The Smart Home Services Diffusion Process: A System Dynamics Model
Giovanni Zenezini ', Maliheh Ghajargar '2 Eleonora Fiore *, Alberto De Marco !
' Politecnico di Torino, Department of Management and Production Engineering
* Politecnico di Torino, Department of Architecture and Design
giovanni.zenezini@polito.it, maliheh.ghajargar@polito.it, eleonora.fiore@polito.it,
alberto.demarco@polito.it
Abstract
The application of smart technologies for domestic environment has been around for a while. But the
market diffusion of such products and services has not seen yet a significant growth. This paper
seeks to provide an overview of the most important factors that influence the diffusion process of
smart home services via literature and a case study. These factors compose a System Dynamics
model showing the diffusion dynamics of three main smart home services (Heating, Monitoring,
Assisted Living).
1. Introduction
During the last two decades, Information Technology and Internet connectivity have been more and
more integrated in people’s daily life. This changed the way people live, work, entertain and
communicate to each other. (Weiser’s Ubiquitous Computing 1991, Ashton’s Internet of Things
2009, Ericsson’s Connected Homes 2015, Fjord’s Living Services 2015, etc.). A promising
application sectoOr of this new paradigm is represented by the home automation, or Smart home. For
instance, a study conducted in 2014 by IDC, a global source of technology market intelligence,
showed that 69 percent of all consumers plan to buy a home automation device in the next five years.
A smart home in literature is defined as a domestic environment equipped with computing and
information technology, which addresses householders’ needs such as comfort, control, convenience,
security and entertainment (Aldrich, 2003) (e.g. switching on/off the heating, lighting on/off,
controlling the appliances remotely and etc.). Smart homes are able to provide users with a large
variety of customizable services in order to tailor the domestic environment to their needs
(Ricquebourg et al., 2006). The main advantage of smart home services is the ability to be
preemptive, ie. being able to predict any inconvenience and errors, thus avoiding unpleasant
drawbacks to the user (Allmendiger and Lombreglia 2005). Therefore the applications of smart home
can be divided in four categories: Energy Saving; Support for elderly or disabled; Security and
safety; User convenience, among them the user convenience has long been associated with smart
services (Holroyd P., Watten and Newbury 2010).
Despite the fact that smart home technologies have been around for a while, their prevalence and
diffusion is not widespread, thus their potential is largely untapped. The greatest obstacles to the
diffusion of smart home applications are related to: (1) difficulty of integration in existing
households; (II) usability, learning and reliability; (III) costs (Tumino, 2015). Hence, the purpose of
this paper is to investigate on some of the factors that might enhance the diffusion process of smart
home services. Consumers today are exposed to a wide range of influences that include word-of-
mouth communications, network externalities, and social signals. In this context, research on
diffusion modeling seeks to understand the spread of innovations throughout their life cycle, and
4
therefore has adapted to describe and model these influences (Peres et al.2010). As Rogers (2003)
pointed out, the diffusion of innovation is the process by which an innovation is adopted by a society
over time. Kalish (1985) devised an innovation model from behavioral theory, composed by two
steps: awareness and adoption. In this model, during the awareness phase an innovation spreads in an
epidemic-like way, and is generated by advertising and word of mouth.
The diffusion of technological innovations follows some complex, unexpected and unpredictable
dynamics. For ICT companies it is important to adopt a holistic approach, in order to be able to
manage the sudden changes in market demand. In this context, SD has proved itself to be an
appropriate approach to study the complexity of innovation diffusion processes and a “suitable
instrument for decision support in innovation management” (Maier, 1998), precisely because
“innovation management comprises all the ingredients of complexity: a large number of variables
involved; tightly interrelated in non-linear fashions; and highly dynamic” (Milling, 2002).
In order to contribute to the literature in the application of SD methodology to the field of innovation
diffusion,this paper presents an on-going project, carried out in collaboration with Swarm Joint Open
Lab of TIM (Telecom Italia S.p.a.) and proposes a System Dynamics model concerning the
awareness phase of the diffusion process of selected Smart Home Services.
The paper is structured as follows. In the next section, a literature review is provided on two main
streams of research: smart home services and diffusion modeling with System Dynamics. Then, the
case study comprising a description of the three services is depicted in Section 3. The SD diffusion
model development is presented in in section 4. In section 5 the practical implications and possible
use of the model are depicted, and in section 6 discussions and conclusions are drawn.
2. Literature Review
This literature review is divided in two parts. The first part is related to the smart home research
arena, and it provides both a contextualization of our research work and the identification of the most
promising services. The second part instead provides the theoretical background for the diffusion
model, building from previous efforts in the field of diffusion modeling with System Dynamics.
2.1. Related studies on smart home services diffusion
The term smart home has been coined by the National Association of Home Builders (NAHB) in the
early 1980s after it set up a group to work for the diffusion of smart technologies in the design of
new homes. However, the smart home life envisioned by many researchers since 20th century has
yet to deliver. Most of the people are still living in homes more similar to their grandparents’ rather
than those conceived by the early smart home pioneers (Harper, 2003).
The diffusion rate of smart home services and the factors that can influence it has been the subject of
many researches in academic and commercial domains.
Balta-Ozkan et al. (2013; 2014) studied different European smart home markets, which are
characterized by different policy and socio-economical contexts and reveals the key barriers to the
diffusion of smart home services such as interoperability, reliability, data privacy, control and the
cost of the smart home technologies. They illustrate how the age of the building, technical and
economical drivers of context can facilitate or create barriers to the diffusion of such services and
finally they highlight the need of adopting a more holistic and integrated approach to smart home
2
services. Wherein smart home services are not limited just to the application of the energy
consumption and management products, but also integrated to the other applications such as health,
security, assisted living, which can be personalized according to user's need and, thus can bring
positive changes in people’s life. Ehrenhard et al. (2014) seek to identify the barriers that are keeping
the smart home services limited to a small, luxury and segmented market with stand-alone
technologies, despite its potentialities to deliver values to a much bigger market. These barriers are
related to meeting end-user requirements, platform management, improved value creation and the
role of the government. Furthermore, the organizational and market aspects of Smart Home platform
diffusion from a business ecosystem perspective have been investigated in order to provide a generic
value network for such platforms (Cusumano, 2010).
In summary, the most important key barriers of the diffusion of smart home services provided by
literature are related to the areas of service, user or context. (Table 1)
Table 1: Key Barriers of the smart home services
Area Key Barriers/Facilitators
Service-related | Interoperability
Reliability
Control of the service
Cost of the service
User-related Data Privacy
Value creation for end-user
Context-related | Technical, economical and social drivers of
context
The features of the building (age, value, etc.)
Policy-related Platform management
Policy making (the role of the government)
2.1.1 Customer motivations — end-user studies
While some sectors, such as security or air conditioning are widely covered by current services, other
areas are still uncovered. In particular, studies show that there is a huge gap between consumer
requirements and the products currently available on the market related to energy management for
smart home. In particular the delta between the functions undertaken by current smart home services
and the functions expected by users is higher than 10% for heating, energy consumption and the
remote control of the appliances. The major concern of the smart services for home environment is
related to the management of everyday tasks, labor saving and task simplification, ease of operation,
remote control and cost reduction (Aldrich, 2003).
Ericsson Consumer Lab (2015) conducted a qualitative and quantitative user research among 1,000
individuals around the world. The result of this study confirms that control, security, interest in new
technology and the ability to make life easier are the greatest motivators for the diffusion and usage
of connected or smart home services. Cost efficiency has not been a motivator factor (5%), as long as
customers need yet to realize the cost saving. Among the factors in more detail, customers declared
that the internet connectivity through household wireless connections is a motivator to adopt the
technology as long as they are annoyed by seeing appliances cables around their homes; this is also
related to the aesthetic factors of the product that might be considered. This study reports also that
five individuals out of ten (+50%) show interests toward such services, four out of ten would like to
have an integrated connected home service and most of them need services and products related to
their health and wellbeing. Women living in large households particularly express high needs for
connected home services.
In another similar user-study conducted in 2015, through an online survey by Osservatori.net in
collaboration with Doxa, among 1,000 final users, it was highlighted that 59% of them would
consider to subscribe services based on information collected by smart objects, with important
evidence on safety and damage prevention, but also on remote medical care and remote assistance in
case of accidents.
2.2. Modeling the diffusion of innovation with SD
Maier (1998) considers a set of four relevant elements for the diffusion of innovation, one of which
can be controlled by the managers. In particular, pricing, advertising, the quality of the product and
the delivery delays due to insufficient manufacturing capacity are among the elements that are
influenced by managerial decisions. Milling (2002) adds that the diffusion of innovation largely
depends on the knowledge of the underlying technology, which he regards as the result of an
evolutionary process. The maturity of the technology is instead a key innovation driver for Tsai and
Hung (2014), along with service maturity, price, perceived risk and the macro-economic situation.
SD has been already applied in various domains of research with the purpose of observing and
exploring diffusion behavior of an innovation system. For instance it is applied to diffusion process
of energy efficiency lighting in households (Timma et.al., 2015), on alternative fuels vehicles in
order to predict the future market of such vehicles (Shen and Ma, 2013), and to the innovation
diffusion of multi-generation products (Lo, et.al., 2011). While Kreng and Weng (2013) introduce an
integrated multi-generation diffusion model, thus considering dynamics of the potential market
accounting the relationship among generations and products, Ryan and Tucker ‘s work (2009)
focused particularly on the diffusion of the videocalling technology. In the latter, authors investigates
how different types of heterogeneity (e.g. adoption cost, network effects, usage) affect network
evolution and so the diffusion of the product. Finally, Tsai and Hung (2014) apply system dynamics
to investigate the diffusion of cloud computing.
The coaching reference model for these previous works as well as this research is the Bass diffusion
model (1969). In this model, users can adopt first as a consequence of the advertising by the
company that provides the services:
Adoption rate from advertising = a*P Equation 1
Where P is the population of Potential adopters at each time step and a is the advertising
effectiveness (adoption fraction from advertising).
The adoption from word of mouth can be modeled as follows:
Adoption rate from word of mouth = c*i*P*A/N Equation 2
Where c is the contact rate between individuals in a population, measured in people contacted per
person per time period, i is the adoption fraction from word of mouth, A is the population of
Adopters, and finally N is the total potential market. Potential adopters generate as much as cP
contacts for each period. Potential adopter interact with adopters and the probability that each
potential adopter has to get in contact with an adopter is the fraction of total adopters A/N.
Therefore, the total amount of interactions between adopters and potential adopters for each time
period is cPA/N. However, only a portion i of these interactions turns out as a successful adoption.
3. Smart home services
This research has been proposed by Swarm Joint Open Lab, a group within TIM Open Innovation, to
challenge the modeling of the diffusion of three main services related to Smart Home: Monitoring,
Heating and Assisted Living. The goal was to investigate how these services could be spread into the
market, with a particular focus on the Italian customers, also taking care of transversal service
interactions (cross-selling) and different factors that might influence this adaption, whether user-
related, context-related or service related. And finally to understand the interaction between the
adopted service and others.
The smart home services under examination are depicted in the following subsections.
Heating service: The energy consumption in buildings has steadily increased since 2008. In
particular, buildings sectors energy consumption has grown at a faster rate than the industrial and
transportation sectors and it requires a considerable amount of the primary energy usage in
developed countries, representing 20% to 40% of the total primary energy consumption (Perez-
Lombard et al., 2008). Thus the performance and efficiency of space heating systems become crucial
not only for improving inhabitants’ comfort but also for reducing energy usage purposes (Ren,
2015).
The service consists in the remote control of the temperature and the possibility to save on energy
bills by avoiding an overheating when the user is out and enabling preheating of the space for user’s
thermal comfort when come back home.
Monitoring service: Home security and personal safety are major concerns for individuals. People
want to protect their valuables and provide a safe haven for family members and loved ones.
Traditional home security systems generally alert the neighborhood with a loud noise warning the
intruder(s) that the invasion has been detected. In addition, home alarms generally inform a home
security central system of the unauthorized entry. The home security central system then may alert
the police and/or third party security companies. Home security devices generally involve a kit of
window detectors, door detectors, motion sensors and other devices (Saylor et al., 2003).
The service within the modelling framework consists in a system and method for connecting a
security system to a wireless communication system to automatically inform the owner and other
authorized entities. This function must be predefined by the user when alarm worthy situations
occur. It allows the individuals to remotely monitor the internal and external of the house at any
time.
Assisted Living service: The increased life expectancy and the growth of the older adult population
have led to new models of aging that empower people to fulfill their lives in their homes (Demiris,
2008). (Pragnell et al., 2000). Researchers have demonstrated that the smart home information
technologies (IT) in residential care (RC) facilities are performing tools to enhance resident quality
of life and safety (Courtney, 2008). Smart home services aim indeed to be a promising and cost-
effective way of improving home care for the elderly in a non-obtrusive way, allowing greater
independence, maintaining good health and preventing social isolation. (Chan et al. 2009).
The service provides assistance for elderly in their home, throughout the day by using devices, the
fast and direct connection to caregivers and the emergency services such as the ambulance.
4. Model development
The main objective of this research project is to understand the levers that enable the adoption
growth of the smart home services by users. To this end, we adopted the Bass diffusion model for
each of the three services. However, the Bass diffusion model has been adopted only as a funding
reference. Our objective is in fact to model the awareness on the service, rather than the adoption.
Hence, we argue that for a very innovative service advertising works as a factor enhancing
awareness in the first phase, and only in a second phase as an adoption factor. Moreover, we do not
introduce the price of the service in our model, given the fact that no historical data is present for
fitting the model, and furthermore price usually is not a factor for increasing awareness during the
diffusion process (Kalish, 1985).
To the original Bass model we introduce the notion that the services are connected so that the
diffusion of one service may support the diffusion process of another one. We model this by
introducing a cross-selling adoption variable, repeated for each service. Finally, we model the
awareness process on the total number of households rather than individuals.
4.1. Structure of the model
The proposed model integrates the basic structure of the Bass diffusion model with the cross-selling
variable. The structure of the model is shown in Figure 1.
! Adoption from
Adopters
service 1
T adoption rate 7
7 .. A 7:
\ Adoption
fa nee /
*. advertising ;
tm Adoption from.
{ Word of Mouth :
Figure | - General structure of the model
In our model, the adoption from advertising resembles the definition by Bass, as seen in Figure 2.
tial adopters Adopters
service 1 service 1
adoption rate hee
\., Adoption from 7
advertising
service 1
\
Advertising
effectiveness
Figure 2 - Adoption from Advertising
Adoption from word of mouth (Figure 3) works similarly to the Bass diffusion model. However, we
have computed two different potential markets depending on the features of each service.
Nevertheless, this service can be delivered only to households with a Wi-Fi connection; hence for all
three services the initial potential market is the following:
N =%wi— ficonnected families * H Equation 3
Where H is the total households. However, a further refinement was made for the Assisted Living
(AL) service, as per equation:
AL potential market = N * % of households with elderly people
Potential adopters =z Adopters
service 1 service 1
adoption rate a
\
"Sa Adoption from
Word of Mouth My
\
adoption
fraction
Figure 3 - Adoption from Word of Mouth
Finally, the adoption from cross-selling should represent the users that adopts one service if they
have successfully adopted, used and liked another one. Adoption from cross-selling works with
similar dynamics for each of three service. In fact, the adoption of one service supports the adoption
of another one by means of a parameter named degree of compatibility. This parameter measures the
frequency to which users integrate the already adopted service with another one.
The adoption from cross-selling is shown in Figure 4.
Degree of
-ompatib
adopter otoption
Figure 4 - Adoption from Cross-selling
Service 1 Adoption rate from cross — selling from service 2 =
Cross — selling Service 2 — Service 1 * Service 1 Adoption fraction Equation 4
Where Cross-selling service 2-service 1 represents the amount of users that have already adopted
service 2 and might integrate with service 1 as well. This is equal to:
Cross — selling service 2 — service 1 =
Adopters Service 2 * Degree of compatibility Service 2 — Service 1
Equation 5
4.2. Adoption from Word of Mouth
For this component of the diffusion model, the original Bass diffusion model was integrated with the
adoption factors peculiar of the service that could enable or hinder its adoption. These factors are
related to the different features and scope of the three services. In fact, they aim to appease different
users’ needs and hence their adoption is driven by various motivations which have to be taken into
account for the model development.
In our model, these differences are accounted to model the adoption fraction, which is not a unique
figure but rather the parametric result of a combination of various effects.
For instance, the Heating service will be as captivating to users as its effectiveness in terms of energy
consumption reduction and level of comfort provided. Hence, the heating adoption fraction is the
summation of two parameters, as seen in equation 6:
e Total energy net savings, measured as the difference between the kwh saved and the price of
the service;
e The level of comfort provided, measured as a dimensionless parameter
i (Heating) = Comfort + Net savings Equation 6
The value of these parameters can be studied from two different perspectives.
On a micro-level made up by the individual adopter, they may represent the relative preferences of
one factor over the other. That is, the threshold level of one parameter over which the individual
potential adopter will not adopt the service. For instance, is there a Comfort level sufficiently high,
able to counterbalance a negative Net savings, so that the service can still be attractive to the user?
On the other hand, on an aggregate market-level composed by all potential adopters, the values of
the parameters might measure the share of users that make their adoption looking at one parameter
over the other one.
The heating adoption from WoM is shown in Figure 4.
Heating Adoption
from Word of Mouth
*
- \
Potential market a
ye ie
ff 4 ,
/ i
‘otal \ Contact rate _—
households Share of wi-fi A
connected families
Net savings
~*~ \
\ Comfort
/ \
| Price Heating
kwh saved
Figure 4 - Heating Adoption from Word of Mouth
The second service in the diffusion model is the Monitoring service. As stated, it uses several sensors
and camera to provide general information about the activities of people and resources in the home
environment. This means that it will be effective on the expense of a little bit of privacy that the user
Squaring the sunny circle? On balancing incentives for solar
prosumers and cost causation
Author: Merla Kubli *?*
“Institute of Sustainable Development, Zurich University of Applied Sciences, Switzerland
“Institute for Economy and the Environment, University of St. Gallen, Switzerland
The manuscript is currently in publication process and cannot be published in full length. For a personal
copy please contact the author.
Abstract
Solar prosumers are about to revolutionize the power sector. Utilities are challenged in recovering the
costs of distribution grids, as parts of their revenue basis decreases through self-consumption. Adjusting
the grid tariff sets off a reinforcing feedback loop that increases the attractiveness of solar in
but also leads to a distribution effect between solar prosumers and conventional consumers. The
question is: How to recover distribution grid costs equitable without hampering the diffusion of solar
power? Can the two criteria be fulfilled at the same time, or is do we aim for squaring a circle? To
d to understand the
address this question, I present a System Dynamics simulation model de
interactions and assess these competing goals. The occurring distribution effect under the volumetric
grid tariff appears to be rather limited. Simulation experiments reveal that grid tariff designs strongly
influence investments for solar power. A capacity tariff can reduce deviations from the cost causation
li: I to reduce
principle of solar prosumers and incentivizes in in dec / storage
peak demand. Nevertheless, also the capacity tariff causes a distribution effect.
*Author contact: Merla Kubli, Institute of Sustainable Development, Zurich University of Applied Sciences,
Technoparkstrasse 2, 8401 Winterthur, Switzerland, merla.kubli@zhaw.ch, +41 58 934 72 59.
is willing to give up on. For the model development this translates into the calculation of the
adoption fraction as a combination of some positive and negative factors.
A parameter named “Privacy” depicts the user willingness to adopt the service even though, it
requires a lower level of data privacy. Then, we argue that security services will be more attractive to
users in location where the police force is quick and effective (Enforcing effectiveness), and the
crime rate is high. Moreover, we introduce a parameter, Building characteristics, that comprises the
factors that might induce the user to feel less safe, such as the distance from one’s workplace or the
economic value, and the age of the building.
Building characteristics =
Distance from workplace + Economic value — Age of the building)/3
Equation 6
The formula for monitoring adoption fraction is shown in Equation 7 below.
i (Monitoring) =
Enforcing ef fectiveness + Crime rate of the area + Building characteristics — Privacy
Equation 7
The monitoring adoption from WoM is shown in Figure 5.
Monitoring
Adoption from
from Word of Mouth
AA ~
A en
i ———— Monitoring <+———_
adoption ,
ue fraction es Privacy
Crime rate | \
of th earea Effectiveness
Contact rate v
Building
Characteristics
f \
Age of the Economic
buiding Distance from value
workplace
Figure 5 - Monitoring adoption from Word of Mouth
Finally, Assisted living provides ambient assisted living for fragile people through real time
information about their health parameters. For this reason, the service leverages on two parameters.
First, we argue that there is a need to install more sensors within the domestic environment and also
use the wearable devices in order to collect health parameters. As a consequence, some users could
find this service too much invasive for the person living in the house.
Second, a risk factor exists to depict the degree of potential threats to the state of health of the person
living in the house. This factor should include the propensity of the elderly people to incur in
domestic accidents or their declining health.
10
The formula for Assisted living adoption fraction is shown in Equation 8 below:
i (Assisted Living) = Risk factor — Invasiveness — Privacy Equation 8
The assisted living adoption fraction is shown in Figure 5.
Assisted Living Adoption
from Word of Mouth
Pi nas. Ass Living
Assisted Living _ -—~ ~——— adoption “—
potential market fraction
/ \ , a %
1
\ Contact rate
Invasiveness
\
Percentage of \
households with
elderly people
\
Risk Factor
Figure 5 - Assisted Living Adoption from Word of Mouth
4.3. Calibration of the model
The model has been calibrated with three types of parameters. The first type refers to the context
where the service has been launched. The second and third types of parameters instead, instead,
pertain to some features of the service and the users’ stand and preferences towards these
characteristics. For the purpose of this paper, we could calculate the values only for the context-
related parameters. The values are computed for the city of Turin, Italy. Turin has the population
consisted of 9027137 inhabitants, is an industrial city located in the north-west of Italy, with an
annually per capita income of roughly 23,000 €.
Type of parameter Parameter Influence on the adoption
process
Context-related Enforcing effectiveness Positive
Crime rate of the area Positive
Risk factor Positive
Service-related Kwh saved Positive
Price Heating Negative
Advertising effectiveness Positive
Incentives from insurance company | Positive
141,
User-related Comfort Positive
Privacy Negative
Age of the building Negative
Economic value Positive
Distance from workplace Positive
Invasiveness Negative
Contact rate Positive
We retrieved the value for the context-related parameters from multiple sources. Crime rate of the
area is computed as the percentage of households victim of theft (elaboration from Urbes, 2015). In
Turin, 5,742,burglary were committed in 2014 (636.5 house robbery per 100,000 inhabitants).
Stating that 415,414 households live in Turin, the parameter Crime rate is equal to:
Crime rate = 5,742/415,414 = 0.014
Enforcing effectiveness is calculated as the percentage of burglars found guilty of house robberies.
This rate is equal to 2.6% for the region of Turin (we could not retrieve more detailed data on the
city level).
Finally, risk factor is computed as the frequency of falls of elderly people, as a proxy of how risky it
is to leave an elderly person alone in the house. This parameter is equal to 10.25% for the region of
Turin (Epicentro, 2013).
5. Implications and practical use of the model
The proposed System Dynamics model could serve a handful of purposes.
Through system dynamics methodology it was possible to frame the diffusion problem, and identify
what are the most important external factors that influence the diffusion of an innovative service.
Regarding this point, the model supports the identification of the most promising markets as a
consequence of these external factors and their influence on the diffusion process. For instance,
smart home services can be more popular in cities where the crime rate or the enforcing effectiveness
represents higher values.
Then, the model provides a proper tool for monitoring the diffusion process. Firms can perform ex-
ante simulation and create a benchmark and then assess the actual performances on the predicted
ones, so to validate hypotheses on the user's behavior. As a matter of fact, the diffusion curve
returned by the model can be fitted against actual data, by modifying accordingly all the user-related
and service-related parameters. This fine-tuning process can be performed periodically for more and
more accurate predictions. However, for this particular implication of the model, we have to assume
that the relation between the context-related parameters and the adoption fraction is linear.
This model can show the dynamics of the diffusion by end-users among different services, showing
that the diffusion of one service over the others can influence the diffusion of another connected
service. Then, this model can be used to evaluate how the process of cross-selling might develop,
12
and therefore from a company’s perspective insight can be drawn over which service should be
introduced or offered first.
6. Discussions and conclusions
During this research work we have identified some enabling factors for the diffusion of three smart
home services, further assessing the relation, positive or negative, between these factors and the
diffusion. We have divided these factors in three categories, and provided numerical calculations for
one category of factors, namely the context-related.
We applied the Bass diffusion model for the awareness phase of a diffusion process, showing that the
proposed model is able to provide insights on the most promising markets. Moreover, it can be used
to monitor the diffusion process, as well as test hypotheses on users’ requirements and behaviors,
especially in terms of interconnections between the services. However, user and service related
parameters can be assessed beforehand, through a sound market analysis on the users’ requirements,
in order to build a more predicting model.
Further development of this project will be the validation of the relation between context-related
parameters and adoption fraction; with real data inputs from end users and field studies. Context-
driven data can be changed according to a particular city with its peculiar characteristics, in order to
understand the dynamics of the innovation diffusion within the city under examination. In this way
the model will be adapted accordingly. Other factors will be also implemented in the model such as
the user’s dropout rate. A model focusing on the adoption step of the diffusion process is out of
scope of this preliminary work and will be tested in future developments, since it requires further
information on the price of the service and the risk-taking attitude of the potential users.
Acknowledgement
This work was done in collaboration with the Joint Open Lab SWARM and was supported by a
fellowship from TIM. We thank our colleague, Elisabetta Listorti from Politecnico di Torino, who
provided insight and expertise that assisted the research work.
References
Aldrich, F. K. (2003). Smart Homes: Past, present and future, in Inside the smart home, ed. Harper
(ed.) (London: SpringerVerlag), 17-39.
Allmendinger, G. and Lombreglia, R. (2005). Four Strategies for the Age of Smart Services.
Harvard Business Review, 83(10), 131— 145.
Ashton, K., (2009). That ‘Internet of Things’ Thing, in the real world things matter more than ideas.
RFID J.
Balta-Ozkan, N., Boteler, B. and Amerighi, O. (2014). European smart home market development:
Public views on technical and economic aspects across the United Kingdom, Germany and Italy.
Energy Research and Social Science, 3(C), 65-77.
Balta-Ozkan, N., Davidson, R., Bicket, M. and Whitmarsh, L. (2013). The development of smart
homes market in the UK. Energy, 60, 361-372.
Bass, F.M. (1969). A new product growth model for consumer durables. Management Science, 15,
215-227.
13
Chan, M., Campo, E., Estéve, D. and Fourniols, JY. (2009). Smart homes — Current features and
future perspectives. Maturitas, 64(2), 90-97.
Courtney, K.L. (2008). Privacy and Senior Willingness to Adopt Smart Home Information
Technology in Residential Care Facilities. Methods Inf Med., 47(1), 76-81.
Cusumano, M.A. (2010). Staying Power: Six Enduring Principles for Managing Strategy and
Innovation in an Uncertain World, Oxford University Press, USA.
Demiris, G. and Hensel, B.K. (2008). Technologies for an Aging Society: A systematic review of
“smart home” applications. Yearb. Med. Inform., 3, 33-40.
Ehrenhard M., Kijl B. and Nieuwenhuis, L. (2014). Market adoption barriers of multi-stakeholder
technology: Smart homes for the aging population. Technological Forecasting and Social
Change, 89, 306-315.
Epicentro-Il portale di epidemiologia per la sanita pubblica (2013). Rapporto nazionale Passi
d'Argento 2012. Available at: http://www.epicentro.iss.it/passi-argento/rapporto2013/cadute.asp
Accessed on 10 September 2015.
Ericsson ConsumerLab Insight Summary Report: Connected Homes (2015). Available at:
http://www.ericsson.com/res/docs/2015/consumerlab/ericsson-consumerlab-connected-homes.pdf
Accessed on January 20, 2016.
Fjord Report The Era of Living Services (2015). Available at:
https://livingservices.fjordnet.com/media-files/2015/09/living-services.pdf Accessed on January
20, 2016
Harper, R. (2003). Inside the smart home, Springer, Heidelberg, 63—78.
Holroyd, P., Watten, Ph., Newbury, P. (2010). Why Is My Home Not Smart? In Aging Friendly
Technology for Health and Independence, Volume 6159 of the series Lecture Notes in Computer
Science, 53-59.
Kalish, S. (1985). A new product adoption model with price, advertising, and uncertainty.
Management science, 31(12), 1569-1585.
Kreng, V.B., Wang, B.J., (2013). An innovation diffusion of successive generations by system
dynamics - An empirical study of Nike Golf Company. Technological Forecasting and Social
Change, 80(1), 77-87.
Lo, S.-H. Fan, C.-Y. Fan, P.-S., Lai, M.-F. (2011). Innovation product multi-generation diffusion
process, A system dynamics model approach. /CIC Express Letters, 5(4), 1235-1241.
Maier, F. H. (1998). New product diffusion models in innovation management—A system dynamics
perspective. System Dynamics Review, 14(4), pp. 285-308.
Milling, P. M. (2002). Understanding and managing innovation processes. System Dynamics Review,
18(1), 73-86.
Peres, R., Muller, E. and Mahajan, V. (2010). Innovation diffusion and new product growth models:
A critical review and research directions, Journal of Research in Marketing, 27, 91-106.
Perez-Lombard L., Ortiz J. and Pout, C. (2008). A review on building energy consumption
information, Energy and Buildings, 40(3), 394-396.
Pragnell, M., Spence, L. and Moore, R. (2000). The Market Potential For Smart Homes. York
Publishing Services, Published for the Joseph Rowntree Foundation.
Ren, X., Yan, D., Hong, T. (2015). Data Mining of Space Heating System Performance in
Affordable Housing. Building and Environment, 89, 1-13.
14
Ricquebourg, V., Menga, D., Durand, D., Marhic, B., Delahoche, L. and Loge, C. (2006). The Smart
Home Concept: our immediate future. Proceedings of 1% IEEE International Conference on
ELearning in Industrial Electronics, 23-28.
Rogers, E.M. (2003). Diffusion of innovations, 5th Ed. Free Press: New York.
Ryan S. P., Tucker C., (2011), Heterogeneity and the dynamics of technology adoption, Springer
Science+Business Media, LLC
Saylor, M. J., Slavin A. and Hugues Martin J.P. (2003). System and method for connecting security
systems to a wireless device. U.S. 6661340 B1, December 9, 2003.
Shen F., Ma T., (2013), System dynamics modeling of diffusion of alternative fuel vehicles, in
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence
and Lecture Notes in Bioinformatics). Volume 8041 LNAI, 2013, Pages 241-251
Timma L., Bariss U., Blumberga A., Blumberga D., (2015), Outlining innovation diffusion processes
in households using system dynamics. Case study: energy efficiency lighting. Energy Procedia 75,
2859 — 2864.
Tsai, J. M., and Hung, S. W. (2014). A novel model of technology diffusion: System dynamics
perspective for cloud computing. Journal of Engineering and Technology Management, 33, pp.
47-62.
Tumino, A. (2015). Internet of Things: l'Innovazione che crea Valore. Working paper for
Osservatorio Internet of Things, School of Management of Politecnico di Milano. Available at:
http://www. osservatori.net/ accessed on 10 October 2015.
URBES (2015). Il benessere equo e_ sostenibile nelle Citta. Available at:
http://www. istat.it/urbes2015 accessed on 10 September 2015.
Weiser M. (1991). The Computer for the Twenty-First Century. Scientific American, 265(3), 94-104.
15
