic process of stock accumulation is poorly understood (Booth Sweeney and
Sterman 2000).
Asset Stock Coordination in Love and Commerce
Asset stock accumulation is at the heart of change over time in business and society. All
business and social systems contain a variety of interlocking asset stocks or ‘resources’ from
which dynamics arise. To illustrate this vital idea I first review the structure and dynamics of the
well-known Romeo and Juliet simulator (Radzicki 1993 and Morecroft 2010). This tiny model
contains just two interlocking stock accumulations whose feedback structure generates elegant
cyclical dynamics - a pattem of behaviour over time that surprises many people.
At first glance it may seem that a tiny simulator of a love relationship is far removed
from business operations and strategy. But really it is not. To explain why, I select a stylised
system dynamics model of a manufacturing firm, and compare it visually with the Romeo and
Corporate an
Entrepreneur: 3: TallMart Maule on on Marketing Expense on
Mattaking Cos per TM Consumer
NeW TM Consumers
+ from Marketing
Marketing Expense on es os i TM Product ee
Rating Credential
perTM Producer ating Creder
‘TM Marketing YY be eal
TM Marketing Budget a,
on Producers a,
SalesfomTM
+
qs TM Distiuton
Ext TM Consumers
| per Product
New TM Producers
from Marketing
a Products Consumer per
— oe: Product Normal
DR tNew TallMart
c oi ftom
Etra Profitfora
TM Product, _+ TM Profitper
Pe Consuner per
‘TM Distibution
Fee per Product
™ he
Figure 3 Causal Loop Diagram of the TallMart Scenario
-
Comat (+f
. WS
Entrepreneur 4: iGuide Applicaton iGuide Marketing ——P> ‘Guide Marketing
iGuide Marketiig ibietig Cok Bueget Budget on Consumers
aintaiing Cos Application
Budget on Producers Buin Cost Marketing Expense on
0s :
iGuide Producer Effort in ra periBuife Co
Y. Information Dislosure
Products per Kg
Producer
iGuide Subsidy per N ew iGuide
Producer Cc 5 Consumers from
(: 4+) C Marketing
SY
VS
Subs iGuide Producer
Producers Subsidy
+ Prot or Ext Guide
Marketing Expense on ‘ + Consumer per Product
per iGuide Producer (' i Th
Sales from
iGuide
+ Distibution Fees +
+ ft iGuide Consumer i)
New iGuide Retailers periGuide Cruahc WU
from Marketi Products per
™ e iGuide Products Cs per
+ Product Normal
Ne
fe iGuide
iGuide Distribution iGuide Retailers Coes PTS iis
Fee per Product e 4+) Consumers from
i) = Imitation
EvaProitfore. =
iGuide Product + iGuide Profit per
Consumer per Product
Figure 4 Causal Loop Diagram of the iG uide Scenario
Markeung expense
Application on per Consumer
Application Maintaining Cost
Building Cost PA Marketing Budget.
+ sa -
: if New PA Consumers
PA Association PA Cost (— 4) from Marketing
Fee ie —_
SP ep No
ey
Association Fee per pe Profit for PA
Producer Prot or PA Producers + :
Producer + PA Cons P) Nei PA Consumers
4) # AE from Imitation
(. (+h Extra Sales for PA ” eee
— Producers _ +
PA Premimum = (Cf
wee | (oh
Products Buying by
per Consumer
Corporate Entrepreneur: 5:
Producer Association (PA)
PA Product Rating
+ Credential
Figure 5 Causal Loop Diagram of the iGuide Scenario
5. Simulation Results
This section reports our simulation results. The parameter values for base runs were calibrated
against empirical data collected in I-Choose project. We also conducted a series sensitivity
experiments to test the robustness of our model and the reliability of simulation results of base
runs.
5.1 Base Run Results
As shown in Figure 6, the base run result of the Consumer A dvocate scenario shows that the
number of consumers who are buying coffee from the platform gradually increases to 3000 (10%
of the total consumer population) and stops growing by the end of the 9" year after the platform
is launched into the marketplace.
° 10 20 30 40 30 60 7oO so 30 100
Figure 6 Base Run Result of the Consumer Advocate Scenario
As for the Virtual Certifier and iGuide scenarios, their base run results show that there are few
consumers who are buying coffee from the platform, and this situation has remained the same
ever since the platform is launched into the marketplace, as shown in Figure 7.
o.1
s 10 20 30 40 50 so 76 0 30° 100
Figure 7 Base Run Results of the Virtual C ertifier and iG uide Scenarios
Figure 8 shows the base run result of the TallMart scenario: In the first eight years after the
platform is launched into the marketplace, the number of LOHAS consumers who are buying
coffee from the platform remains at the level of 3000 (10% of the total LOHAS consumer
population). The LOHAS consumer number starts growing rapidly around the 9" year and
increases to 30,000 (100% of the total LOHAS consumer population) by the end of the 24" year.
30,000
° 6 12 18 24 30 36 42 48
8
Figure 8 Base Run Result of the TallMart Scenario
As for the Producer Association scenario, in the first 15 years after the platform is launched into
the marketplace, the number of consumers who are buying coffee from the platform remains at
the level of 3000 (10% of the total consumer population). The consumer number starts growing
rapidly around the 16” year and increases to 30,000 (100% of the total consumer population) by
the end of the 33" year, as shown in Figure 9.
30,000
22,500
15,000
7,500
°o
es Ss 12 18 24 30 36 42 4s 54 so
Figure 9 Base Run Result of the Producer Association Scenario
5.2 Sensitivity Test Results
We normalized model parameters and conducted sensitivity experiments on a series of
parameters to strengthen our confidence in the model and test the reliability of simulation results
of base runs. These parameters include the ratio of the number of producers to the number of
consumers, the ratio of the number of producers to the number of retailers, the market size, the
ratio of the size of customer base to fixed Cost, the marginal cost for a producer or retailer, the
rating-subscription fee per consumer, and the subsidy per producer. Detailed experiment results
are as follows.
Sensitivity tests on the ratio of the number of producers to the number of consumers: As for the
iGuide and Virtual Certifier scenarios, test results show the same diffusion pattern. The market
behavior is hardly influenced by the variation of the ratio of the number of producers to the
number of consumers, as shown in Figure 10. As for the TallMart and Producer Association
scenarios, test results show the same general diffusion pattern as the ratio of the number of
producers to the number of the consumers changes within a certain range, however, as the ratio
gets bigger in this range, the initial growth rate of the rapid- growth period becomes smaller. As
the ratio of the number of producers to the number of the consumers changes outside of this
range, the diffusion process cannot take off, as shown in Figure 11.
ict Pp: atio
Base
50% 75% fi osec J 100°
LOHAS Consumers Buy Coffee from the Platform
0.1
0.075
° 25 50 75 100
Figure 10 Sensitivity Tests on Producers to Consumers (iG uide and Virtual C ertifier)
sen-pro-corr-ratio
Base
50% —- 75%} 95° 100°. ay
LOHAS Consumers Buy Coffee from the Platform
30,000
15,000
0 £
wh
30 4
oO
60
Figure 11 Sensitivity Tests on Producers to Consumers (TallMart and Producer
Association)
Sensitivity tests on the ratio of the number of producers to the number of retailers: As for the
iGuide scenario, test results show the same diffusion pattern. The market behavior is hardly
influenced by the variation of the ratio of the number of producers to the number of retailers, as
shown in Figure 12. This test is not applicable for other scenarios, since there is no retailer agent
in other scenarios.
i. rep ti
Base
50% 75% (95° NN 100°
LOHAS Consumers Buy Coffee from the Platform
O14
0.075
25 50 75 100
Figure 12 Sensitivity Tests on Producers to Retailers (iG uide)
Sensitivity tests on the market size: As for the iGuide, Virtual Certifier (as shown in Figure 13),
and Consumer A dvocate (as shown in Figure 14) scenarios, test results show the same diffusion
pattern. The market behavior is hardly influenced by the variation of the market size. As for the
TallMart and Producer Association scenarios, test results show the same general diffusion
pattern as the market size changes between a certain range, however, as the market size gets
bigger in this range, the take off defers. As the market size changes outside of this range, the
diffusion process cannot take off, as shown in Figures 15 to 18.
gude-sen-scale
Base
50% 75% NN 95° NN 100° EB
LOHAS Consumers Buy Coffee from the Platform
0.1
0.075
ce) 25 50 75 100
Figure 13 Sensitivity Tests on Market Size (iG uide)
Base
50% 75% (95% J 100°.
LOHAS Consumers Buy Coffee from the Platform
4.000
3,000
2,000
1,000
0 25 50 75
Figure 14 Sensitivity Tests on Market Size (Consumer A dvocate)
tallmart-sen- scale
50% -75°%fos°< [100°
LOHAS Consumers Buy Co ffee ffom the Platform
300,000
225,000
150,000
75,000
0 1
O
30 4
oO
60
Figure 15 Sensitivity Tests on Market Size (TallMart)
oO 6 12 is 24 30 36 42 48 54 60
Figure 16 Sensitivity Tests on Market Size (TallMart)
P
Base
50% —- 75% I 95° I 100%
LOHAS Consumers Buy Coffee from the Platform
2M
500,000
Oo 15 30 45
Figure 17 Sensitivity Tests on Market Size (Producer Association)
oO 6 12 18 24 30 36 42 48 54 60
Figure 18 Sensitivity Tests on Market Size (Producer Association)
Sensitivity tests on the ratio of the size of consumer base to fixed cost: As for the iGuide, Virtual
Certifier (as shown in Figure 19), and Consumer A dvocate (as shown in Figure 20) scenarios,
test results show the same diffusion pattem. The market behavior is hardly influenced by the
variation of the ratio of the consumer number to the fixed cost. As for the TallMart and Producer
Association scenarios, test results show the same diffusion pattern as the ratio of the consumer
number to the fixed cost changes in a certain range, however, as the ratio gets smaller in this
range, the take off defers. As the ratio of the consumer number to the fixed cost changes outside
of this range, the diffusion process cannot take off, as shown in Figures 21 to 22.
id
Base
50% 75% (os°- MB 100°:
LOHAS Consumers Buy Coffee from the Platform
oO.1
0.075
o 25 50 75 100
Figure 19 Sensitivity Tests on Consumers to Fixed C ost (iG uide and Virtual C ertifier)
ca- fixed ‘atio
$cc 75> 957 M1007
LOHAS Consumers Buy Coffee from the Platform
4,000
3,000
o 25 30 75 100
Figure 20 Sensitivity Tests on Consumers to Fixed C ost (Consumer A dvocate)
° ,
Base
50% 75% IN OS° INN 100%
LOHAS Consumers Buy Coffee from the Platform
30,000
oO 15 30 45 60
Figure 21 Sensitivity Tests on Consumers to Fixed C ost (TallMart and Producer
Association)
Juliet model, I demonstrate striking similarity in the feedback structure of the two models. This
visual similarity helps to explain why there is identical cyclicality in accumulations of love,
inventory and workforce. I then review common and enduring features of the two coordinating
networks that define the relationship between lovers on the one hand and functional areas of the
firm on the other.
Operating Policy, Bounded Rationality, Dominant Logic and Asset Stock C coordination
l use structural insights from the factory model to envisage changes to standard operating
policies (such as inventory control, forecasting and workforce planning) that, when implemented,
should improve cross-functional coordination and boost business performance. Beneficial
changes normally modify information flows within and between functions of the firm. There
follows a brief review of the information processing assumptions behind operating policies. The
degree of success that firms achieve in asset stock coordination reflects the bounded rationality
of decision makers, the psychological environment in which their decisions are made and the
resulting ‘dominant logic’ of each operating policy.
The Rise and Fall of People Express
From the discussion of operating policy and dominant logic it is but a small step to a
behavioural understanding of covert asset stock coordination problems that lay behind the
dramatic rise and fall of People Express (a visionary low-cost start-up airline) in the US airline
industry of the 1980s.
References
Booth Sweeney L, Sterman JD (2000). Bathtub dynamics: initial results of a systems thinking
inventory. System Dynamics Review 16(4): 249-286.
Morecroft JDW (2010). Romeo and Juliet in Brazil: Use of Metaphorical Models for Feedback
Systems Thinking, Chapter 6 in Tracing Connections: Voices of Systems Thinkers, an edited
volume in tribute to Barry Richmond, isee systems publications, Lebanon NH and The Creative
Leaming Exchange, Acton MA
Radzicki MJ 1993. “Dyadic Processes, Tempestuous Relationships, and System Dynamics,”
System Dynamics Review, 9 (1), 79-94.
List of Learning Support Resources Mentioned or Used in the Talk
Morecroft JDW (2015). Strategic Modelling and Business Dynamics: A Feedback Systems View
(2™ edition), Wiley, Chichester.
- The Romeo and Juliet simulator is in the online leaming support folder for Chapter 10.
- Stylised models of a manufacturing firm are described in Chapter 5 of the book and are
available as simulators in the online leaming support folder for Chapter 5.
- Materials on policy structure and bounded rationality are in Chapter 7 of the book
- Materials and mini-simulators for the rise and fall of People Express are in the online
leaming support folder for Chapter 6.
0 10 20 30 40 50 60 70 80 90 100
Figure 22 Sensitivity Tests on Consumers to Fixed C ost (TallMart and Producer
Association)
Sensitivity tests on the marginal cost for a producer or retailers show the similar results as the
sensitivity test result on the ratio of the size of consumer base to fixed cost: As for the iGuide
and Virtual Certifier scenarios, test results show the same diffusion pattem. The market behavior
is hardly influenced by the variation of the marginal cost for a producer or retailer. As for the
TallMart and Producer Association scenarios, test results show the same diffusion pattern as the
marginal cost for a producer changes in a certain range, however, as the marginal cost gets
bigger in this range, the take off defers. As the marginal cost for a producer changes outside of
this range, the diffusion process cannot take off. Sensitivity tests on the rating-subscription fee
per consumer and the subsidy per producer show the same diffusion pattern. The market
behavior is hardly influenced by the variation of these parameter values.
6. Discussion and C onclusion
Our base run simulation results show that the diffusion path varies in these five scenarios,
although agent-behavior assumptions are the same across scenarios. For the Consumer A dvocate
scenario, the number of consumers who are buying coffee from the platform increases and then
stops growing. For the Virtual Certifier and iGuide scenario, the system never takes off. For the
TallMart and Producer Association, the number of consumers grows and eventually reaches the
saturation. Sensitivity test results show the same general diffusion patterns as a series of
parameter values change. In brief, our simulation results show that different designs of business-
model elements (i.e., the value proposition, customer segments, the revenue generation
mechanism, the cost structure, and the partner network) lead to different diffusion paths.
A possible explanation for this could be different designs of business-model elements result in
different system structures within the simulation boundary, which in turn lead to different system
behaviors (diffusion paths). We speculate some designs are superior to other designs since they
will form a structure that has moderate negative effects to hinder the market take-off. If we try to
20
capture the structure difference between scenarios that take off and scenarios that do not take off,
it seems that when the business model represents a one-way relationship, it has a poor
performance in realizing the market take-off. In the iGuide scenario, iGuide receives distribution
fees from retailers, retailers receive premiums from consumers, and iGuide subsidizes producers.
This scenario can be visualized as one-way relationship:
consumers~retailers>iG uide> producers. In the Virtual Certifier scenario, Virtual Certifier
receives distribution fees from producers, and producers receive premiums from consumers. This
scenario can also be visualized as a one-way relationship: consumers producers virtual
certifier. Similarly, in the Consumer A dvocate scenario, Consumer A dvocate receives rating
subscription fees from consumers: consumers consumer advocate, also a one-way relationship.
In contrast, in the Producer Association scenario, producers share costs, and receive premiums
from consumers. Producers form an alliance and interact with consumers. This scenario can be
visualized as a multi-sided relationship: consumers producers; producers€~ producers. In the
TallMart scenario, TallMart receives distribution fees from producers, receives premiums from
consumers, and pays dividend to producers. This scenario can also be visualized as a multi-sided
relationship: producers>TallMart<consumers; TallMart> producers. Nevertheless, these ideas
need to be further investigated and tested.
Our next steps include an in-depth explanation of our findings by examining the system structure
of each scenario, more comprehensive scenario design based on a business-model typology to
include various types of business models, looking for empirical cases to support building
simulation models that represent newly designed scenarios, and further investigating the
relationship between diffusion related market behaviors and structure features derived from
business-model designs.
References
Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human
Decision Processes, 50, 179-211.
Amit, R., & Zott, C. (2001). Value creation in e-business. Strategic management journal, 22(6-7),
493-520.
Bass, F. M. (1969). A new product growth model for consumer durables. Management Science
15, 215-227.
Brown, L. (1981). Innovation diffusion. New Y ork: Methuen.
Bruner, G. C., Il, & Kumar, A. (2005). Explaining consumer acceptance of handheld intemet
devices. J ournal of Business Research, 58(5), 553-558.
Chen, J. V., Yen, D. C., & Chen, K. (2009). The acceptance and diffusion of the innovative
smart phone use: A case study of a delivery service company in logistics. Information &
Management, 46, 241-248.
21
Chesbrough, H. (2010). Business Model Innovation: Opportunities and Barriers. Long Range
Planning, 43, 354-363.
Chesbrough, H., & Rosenbloom, R. S. (2002). The role of the business model in capturing value
from innovation: evidence from X erox Corporation's technology spin-off companies.
Industrial and corporate change, 11(3), 529-555.
Cole (1949). A purposeful activity to initiate, maintain and aggrandize a profit-oriented business.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User Acceptance of Computer
Technology: A Comparison of Two Theoretical Models. Management Science, 35(8),
982-1003.
Dattee, B., FitzPatrick, D., & Weil, H. B. (2007). The dynamics of technological substitutions. In
Proceedings of the 25" International Conference of the System Dynamics Society.
Dattee, B. & Weil, H. B. (2005). Dynamics of social factors in technological substitutions.
Massachusetts Institute of Technology, Sloan School of Management (Working Paper
4599-05).
Dubosson-Torbay, M., Osterwalder, A., & Pigneur, Y . (2002). E-business model design,
classification, and measurements. Thunderbird International Business Review, 44(1), 5-
23.
Eastin, M. S. (2002). Diffusion of e-commerce: an analysis of the adoption of four e-commerce
activities. Telematics and Informatics, 19, 251-267.
Gruhl, D., Guha, R., LibenNowell, D., & Tomkins, A. (2004). Information Diffusion Through
Blogspace. Paper presented at the 13th international conference on World Wide Web
Harno, J. (2010). Impact of 3G and beyond technology development and pricing on mobile data
service provisioning, usage and diffusion. Telematics and Informatics, 27, 269-282.
Homer J. A Diffusion Model with A pplication to Evolving Medical Technologies. Technological
Forecasting and Social Change, 31(3): 197-218, 1987.
Hung, S.-Y., Ku, C.-Y., & Chang, C.-M. (2003). Critical factors of WAP services adoption: An
empirical study. Electronic Commerce Research and Applications, 2(1), 42-60.
Knutsen, L. A., & Lyytinen, K. (2008). Messaging specifications, properties and gratifications as
institutions: How messaging institutions shaped wireless service diffusion in Norway and
Japan. Information and Organization, 18, 101-131.
Liao, C., Chen, J.-L., & Yen, D. C. (2007). Theory of planning behavior (TPB) and customer
satisfaction in the continued use of e-service: An integrated model. Computers in Human
Behavior, 23, 2804—2822.
Luna-Reyes, L.F., Andersen, D.F., Andersen, D.L., Derrick, D., & Jarman, H. (2012). Full
Information Product Pricing (FIPP) Regimes: Policy Implications for US-Mexico
Sustainable Commerce.
Luna-Reyes, L., Sayogo, D. S., Zhang, J., Pardo, T., Tayi, G. K., Hrdinova, J., et al. (2012).
Beyond Open Government: Ontologies and Data Architectures to Support Ethical
Consumption. Proceedings of the 6th Intemational Conference on Theory and Practice of
Electronic Governance (ICEGOV), Albany, USA.
22
Mahadevan, B. (2000). Business models for Internet-based e-commerce. California management
review, 42(4), 55-69.
Miller, D., & Garnsey, E. (2000). Entrepreneurs and technology diffusion: How diffusion
research can benefit from a greater understanding of entrepreneurship. Technology in
Society, 22, 445-465.
Oliver, R. L. (1993). Cognitive, Affective, and Attribute Bases of the Satisfaction Response.
Journal of Consumer Research, 20, 418-430.
Osterwalder, A. (2004). The business model ontology: A proposition in a design science
approach. /nstitut d’Informatique et Organisation. L , Switzerland, University of
Lausanne, Ecole des Hautes Etudes Commerciales HEC, 173.
Pavlou, P. A., & Fygenson, M. (2006). Understanding and predicting electronic commerce
adopton: An extension of the theory of planned bahavior MIS Quarterly, 30(1), 115-143.
Rogers, E. M. (1995). Diffusion of Innovations (4th ed.). New Y ork: Free Press.
Ruyter, K. d., Wetzels, M., & Kleijnen, M. (2001). Customer adoption of e-service: an
experimental study. International Journal of Service Industry Management, 12(2), 184-
207.
Schumpeter, J. (1947). The creative response to economic history. J Econ Hist.
Timmers, P. (1998). Business models for electronic markets. Electronic markets, 8(2), 3-8.
Timmers, P. (1999). Electronic commerce. John Wiley & Sons, Inc..
Vijayasarathy, L. R. (2004). Predicting consumer intentions to use on-line shopping: the case for
an augmented technology acceptance model. Information & Management, 41(6), 747—
762.
Weil, H. B., & Utterback, J. M. (2005). The Dynamics of Innovative Industries. In Proceedings
of the 23 International Conference of the System Dynamics Society. Boston, MA.
Wiklund (1998). Taking advantage of opportunity by novel.
Yi, M. Y., & Hwang, Y. (2003). Predicting the use of web-based information systems:
selfefficacy, enjoyment, learning goal orientation, and the technology acceptance model.
International J ournal of Human-Computer Studies, 59(4), 431-449.
Zott, C., Amit, R., & Massa, L. (2011). The business model: recent developments and future
research, J ournal of Management, 37 (4), 1019 - 1042.
23
