ArticlePDF Available

What is the Internet of Things? An Economic Perspective

Authors:

Abstract and Figures

This paper tryes to give answers to the question "What is the Internet of Things?" and the way how the author got to these answers
Content may be subject to copyright.
What is the Internet of
Things?
An Economic Perspective
Elgar Fleisch (ETH Zurich / University of St. Gallen)
Auto-ID Labs White Paper WP-BIZAPP-053
January 2010
Elgar Fleisch
Professor of Information and Technology
Management
ETH Zurich / University of St. Gallen
Contact:
Email: efleisch@ethz.ch
Website: www.autoidlabs.org
Business
Processes & Applications
What is the Internet of Things -
An Economic Perspective
I was always skeptical about the buzzword Web 2.0 - at least, it seemed like nothing more
than a buzzword to me, until I read the paper from O`Reilly titled What is Web 2.0? (O'Reilly,
2005). Until then, I thought of Web 2.0 as a collection of a few fuzzy concepts some people
gave a new name, just to plant a new tree in an already crowded Internet garden in order to
attract unjustified attention. However, after reading the paper and understanding the
concepts of user participation and service orientation more deeply, I became a convert. To
me, the term Web 2.0 now provides a natural and important bracket around the design
patterns and business models of the next level in Internet technology and usage.
Many people may share the same feelings I once had for Web 2.0 for the term Internet of
Things (IOT). Is it just another skin around well-known concepts such as ubiquitous
computing, pervasive computing, cyber physical systems, ambient intelligence, or
technologies such as sensor networks and RFID? If not, what value does it add? And frankly,
what is it really? What are its main building blocks?
In recent years, the usage of the IOT-idiom has grown considerably. It has become a leading
theme in conferences, books, academic and professional journals, university courses,
research summer schools, research programs of companies, universities, applied research
organizations and government-funded research programs, as well as reports on global future
developments and industry analysis. However, the relevance of the term IOT is still not
comparable to, for instance, Web 2.0 when measured in usage (e.g., in Google searches and
hits) or in global spread, which is still somewhat European centered.
With this paper, I first and foremost want to render an account of what I think the IOT is, what
its constituting concepts are and which main impacts on society and economy we can see
today and expect in the near future. Doing so, I want to provide my research team, students
and, perhaps colleagues in academia and industry with a baseline and some directions for
ongoing and future research and development endeavors. Thus, this paper is targeted
towards students, practitioners and researchers who are interested in understanding and
contributing to the ongoing merge of the physical world of things and the Internet.
The conclusions of this work are based on information compiled from three sources: on vast
numbers of academic and industrial publications, on numerous interviews and talks with
colleagues, and on the personal experience I was lucky enough to gain within the last eight
years.
The paper is structured as follows: The first section identifies the differences between
Internet applications and applications that would probably belong to a cloudy IOT. It is written
to sharpen the understanding of what the unique characteristics of an IOT would be. The
next section searches for patterns in the investigated IOT-applications. It does so by
identifying the root causes that drive the value of IOT-application to users and companies.
The resulting common theme, the reduction of the real world-virtual world transaction costs,
provides the base line for the overall economic energy of the IOT, which is discussed in the
subsequent section. The paper closes with a description of some less obvious and therefore
juicy patterns of how companies approach the IOT, followed by a brief summary and outlook.
What is the IOT and how does it differ
from the Internet?
The basic idea of the IOT is that virtually every physical thing in this world can also become a
computer that is connected to the Internet (ITU, 2005). To be more accurate, things do not
turn into computers, but they can feature tiny computers. When they do so, they are often
called smart things, because they can act smarter than things that have not been tagged.
Of course, one could question whether things would really have to feature computers to
become smart. For instance, a consumer good could be considered to be already smart,
when tagged with a visual code such as a bar code or equipped with a time-temperature-
indicator that, say, a mobile phone can use to derive and communicate the product's state of
quality, dynamic carbon footprint, effect on diabetics, or origin. Certainly the boundary
between smart things, which autonomously can derive and transform to different states and
communicate these states seamlessly with their surroundings, and not so smart things,
which only have a single status and are not very active in communicating it, is blurring
(Meyer et al., 2009). For pragmatic reasons, however, I will focus in this paper on smart
things that are smart because they feature tiny low-end computers.
The IOT-idea is not new
1
. However, it only recently became relevant to the practical world,
mainly because of the progress made in hardware development in the last decade. The
decline of size, cost and energy consumption, hardware dimensions that are closely linked to
each other, now allows the manufacturing of extremely small and inexpensive low-end
computers (Payne, MacDonald, 2004).
As mass adoption of these tiny networked computers becomes a real option, new questions
surface. What, if anything, would discern the IOT from existing computing realities, in
particular the Internet? What new values and risks would it generate? And what new
infrastructure would it have to rely on? The following paragraphs describe the most important
differences between the Internet and the IOT:
1. Invisible versus flashy hardware. First of all, the hardware in the IOT looks considerably
different and serves a different purpose. Whereas the nerve ends of the Internet are full-
blown computers, from high capacity work stations to mobile phones, that require regular
access to the power grid, the nerve ends in the IOT are very small, in many cases even
invisible, low-end and low energy consumption computers. They typically feature only a small
fraction of the functions of their bigger Internet-siblings, often including sensing, storing and
1
For early mentors of the IOT and similar concepts see (Gershenfeld, 1999), (Ferguson, 2002), (Kindberg et al.,
2002), (Schoenberger, Upbin, 2002), (Wright, Steventon, 2004). For an overview of the history of the IOT see
(Mattern, Flörkemeier, 2009).
communicating a limited amount of information. In most cases, they cannot interact directly
with human beings.
2. Trillions versus billions of network nodes. Today, about five billion devices such as mobile
phones (3.3 billion), personal computers (1.2 billion), MP3 players (220k), digital cameras
(120k), web cams (100k), PDAs (85k), and data servers (27k)
2
serve a population of about
6.7 billion people, of whom only 1.5 billion are currently using the Internet
3
. These numbers
seem huge, but in comparison to the number of things we constantly create on this earth,
they are not. One can grasp an idea of the order of magnitude by estimating the number of
consumer products that are produced every year. To do so, I divided the revenue of a
leading consumer goods company by the estimated average product price, expanded the
reported market share to 100%, and multiplied that result by the estimated lifespan. The
resulting number (84 billion) only begins to hint at dimensions, as many of these products
might never be equipped with minicomputers, and as consumer products only account for a
fraction of the things we create each year. Sanjay Sarma even estimates 555 billion units in
an Auto-ID Center-specific selection of supply chains (Sarma, 2001). However these figures
may exactly look like, these estimates already suggest that there will be so many computer-
enabled things around us that, firstly, people will not be willing and able to directly
communicate with them, and secondly, a new network infrastructure, the IOT, might be
required.
3. Last mile bottleneck versus highway. The last mile in a communications infrastructure
refers to the communication link between the nerve endings, or the leaves in a local tree, and
its next tier or layer. Driven by user demands, e.g., streaming videos, and technology
progress, the speed of the last mile in the Internet has been increasing tremendously over
recent years. Today, an average household in many countries can expect to have cable-
based Internet access with a bandwidth of at least 1 MBit/s. With the implementation of
emerging technologies such as fiber optics to the home, the bandwidth will soon become as
high as 50 - 100 MBit/s. By contrast, the speed of the last mile towards an average low
energy consuming radio frequency identification tag is only about 100 kBit/s.
4. Babylon versus global identification and addressing. The low end hardware of the IOT is
responsible for another difference: the identification and addressing of the nerve endings. In
most cases, the Internet-based identification and addressing schemes require too much
capacity to become part of low-end smart things. Therefore, academic and industrial
communities are searching for alternative technologies and standards (e.g. EPC, ucode,
IPv6, 6LoWPAN, Handle System, or Internet0) to number and address the smartening
physical world. So far, most identifiers for smart things and technologies bridge the last mile
based on local, technology vendor-specific closed-loop schemes. However, if the IOT would
like to follow the successful path of the classical Internet, its architecture would have to make
sure that any tagged object could in principle be accessed by any computer. For that, a
global standard protocol, identification and addressing scheme for bridging the last mile from
the Internet to the smart things would be required.
2
Data Source: Wired 7/2008
3
Data Source: http://www.internetworldstats.com/stats.htm
5. Machine-centric versus user-centric. The characteristics of the Internet and the IOT define
the range of services they support. The vast majority of Internet-based services are targeted
towards human beings as users, e.g., the World Wide Web (WWW), email, file sharing,
video, online chat, file transfer, telephony, shopping, or rating. The attributes of the IOT
almost completely exclude humans from direct intervention. That is why Marc Weiser called
for a paradigm shift towards human-out-of-the-loop-computing when he introduced the vision
of ubiquitous computing. In most IOT-applications, the smart things communicate amongst
each other and with computers in the Internet in a machine-to-machine way (Mattern, 2004).
When users need to be involved, e.g., for decision making, they currently contribute via
personal computers and mobile phones.
6. Focus on sensing versus on communication. The economic success story of the Internet
started with the WWW, which allowed companies and individuals virtually for the first time to
reach out to a global customer base at ridiculously low cost. Looking closely at the WWW’s
communication abilities (distribution and presentation of content), it seems no big surprise
that the first economic success stories were made in the areas of advertising (Google) and
shopping (e.g., eBay or Amazon). However, looking backwards has always been far simpler
than forecasting.
The second boost of the Internet was largely driven by adding the ability to deal with user-
generated content, i.e., data that is not only consumed by users but also provided by users.
The success stories of these Web 2.0-based services include Facebook, YouTube, Twitter,
and Wikipedia.
The IOT adds another data dimension. It allows the physical world, things and places, to
generate data automatically. In fact, in my view, the IOT is all about sensing the physical
world. It provides the infrastructure that for the first time enables us to measure the world,
just as Gauss did around 200 years ago, but far more powerful. It is a cost-efficient means of
growing a very finely granulated nerve system with trillions of new nerve endings. Linked
together, they can provide humans with a measurement tool that opens the door to many
new findings, applications, benefits, and risks.
Internet of Things, or Web of Things?
After reading through the differences, one can argue that the IOT is not on the same level as
the Internet, that it is in fact “only” an application of the Internet, in very much the same way
as many existing Internet-enabled services. Following that path, the term IOT would be an
exaggeration, and should be renamed something like Web of Things.
On the other hand, one can easily argue that a Web of Things would require low level
building blocks, e.g., for addressing the smart things, bridging the last mile and linking it with
the Internet, that are peers with some Internet building blocks. As a consequence, the IOT
may rightly be conceptualized as an extension of the Internet to reach out to the physical
world of things and places that only can feature low-end computers (Gershenfeld et al.,
2004).
In fact, often enough when we use the term IOT, we do not differentiate between the
infrastructure and application levels. We use IOT as a bracket-term to refer to one or both at
the same time.
On the infrastructure level, the IOT can be viewed as an extension to the Internet as we
know it today. The IOT expands the technical Internet building blocks such as DNS, TCP,
and IP with identification and addressing schemes, last mile communication technology, and
an Internet gateway that matches the IOT requirements, foremost among them low energy
consumption, low cost, and mobility (Sarma, 2004).
Identification and addressing scheme. In many applications, the identification and addressing
of an IOT-leaf via IP numbers or MAC identifiers requires too much computing power to be
handled by a minicomputer that can operate autonomously on a sustainable basis. Current
developments in the areas of 6LoWPAN (IPv6 over Low Power Wireless Area Networks)
(Hui, Culler, 2008), mini IPv6 standard stacks for sensor networks, energy-harvesting,
energy-storage, and energy-consumption are likely to change this equation, but for the time
being almost all up-and-running IOT-solutions use alternative ways to identify and address
their sensors.
Last mile communication technology. Furthermore, the communication technology that
bridges the air from a sensor to a regular node in the Internet has to bear up to the typical
restrictions of a last mile in the IOT. It has to be wireless, robust, and, most of all, energy
efficient. In some cases, the communication protocol must also enable security features,
transport energy to run the sensor, or allow measurement of the distance (ranging) and
localization. The proposed methods and standards for corresponding communication
protocols are as manifold as the IOT application areas.
Gateway to Internet. And last but not least, once the identifier of a tag, along with other
sensor information, has been successfully communicated to a node in the Internet -- a node
that can operate based on Internet technology because, amongst other considerations, it
regularly has access to the power grid -- it frequently has to be resolved against other
resources in the Internet. For instance, in a very simple application, the gateway only has to
find the digital proxy of a tagged thing. For this task, the identifier of the tag needs to look for
a corresponding IP address, sometimes referred to as the “homepage” of the tagged thing, in
much the same way as the Domain Name System (DNS) resolves a domain name into a
corresponding IP address. A gateway based on the DNS, called the Object Name System
(ONS), was proposed by the Auto-ID Labs.
In more complex and also more realistic scenarios, there will be more than a single
“homepage” attached to a thing or place, and they will not really be homepages but web
services. In an ideal open IOT-architecture, not only can every sensor be reached by every
authorized computer or person, but in addition, every person and organization can set up
their own services, link them with identifiers, and offer them to the public. For instance, a tag
on a consumer good would not just provide a link to the product homepage provided by the
producer (in the world of EPCglobal, which would be a pure ONS-based service). Rather, if
brought close enough to an RFID reader, it would generate an additional list of alternative
services provided by independent firms or not-for-profit organizations from which the user or
the user's computer system can choose. This list could include services such as product
rating, fair trade check, counterfeit check, proof-of-origin, replenishment alert, political
shopping (do I, by buying the product, support labor in a foreign country or in my beloved
home country?), or self check-out. The alternative services do not necessarily have to be in
alignment with the interests of the consumer goods company. In the world of EPCglobal, the
Discovery Service would generate the list of services that are available given an EPC.
On the application level, it can be observed that IOT-applications never work stand-alone,
but always also use Internet-based services. So IOT-applications might simply be regarded
as a special set of Internet applications that also leverage the IOT-infrastructure. They
recently have been subsumed under the term “Web of Things,” in retail environments “Web
of Goods,” or in closed-loop scenarios even “Intranet of Goods.” In the “Web of Things,”
tagged items or spaces serve as additional triggers and actuators to re-invent classical web
applications such as product rating, or to enable new services such as pet-tracking. As on
the infrastructure layer, they extend classical Internet applications to the real world.
IOT-standards: One global one-size-fits-all?
Unlike in the Internet, there is currently no single global set of standards for the IOT, and in
all likelihood, there never will be. The most important reason is that the IOT leaves the clean,
closed, logically consistent and self-sufficient digital world. IOT projects suddenly have to
deal with physical properties such as distance (should a tag ideally be readable within a few
hundred meters, a few meters, or a few millimeters?) and characteristics of neighborhoods
(e.g., materials that absorb or reflect radio waves).
4
. These properties depend on concrete
applications, and these applications are almost as manifold as the physical world itself,
resulting in a rich variety of technological forms of appearances.
In some industries, however, de facto standards emerge. For instance, mainly due to the
mandates of Walmart, Metro, and other large retailers, the EPCglobal standard stack is the
de facto standard in the retail and consumer goods industries (Thiesse et al., 2009). And
since retailers do not only sell consumer goods, the EPC standard likely expands to other
related industries, such as the textile or the pharmaceutical industry. Once a standard drives
large quantities, and the EPCglobal standard stack certainly does that, the cost of standard-
compliant technology declines dramatically (after all, size matters) and is likely to draw
additional industries to join in, which will be further propelled by the availability of open
source implementations of the EPC stack (Flörkemeier et al., 2007).
4
In classical IT projects “only” two general types of skills have to cooperate to create a working solution: the
people who understand the business or user side of an application, and the IT guys. In an IOT-project, at least
for the time being, an additional party who can deal with the physical challenges, typically electrical
engineering technicians, is needed (Henzinger, Sifakis, 2007).
When things add value - IOT value drivers
Another approach to isolating the essence of the IOT is to look at the value that its
applications add for both businesses and consumers
5
. I learned quickly that trying to
structure IOT applications is as impossible as modeling the entire world, because essentially
every business process in essentially every industry on this globe is embedded in the
physical world. Thus, the IOT is potentially relevant for every step in every value chain. So I
switched to searching for the origins of the value an IOT application would provide to its
milieu. To do so, I took a list of about a hundred existing and emerging applications that
leverage the IOT concept. It turned out that every investigated application sports one or more
of the seven main value drivers identified below. The first four drivers are dedicated to root
causes based on machine-to-machine communication, while the latter three show root
causes based on the integration of users.
1. Simplified manual proximity trigger. The first driver in the proposed value driver stack is
very basic and is part of numerous applications such as self check-out and stock-taking in
libraries, access control in buildings and sporting facilities, basic payment procedures, even
pet tagging. Its business value stems from the fact that some smart things can communicate
their name, i.e., their unique identification number, in a very robust, fast and convenient way
when they are manually (and usually consciously) moved into the roaming space of a
proximity sensor such as an antenna or a camera that sits and waits for something to pass
by. As soon as the smart thing is close enough to the hot spot, a transaction, e.g., a payment
procedure, a validity check or the creation of an entry record, is automatically triggered.
Businesses include this value driver in their applications because it makes the life of their
employees more convenient (e.g., moving an RFID-loaded access card across a hot spot is
far more convenient than entering a six digit personal security number), enables customer
self service (i.e., outsourcing of costly tasks such as check out to customers) and as a
consequence reduces labor costs. Consumers value this driver for some of the same
reasons. It helps them to save time, to gain independence via self-serving, and finally to
increase their perceived convenience.
2. Automatic proximity trigger. This value driver adds a single but important feature to the
previous one: it triggers a transaction automatically when the physical distance of two things,
let us say a pair of Levi's jeans and a gate in a department store, drops below a threshold,
e.g., when a consumer steps out of a store with a purchase he forgot to pay for. Many
business applications in production and the supply chain management sport this value driver,
from asset management to inventory management. Whenever a smart thing such as a
tagged truck, forklift, pallet, carton, work-in-progress bin, or consumer product does not
remain at a distance from some other smart thing or place that can sense it, a transaction
5
In the European Union it recently became common to replace the term “consumer” with “citizen”, possibly to
stress the fact that the concept of men is richer than that of consumer, and that new technology development
should address human beings as a whole. In this paper, I deliberately use all such terms synonymously to
convey that I never saw humans as pure shopping mammals.
such as an update of a bookkeeping record, the initiation of a replenishment task, or the ring
of an alarm bell is triggered. In other words, IOT applications using this value driver leverage
the powerful qualities of physical neighborhoods to build new and better business processes.
In the pure digital world of classical supply chain management systems, production planning
systems, or enterprise resource planning systems, this was, of course, simply impossible
(Bullinger, Ten Hompel, 2008), (Vitzthum, Konsynski, 2008).
The implementation of this value driver leads to an increase in speed, accuracy, and
convenience that allows companies to reduce their labor costs, process failure costs and
costs of fraud
6
. In addition, it delivers massive new data that can be used to improve
processes constantly over the time. Consumers may directly profit from physical self triggers
via an additional level of convenience, for instance, when a new BMW car opens its doors on
its own as the bearer of the car key approaches. Further, several manufacturers in the
automotive, aircraft and computer assembly sector are developing systems to link the
informatory with the physical world on the shop floor by means of augmented reality
applications (Ong et al., 2008), (Regenbrecht et al., 2005). Proximity triggers are applied to
support workers with work instructions, assembly plans and other information they just
require to fulfill their current task. Assembly steps can even be documented automatically,
which may eliminate almost any manual information processing on the shop floor.
3. Automatic sensor triggering. Value drivers one and two create benefits by manually and
automatically sensing and communicating the name of a thing. Value driver three expands
the ID by any data a smart thing could collect via any sensor. Examples for sensor data
include temperature, acceleration, localization, orientation, vibration, brightness, humidity,
noise, smell, vision, chemical composition, and life signals. This driver allows a smart thing to
constantly sense its condition and environment for relevant movements and initiate actions
based on preprogrammed rules. For instance, it would allow a smart olive tree to constantly
check temperature, brightness, and humidity (of soil and air) to adjust the optimal water feed.
Automatic sensors enable local (therefore individual) and prompt (therefore event-based)
decision making. They rapidly increase the quality of processes, which results in more
efficient (better input/output relation) and more effective (better output) ways of doing things.
In the case of the olive farm, so-called Precision Agriculture would translate into better, or at
least bigger, olives since the watering over time would be closer to a theoretical optimum
(Wark et al., 2007). It would probably also lead to a more environmentally friendly usage of
water, since the tight process control would eliminate unnecessary irrigation.
The fields of application are manifold. They reach from condition monitoring throughout the
entire supply chain to networked smoke detectors in private homes, from the management of
perishable goods to the production of sweet wine, from the monitoring of manmade
construction to early-warning systems for forest fires or earthquakes, from smart meters to
increase the efficiency of the electric grid to the monitoring of life signals of patients in
hospitals and at home. This value driver represents the options that develop when
computers, the IOT, can measure the world in detail at reasonable cost. Then, the IOT
serves as a network of sensors for far more senses than those of human beings. And it can
do so continually, at a ridiculously high resolution, and across the globe.
6
See (Lee, Oezer, 2007) and (Sellitto et al., 2007) for a review of RFID-related value drivers
4. Automatic product security. Another value driver that is part of many applications such as
proof-of-origin, anti-counterfeiting, product pedigree, and access control is product-related
security. The thing to be secured can bear a minicomputer that is equipped with some
security technology such as cryptography. The space or user confronted with such a smart
thing can check the validity of it by walking through the implemented method, for example, a
challenge-response operation. These methods are well established and well understood. For
instance, they are building blocks of every ATM card or car key. However, they require
expensive and power-intensive computing resources. In addition, they often demand costly
handling of digital keys. That is why this method is limited to applications where high values
and risks are at stake.
For inexpensive mass-produced products, another method surfaces: smart things can
provide some level of derived security based on the interplay between a smart thing and its
digital proxy. Imagine that every smart thing has its own homepage (the digital proxy) that is
constantly updated whenever a physical artifact has triggered some action as described
above. This homepage, which looks very much like a curriculum vitae or a pedigree, can be
used to derive with some level of confidence whether the thing under investigation is the
rightful owner of the homepage or not. For instance, if two products point to the same
homepage, one must be a fake (Staake et al., 2008).
In both cases, computers can check the validity of a product automatically, without human
intervention. Whereas the first method works with costly high-end security features built into
the hardware of the nerve endings of the IOT, the second method approaches the security
problem by leveraging the network, i.e., it constantly collects and updates data from the IOT
and then, upon request, uses software to calculate the likelihood of a product being
counterfeit. The network-based method is fuzzier than the hardware-based one, but it is so
inexpensive that it can be applied to every good, and that checks can be carried out on a
constant basis. This, in the end, leads to a new level of security. If every truck, shelf, sales
rep, and consumer checks every drug (because it is simple), the business of counterfeit
producers breaks. With enough eyeballs, all fakes are shallow.
7
5. Simple and direct user feedback. Although the IOT nerve endings are usually very small,
usually even invisible, sometimes smart things feature simple (which translates in this
context to small and energy-efficient) mechanisms to give feedback to the humans who
interact with them at the point and time of action. Often they give feedback to reassure, for
instance, an employee that the manual or automatic proximity trigger actually worked. They
do so by producing an audio signal such as a beep (e.g., when a pallet was properly
identified by a gate), or a visual signal such as a flashing LED (e.g., when a virtual Kanban-
card was sent and received wirelessly). In more entertaining consumer-oriented applications,
the feedback may even produce funny sounds (look for Friedemann Mattern’s Knight’s castle
in (Lampe, Hinske, 2007)), haptic effects such as those we know from playing Wii, or even
smell. In applications that deal with perishable goods, a simple automatic sensor trigger can
show its finding on tiny traffic signal-like displays that tell a consumer whether the product is
7
Eric Raymond once said, on the quality of open source software, that “with enough eyeballs, all bugs are
shallow.” His statement is often cited when the quality of user-generated content, e.g., in Wikipedia, is
discussed.
still worth its not so dynamic price. Advanced car keys can sense where their car is and
indicate the direction to the driver. In production environments, for instance at the production
facilities of Infineon and ST Microelectronics, smart assets even feature a low-energy-
consumption display that tells the operator, amongst other useful information, the next
destination, a machine tool or a shelf, they are to be brought to. This feature, in combination
with identification, localization, and connection to the production execution system, allows a
new level of nearly error-free production of logic chips that is also flexible and cost efficient.
6. Extensive user feedback. This value driver extends the output from simple user feedback
to rich services. To cope with the limitations of the last mile of the IOT, a user friendly
computer, most often a mobile phone, has to serve as a gateway that links the smart thing
with its homepage or any other resource in the Internet that is relevant to the user and the
thing in context. Applications that leverage this value driver are manifold. One could easily
imagine a service that augments product information on physical products, such as a bottle
of wine, with additional information, for instance, from the producer, the dealer, the Wine
Spectator, the Johnson wine guide, or consumer forums (van der Heijden, 2006), (Keegan et
al., 2008). Other consumer product-related service ideas, some already implemented,
include on-the-spot price comparisons (should I buy the product here in this retail store,
when I can get it for a two-dollar discount three blocks down the road or a four-dollar
discount at Amazon?), political shopping advice (which country’s labor produced this
product?), allergy and health warnings (will this product harm me if I have, say, an hazelnut
allergy, or if I suffer from a particular type of diabetes?), or product rating (How did my friends
like this wine? Would they be happy to drink it this evening?) (von Reischach et al., 2009).
The augmentation application is also being used to create new tourist services in cities and
museums, where artwork and points of interest are tagged to link mobile phones that pass by
with audio and video streams that explain the foreign world in the language and at the
individual level of detail and expertise selected by the interested connoisseur. It also is
helpful for linking products such as coffee machines and machine tools that are already in
operation with their operation and repair manuals or individual maintenance records.
In all applications, the mobile phone is the primary means for providing the window to thing-
specific content and services that run on the web. For this and other reasons (simplicity,
mobility, computing power, sensor richness, security level, network infrastructure, adoption
rate, etc.), many, including myself, believe that the mobile phone is bound to be the mass
computer of the future. The recent investments of high tech companies that had historically
been focusing on the personal computer market, e.g., Apple, Google, and Microsoft, in the
mobile phones market may provide additional evidence for this trend.
Businesses profit from this value driver by establishing a new channel to maintain contact
with consumers, offer new services, and gain consumer attention (Allmendinger, Lombreglia,
2005). Services such as product rating and price comparison are nothing but packaged
advertisement. This explains why Google has to be bold regarding Android, its mobile phone
computing platform. Consumers profit from the fact that they can access personalized (the
mobile phone likely knows who they are) services in a very simple and fast way (no need to
start up the computer, the browser, search for a producer's homepage, drill down to the
information they are looking for) right at the place (on the spot, e.g., in a CVS store) and time
(now!) they need to act.
7. Mind-changing feedback. This value driver is at odds with the drivers explained above and
is not based on new technical features of the IOT. Its motivation stems from the concern that
the combination of real world and virtual world computing might generate a new level of
manipulating people. Most humans still spend most of their lives in a physical world. In light
of the how many people spend their leisure time in front of TV sets, gaming consoles, or the
Internet trying to kill a monster or find a second life, one could doubt that. However, I believe
that the physical world is still by far the greatest location for joie de vivre. Physical
experiences such as touching well-designed objects (and humans) or staying in bleak
buildings does something to us emotionally. Now, as computing becomes physical, e.g.,
when computers and the Internet grow a physical nerve endings, some of that power is
accessible in IOT applications and will hopefully be used for good.
Examples of such applications include a toothbrush that interacts with a comic figure on the
bathroom mirror to motivate children and grownups to seriously take care of their teeth, or
smart meter-based applications that show a consumer how much electric power and water
he is consuming, how much peers consume, and what he could do about his resource
consumption not only to increase the cash in his wallet, but also to satisfy his ecological
conscience. Companies, from the utility and insurance industries can use this driver to
design new products and services that align their business goals with consumers who want
to improve their lives and act more responsibly.
In another insurance application, a consumer gets a discount on his car insurance when he
equips his car with a crash recorder that acts like a flight recorder in an airplane. In case of
an accident, the crash recorder can help the insurance company reconstruct the exact
course of events. This fact generates two benefits on the insurance company's side: First, it
helps insurance companies to attract risk-averse customers, who usually generate above-
average margins. So far in their self-perception unjustifiably unrecognized race drivers
certainly would not sign a contract like that. Secondly, it would help to maintain the risk-
averse driving style because incorruptible, honest crash recorder data would not be able to
spin their content in either direction
8
.
All these applications leverage effects that are currently being studied by a discipline called
behavioral economics, which operates on the premise that humans act in every way but
rationally. However, as Dan Ariely shows, they act predictably irrationally (Ariely, 2008).
8
For further information on usage-based insurance models see (Filipova, Welzel, 2005), (Coroama, 2006).
Value driver Value root Business
value
Consumer
value
Example
applications
1. Simplified
manual proximity
trigger
Self-talking ID eases
the triggering of a
transaction and leads to
increase in transaction
speed, accuracy &
convenience
Increased job
satisfaction,
enables consumer
self-service &
reduces labor
costs; increases
data accuracy
Increase in self-
service, speed,
and
convenience
Self check out in
libraries, stock-taking
in libraries,
Access control in
buildings, sporting
facilities and such,
Pet tagging
2. Automatic
proximity trigger
Self-talking ID
automatically triggers a
transaction when in a
roaming area; Leads to
additional speed,
accuracy &
convenience
Reduced fraud-
related costs,
process failure
costs, and labor
costs; New high-
granularity data for
process
improvement
Increase in
convenience
Asset tracking;
Robot stock-taking in
libraries
Theft prevention in
stock-taking;
Car keys
3. Automatic
sensors trigger
Smart thing monitors its
local surroundings,
applies sensor data to
process rules and self-
triggers actions if
required; Enables
event-based actions
based on local data;
increases process
quality
Individual and
prompt process
control increases
process efficiency
and effectiveness;
Additional level of
data granularity for
further process
improvement
Leap in quality
of products and
services
Olive oil production,
condition monitoring,
networked smoke
detectors,
management of
perishable goods,
compliance
monitoring, smart
meters
4. Automatic
product security
Built-in cryptography
and interplay between
physical things and
their digital
representations
enables new level of
security of things
Reduction of cost
of process failure
due to fraud;
reduction of
process security
cost
Increase in
consumer trust
New trust-
related services
Anti-counterfeiting,
proof of origin,
pedigree, access
control
5. Simple direct
user feedback
Smart things provide
direct feedback to users
to increase confidence
and local process
control
Processes become
more accurate,
more flexible, and
faster
Increase in
convenience
and
entertainment
value
Production lot that
shows next job,
perishable good that
tells its quality
status, feedback
gate, digital
enhanced games,
direction indicating
car keys
6. Extensive user Real-world object New customer Increases Deep product
feedback
serves as a link to a
broad range of services
relevant to user and
object; User handles
services, typically via
mobile phone
contact, new
advertisement
opportunity,
additional service
revenues
convenience
because
individualized
information is at
hand exactly at
the point of
decision
information, price
comparison, political
shopping, allergy
test, product rating,
audio tagging, city
and museum guide,
mobile operation and
repair manual,
maintenance record
7. Mind changing
feedback
Technology that is
targeted to influence
the behavior of users,
hopefully for the good
Enables new
emotional product
features and new
services; enables
active selection of
attractive customer
segments; helps to
align business
goals with green
goals
Helps to
improve life and
act responsible
in many
different ways
Improve health, e.g.,
via smart toothbrush,
avoid risks, e.g., via
crash recorders or
pay-as-you-drive
models, save energy
via smart meter
apps, save water via
water metering
Table 1: Summary of IOT value drivers
The economic energy of the IOT
The value drivers are a result of a fundamental economical principle of the IOT: The IOT,
with its technologies to automate the bridging of the last mile between the Internet and the
physical world, dissolves the transaction costs that are caused by real world-virtual world
media breaks. A real world-virtual world media break occurs when a piece of information is
transferred from one carrier medium, e.g., a bar code, to another, e.g., a database that
serves a warehouse management system. When things become computers, these media
breaks, along with their attached costs, fade away (Figure 1).
Dissolving real world-virtual world transaction costs - The power of
avoiding media breaks
Making media breaks disappear might seem to be a small thing, but it is not. In fact, one
could characterize the entire computerization of business and society that has happened in
the last 60+ years as a consequence of the ongoing avoidance of media breaks. With every
new generation of information systems, a set of media breaks vanished. For example, with
the introduction of department-wide information systems, e.g., in accounting, all accounting-
relevant data only needed to be entered into an accounting system once, resulting in one
media break per digitalization. After that, the accounting system could use the data as often
as required without another media break. Before the introduction of accounting systems,
every calculation resulted in several media breaks because an accounting clerk had to
transfer information from a piece of paper into his electronic calculator and then the outcome
back to another, usually paper-based, medium.
With the introduction of company-wide enterprise resource planning systems (enabled,
among other things, by the Ethernet), again millions of company-internal media breaks
vanished. With cross-company information systems (enabled by the Internet) such as supply
chain management systems, another large set of media breaks faded away. The same held
true when content management systems enabled the integration of weakly structured
information such as text, presentations, and videos, or when Internet-enabled information
systems started to integrate not only business partners but also consumers.
Of course, there is a reason that dissolving media breaks are a constant in the history of
business computing: media breaks usually require humans to be resolved. While humans are
the reason for living in general, they are not very good at dealing with media breaks, which
results in error-prone, slow and costly procedures. Error-prone, because men are not built for
replicating simple, boring and tiring tasks, such as keying in data, thousand times a day. So
they are bound to make errors, which may sum up to an average master data accuracy of
about 70% (!) (DeHoratius, Raman, 2008). Slow, because our abilities to do parallel
processing are very limited. And costly, either because labor is in many countries for some
reason taxed more than capital, or because we are biologically overqualified for replicating
tasks simple machines could do for us.
Physical world („atoms“), humans, things, places
e.g. RFID
No human intervention required
Transaction costs
Punch card Key board Bar code
Manual interventon required
Digital world („bits“), information systems, Internet
Time, technology
Merging worlds (c) E. Fleisch
Figure 1: Merging Worlds
Adding the rebound effect: high-resolution data becomes economical
The IOT technologies are not the first attempt to reduce the cost of the last mile. In fact,
every data entry method, from punch card to keyboard to barcode, has pursued the goal of
reducing real world-virtual world transaction costs. However, when things and places also
sport minicomputers, the variable transaction costs
9
converge toward zero. And that
produces a rebound effect
10
: as the price of a sensing event declines, it becomes more
attractive to sense more often (Figure 2).
In different applications, one can study the movement from low-resolution sensing to high-
resolution sensing along three axes. First of all, when the transaction costs of real world
sensing are high, companies tend to sense only when it is inevitable to do so. For instance,
they check their inventory only once a year or when something unusual with high financial
9
Variable transaction costs as opposed to fixed transaction costs refer to the marginal cost of one reading
event, not including the one-time cost of hardware, software and such.
10
We probably all know the rebound effect shrinking telephone charges cause: as it becomes cheaper for us to
place a call, or surf on the Internet, we check our mail and make long distance calls more often. As a
consequence, the monthly phone bill is likely to increase instead of decrease.
Figure 2: Towards real world high resolution data
consequences happens. However, when sensing is free, they might want to sense all the
time. Why? Because it allows processes to react in real-time, and that is often a precondition
to achieving optimal process efficiency. Second, with low sensing costs, it becomes
economical to check real world status not only within the business's own four walls at a few
gates (e.g., asset tracking in a closed-loop bar code-based application), but throughout the
entire supply chain via an infrastructure that works everywhere (e.g., asset tracking by
leveraging GPS/GSM-technology). And finally, with sensing costs fading, companies start to
increase the richness of the data they sense, from simple automatic reading of identification
numbers of large things such as containers to any status of a single item or its surroundings
that new sensing technology can measure.
“Trusteddata
Machine sensing not only leads to a new level of data resolution, it also produces “trusted,”
or as Pentland calls it in a slightly different context, “honest” data. Trusted data is data that is
difficult to influence because it is quietly and continuously collected by machines all the time;
employees and users cannot deliberately choose the time and place of sensing events, as
these happen silently as business processes are executed, e.g., when a work-in-progress
asset on a shop floor is moved to the next machine tool, or when an express mail package is
delivered. The price of this silent monitoring is, of course, loss of privacy
11
. However, this loss
of informational self-determination (which is another definition of privacy) also has a
somewhat positive aspect: it generates data that is more trustworthy. For instance, a retailer
might trust sensor-collected delivery data of its logistic partner far more than questionnaire-
based data that only collects statements from truck drivers whether they have been on time
or not. In fact, the power of trusted data may even modify the business relationship between
the retailer and its logistic partner towards sensor data-based quality control, and in
consequence, remuneration.
The magic of measure and manage - generating the MRI for business
administration?
You can manage only what you can measure. This statement is credited to Peter Drucker,
one of the most influential management thinkers to date. It states the fact that measuring the
effects of a system is a condition of being able to understand and improve it. It serves as a
basic principle in the management of any type of man-made organization. For instance, it is
impossible to effectively guide a legion of firemen to fight a forest fire without knowing where
the teams and the fire sources are, or to steer an army of sales representatives without
methods to measure their achievements.
11
See (Langheinrich et al., 2005), (Thiesse, 2007) for privacy issues in the context of the IOT.
Low ResolutionHigh Resolution
The truth of Drucker's statement is not limited to the management domain. It also holds true
for disciplines such as physics and medicine, and it explains why many Nobel prizes have
been awarded to people who invented new breakthrough measuring instruments: with new
means of measurement, e.g., the X-ray
apparatus of Wilhelm Conrad Röntgen, the
magnetic resonance imaging (MRI)-
technology of Paul Lauterbur and Peter
Mansfield, and the scanning tunneling
microscope of Heinrich Rohrer and Gerd
Binnig, phenomena could be seen that
were invisible before, new connections
could be made, new diagnoses derived,
new therapies tried out, etc.
These new measuring technologies had
one thing in common: they advanced their
disciplines. Now the question is whether
the IOT, this giant global network of
sensors, plays in the same league,
whether it truly has the power to advance
business administration, economy,
especially the behavioral part of it, or any
other disciplines that are linked to the
management of man-made organizations.
Things cannot not communicate
leveraging the power of the
physical world
Paul Watzlawick once famously stated that
humans cannot not communicate. They
communicate constantly on a functional
(e.g., talking) and/or emotional (e.g., blank
glance or knowing smile) level. Now, let us assume that goods also cannot not communicate.
On the functional level, the bottle of water on my desk, for instance, communicates via its
label its name, origin, exact ingredients, capacity, expiration date, EAN-code, etc. The design
of the bottle, including the label, carries emotional values such as “I am from the mountains,”
“I am healthy, fresh, and I taste great,” “Touch me, buy me, open me, drink me.” That is what
product designers are paid for.
Figure 3: Low resolution versus high resolution
management
IOT-technology can increase both the functional and the emotional communication
capabilities of things and places. For instance, on the functional level, it can generate new
information on product quality (was the cool chain broken?), authenticity (is it a fake?), rating
(do my friends like it?), and price (is it cheaper next door?). On the design level, a smart
bottle could, for instance, show its newest commercial, or change its color with its
temperature, when it is opened, or when it is touched. And most importantly, it could do so in
a relatively unobtrusive way: the bottle still could sport a lean design because most of the
additional functionality is only brought to life at the will of a user, man or machine, via a wand
such as a mobile phone.
Smart things enable rich but hidden functionality that can be built to serve all parties in their
value chain, including producer, consumer, transportation, customs, repair centers, and
financial service providers. As things become smart, they turn into physical anchors for
various services. At the same time, they maintain or even increase the emotional
attractiveness only physical things can offer.
How companies make use of the IOT
This section summarizes some intriguing observations my colleagues and I made when
shadowing companies as they started to introduce IOT-technologies. We subsume the
findings under the term High Resolution Management (HRM), which stands for a
management that consequently leverages the power of sensor data to increase visibility and
exploit it for business excellence
12
.
Go for complex problems nuts do not require a sledgehammer
The first question every organization that is aware of a new tool has to ask itself is: where
would it make sense, if at all, to utilize the power of the tool?
13
Which criteria must a problem
satisfy to be eased by the new tool? In the case of HRM, the answer is: do not go for simple
challenges, go for complex ones. You do not use MRI to diagnose a scratch.
A problem is simple when, thinking in terms of a socio-technical system, it only involves a
few nodes with a few states, which behave deterministically. An example would be a mass-
production facility that runs only a few stable machine tools, linked by a fully automated
conveyor belt. In this case, simple rules and management tools are the most cost-efficient
means to control the problem.
12
see (Fleisch, Müller-Stewens, 2009) for an introduction to High Resolution Management.
13
Yes, it is true: in many cases the problem not only triggers the search for a solution, but also the availability
of the potential cure.
However, if the number of nodes (and thus edges) in a problem is high, and the nodes
behave non-deterministically, the complexity of the entire system explodes.
14
It is easy to find
complex problems in every industry. One example is the management of a textile retail chain
where every day in every store, a dozen or so sales assistants serve hundreds of customers
and move around thousands of articles of clothing. Without a strong organizing power, a
retail chain would end in chaos within a few days, simply because it is bound to the same
forces of entropy that seem to turn your children’s rooms into a mess without anybody even
entering them.
The first thing companies try to do is to avoid complexity, e.g., by reducing the number of
product variants
15
. Whenever that is not possible, e.g., for competitive reasons when a
company wants to offer a rich set of individualized products, services, or experiences, the
managing organization has to increase its management capability until the power of the
solution matches the power of the problem. One way to increase the management capability
is to leverage sensor-based data, because they increase the number of potential states in
the management system dramatically. Thus, they help the management system to absorb
the complexity of the managed system
16
and, as a consequence, lead to more efficient and
effective results.
This explains why the most discussed applications of the IOT deal with complex problems,
e.g., the management of large numbers of assets in a supply chain, and are hard to
implement. It also explains why it seems not to be wise to use IOT-technologies to solve
simple problems that have already been worked out with simpler means.
Search for blind spots
Companies using IOT-technologies often experience an effect that physicians might have
learned when they used MRI for the first time: they identify problems they were not looking
for but that just became visible because of the newly available high-resolution data. For
instance, when a car manufacturer tagged his work-in-progress automobiles to better control
the just-in-time deliveries during the assembly process, he was extremely surprised to see
that problem-cars, cars that generated more challenges than usual during assembly, were
regularly put on a holding track. The assembly teams did so to push off the problem-car to
the next shift so they could manage their performance metrics, e.g., how many cars they
assemble in a shift. This approach was not only unfair to the subsequent assembly teams,
who, by the way, did the same. It also generated a costly and ever-growing mismatch
between the order of the highly individual cars on the assembly line and the order of the just-
in-time delivered parts such as doors or windows to be mounted. A logistics company was
14
The number of potential system states is a measurement for the complexity of a system in systems theory.
15
Check the number of product variants Toyota and, say, Daimler or BMW offer. And then compare the
profitability of both companies.
16
The thinking of this paragraph is based on Ashby’s law that states the “only variety can absorb variety”.
flabbergasted when they learned the trusted numbers and reasons for their far higher than
expected rate of delayed deliveries: one of their truck drivers did not like to drive west into
the sunset, so he regularly took a lengthy and costly detour north and only turned west (and
back south) when it was dark. A consumer goods company was surprised to learn that only
20% of the promotion displays they shipped to one of their retailer’s stores were treated as
negotiated and planned.
Diagnose and improve
This blind spot phenomenon is also responsible for some of the difficulties when compiling a
business case: it is simply impossible to calculate the return of an investment when the
problems that are to be solved are partly unknown. That is why some companies utilize IOT-
technology in a first step simply as a diagnostic tool, i.e., to generate a trusted picture of the
reality of their practiced routines. They then use the resulting extensive sets of sensor data to
improve their processes. For instance, one retail company measured the impact of different
types of merchandise presentation on sales and, for example, found out that some trousers
sold better when they were presented hanging on racks than when they were lying on
shelves (Thiesse et al., 2009). After learning a lesson like that, the retailer only needs to
change his routine. He does not need to keep the IOT-system running to gain the benefit.
Automate low level management
In many cases, companies use IOT-technology as a tool for the next step in industrialization:
they automate simple manual tasks such as signing in cargo, updating stock keeping
records, initiating replenishment processes, detecting failures, and sending notifications.
Thus they eliminate very low-level coordination work that was previously executed by
humans.
At the core of the matter, IOT-technology drastically reduces the cycle time of an operational
management cycle with the three steps “do” (perform a task), “check” (compare the results of
the task with the expected values) and “act” (introduce a correction if needed) by automating
the “check”-step (e.g., measuring a tire pressure) and often also the “act”-step (e.g., sending
a notification). This allows a continuous comparison of actual with expected values and
enables, as a consequence, the early detection of deviations, which is key in keeping the
impact of an error as small as possible (as the impact of an error often grows exponentially
over the time it remains undetected).
Constant sensor-based checks eventually enable information systems to automatically detect
relevant real-world events and build the basis for an operational management by exception,
in which the installed information systems and routines deal on their own in all foreseen
situations. They only call human managers for help when they detect an unknown state of
affairs.
Measure, manage, and innovate
Of course, companies use IOT-technology for more than just diagnosis and low-level
automation. When we define using IOT-technology as a diagnostic tool as business
innovation level 0, then purely automating but not changing business processes would be
level 1.
The next level, level 2, includes the IOT-enabled modification of business routines, e.g.,
changing from a clerk-operated check-out process to an RFID-enabled self-check-out in a
library. In level 3 business innovations, companies integrate IOT value drivers into their
product or service offerings. For instance, a toothbrush company turns some of its products
into smart products by equipping the shafts and brushes of its electric produces with tiny
computers. These computers align the shaft’s movements with the shape of the brush, to
measure how often and how persistently each family member (each of whom is using the
same expensive shaft but his own plug-in brush) cleans his teeth, to encourage both children
and grownups to keep on brushing by interacting with a brushing Mickey Mouse in the smart
bathroom mirror, and to prevent customers from using cheaper brushes.
On level 4, in terms of change, the highest level of business innovation, companies use IOT
to transform their business model. Once a company has absolute visibility (think of it as a
real-time video stream, in contrast to weekly black and white photos) of its most important
objects, such as beer kegs in the case of a logistics provider for breweries, cars in case of an
insurance company, or drilling machines in case of a machine tool company, it can, for
instance, switch from selling its products to renting them to its customers on a pay-per-use
basis with a huge impact for both vendor, and customer.
Summary
In this paper, I tried to answer the question “What is the Internet of Things?” I did so by
digesting a careful study of hundreds of applications that automatically or semi-automatically
integrate real-world objects and places with the Internet. In the first step, I looked at the
differences between Internet applications and applications that probably belong in the
category of the IOT, mainly to sharpen my own understanding of what the unique features of
an IOT would be and where IOT-specific challenges could be found. I identified six
characteristics that suggest that integrating the real world with the Internet requires a new set
of infrastructure building blocks.
In the next part, I tried to group the IOT-applications to derive some common design-
schemes. After several unsuccessful attempts, I started to look at the value drivers of each
application, i.e., for each application I searched for the IOT-related root-cause of a benefit for
businesses and users. I identified seven value drivers. Each investigated application used
one or more of these value drivers.
All value drivers were related to the reduction of the real world-virtual world transaction cost.
Taking that as a starting point, I looked for the main economic energy of the IOT. The result
of this endeavor suggests that the IOT will eventually provide management systems with low-
cost, high-resolution data about the real world. The IOT therefore has the potential to
become an MRI-technology for businesses and society, with all its attached advantages and
drawbacks: it might become a tool that advances the entire discipline of how to manage
organizations and complex systems.
In the last section, I looked at some patterns in how companies make use of IOT and found
some intriguing observations that will hopefully help readers to shorten the learning curve of
their organization.
Limitations
Although I studied many IOT-applications, I almost certainly left out entire sets of applications
and technologies that could potentially influence the derived results, i.e., the value drivers
and the building blocks. Future work with access to additional scenarios will test the
robustness of the proposed frameworks.
In this paper, I focused very heavily on the questions of how and where the IOT could add
value to users and organizations. I deliberately did not analyze where the real-world MRI
would add risks, although the potential negative aspects are obvious. Since Pandora opened
the box, we all know that every technology hat two sides. Given the proposed
momentousness of the IOT, an in-depth investigation of the consequences that must end in
concrete applicable concepts
17
seems to be inevitable.
Outlook
If the comparison with MRI holds true, the IOT will feed legions of academic and industrial
researchers and developers with challenging and fascinating questions for many years.
Example questions across all layers include: How can organizations efficiently derive insights
from massive sensor data to improve their offerings and operations (the key phrase here is
“real-world mining”)? How can I protect myself from being scanned all day long? How can
society leverage the power of this new insight to change the world for good, e.g., by
developing mind-changing applications to help consumers to use scarce resources such as
water and electrical power in a responsible way? What if robots join the IOT and things not
only become massive sensors but also actuators? How can we build even more energy-
efficient and autonomous minicomputers? The race is certainly on.
17
Pentland, for instance, suggests in his “new deal on data” to treat data like money (Pentland, 2009).
Acknowledgement
Thanks to my colleagues Friedemann Mattern, Sanjay Sarma, Günter Müller-Stewens,
Stephan Haller, Hubert Österle, Florian Michahelles, and Frederic Thiesse for their
invaluable input.
About the Author
Dr. Elgar Fleisch, professor of information and technology management, ETH Zurich and
University of St. Gallen (HSG)
The research of Prof. Fleisch focuses on the economic impacts and infrastructures of
ubiquitous computing. In the Auto-ID Lab, he and his team develop, in concert with a global
network of universities, an infrastructure for the Internet of Things. In the Bits-to-Energy Lab,
which he co-chairs with Prof. Mattern of ETH Zurich, he investigates and designs
technologies and applications to save electrical power, and in the Insurance Lab, Elgar
Fleisch, together with Prof. Ackermann of HSG, drives forward technology-based innovation
in the insurance industry. All research projects are joint efforts of industry and academia;
their results have been published in more than 200 scientific journals and books.
Elgar Fleisch is a co-founder of several university spin-offs, e.g., Intellion, Synesix, Coguan,
Dacuda and Amphiro, and he serves as a member of multiple management boards and
academic steering committees.
References
Allmendinger, G.; Lombreglia, R. (2005): "Four strategies for the age of smart services“. In:
Harvard Business Review. 83 (10), pp. 131-145.
Ariely, D. (2008): Predictably Irrational. New York, New York, USA: Harper Collins.
Bullinger, H.-J.; Ten Hompel, M. (2008): Internet der Dinge. Berlin: Springer.
Coroama, V. (2006): "The Smart Tachograph – Individual Accounting of Traffic Costs and its
Implications“. In: Pervasive 2006. Springer, pp. 135-152.
DeHoratius, N.; Raman, A. (2008): "Inventory Record Inaccuracy: An Empirical Analysis“.
In: Management Science. 54 (4), pp. 627-641.
Ferguson, G.T. (2002): "Have your objects call my objects“. In: Harvard Business Review.
Harvard Business School Publishing Corporation 80 (6), pp. 138–144.
Filipova, L.; Welzel, P. (2005): "Reducing Asymmetric Information in Insurance Markets:
Cars with Black Boxes “. In: Volkswirtschaftliche Diskussionsreihe / Inst. f.
Volkswirtschaftslehre der Universität Augsburg, Vol. 270 .
Fleisch, E.; Müller-Stewens, G. (2009): "Der Einfluss der RFID-Technologie auf die
Unternehmensführung“. In: Hünerberg, R.; Mann, A. (2009): Ganzheitliche
Unternehmensführung in dynamischen Märkten: Festschrift für Univ.-Prof. Dr. Armin
Töpfer. Gabler Verlag, pp. 69-91.
Flörkemeier, C.; Roduner, C.; Lampe, M.(2007): "RFID Application Development With the
Accada Middleware Platform“. In: IEEE Systems Journal. 1 (2), pp. 82-94.
Gershenfeld, N.; Krikorian, R.; Cohen, D. (2004): "The Internet of Things“. In: Scientific
American. 291 (4), pp. 76-81.
Gershenfeld, N. (1999): When Things Start to Think. New York: Henry Holt.
Henzinger, T.; Sifakis, J. (2007): "The Discipline of Embedded Systems Design“. In:
Computer. 40 (10), pp. 32-40.
Hui, J. W.; Culler, D. E. (2008): "IP is dead, long live IP for wireless sensor networks“. In:
Proceedings of the 6th ACM conference on Embedded network sensor systems. New
York: ACM Press pp. 15-28.
ITU (2005): "The Internet of Things“. ITU Report, Genf.
Keegan, S.; O'Hare, G.; O'Grady, M. (2008): "Easishop: Ambient intelligence assists
everyday shopping“. In: Journal of Information Sciences. 178 (3), pp. 588-611.
Kindberg, T.; Barton, J.; Morgan, J. et al. (2002): "People, places, things: Web presence
for the real world“. In: Mobile Networks and Applications. 7 (5), pp. 19-28.
Lampe, M.; Hinske, S.; Ryu, H.; Milrad, M.; Ogata, H. (2007): "Integrating Interactive
Learning Experiences into Augmented Toy Environments“. In: Workshop on Pervasive
Learning. pp. 1-9.
Langheinrich, M.; Coroama, V.; Bohn, J.; Mattern, F. (2005): "Living in a smart
environment - Implications for the coming ubiquitous information society“. In: IEEE
International Conference on Systems 2004. pp. 5633-5638.
Lee, H.; Oezer, O. (2007): "Unlocking the value of RFID“. In: Production and Operations
Management 16 (1), pp. 40–64.
Mattern, F. (2004): "Wireless future: Ubiquitous computing“. In: Proceedings of Wireless
Congress 2004.
Mattern, F.; Flörkemeier, C. (2009): "Vom Internet der Computer zum Internet der Dinge".
URL: http://www.vs.inf.ethz.ch/publ/papers/Internet-der-Dinge.pdf. Zurich, 2009.
Meyer, G.; Främling, K.; Holmström, J. (2009): "Intelligent Products: A survey“. In:
Computers in Industry. 60 (3), pp. 137-148.
Ong, S. K.; Yuan, M. L.; Nee, A. Y. C. (2008): "Augmented reality applications in
manufacturing: a survey“. In: International Journal of Production Research. 46 (10), pp.
2707-2742.
O'Reilly, T. (2005): "What Is Web 2.0“. URL: http://oreilly.com/web2/archive/what-is-web-
20.html.
Payne, R; MacDonald, B (2004): "Ambient Technology — Now You See It, Now You Don't“.
In: BT Technology Journal. 22 (3), pp. 119-129.
Pentland, A.; Dutta, S.; Mia, I. (2009): "The Global Information Technology Report 2008-
2009“. In: The Global Information Technology Report 2008-2009. World Economic
Forum, pp. 75-80.
Regenbrecht, H.; Baratoff, G.; Wilke, W. (2005): "Augmented reality projects in the
automotive and aerospace industry“. In: IEEE Computer Graphics and Applications. 25
(6), pp. 48–56.
Sarma, S. (2004): "Integrating RFID“. In: ACM Queue. 2 (7), pp. 50-57.
Sarma, S. (2001): "Towards the Five-Cent Tag“. In: MIT Auto-ID Center White Paper. MIT
Auto-ID Center.
Schoenberger, C.; Upbin, B. (2002): "The Internet of Things“. In: Forbes Magazine. 169 (6),
pp. 155-160.
Sellitto, C.; Burgess, S.; Hawking, P. (2007): "Information quality attributes associated with
RFID-derived benefits in the retail supply chain“. In: International Journal of Retail &
Distribution Management. 35 (1), pp. 69-87.
Staake, T.; Michahelles, F.; Fleisch, E. et al. (2008): "Anti-Counterfeiting and Supply Chain
Security“. In: Cole, P. H.; Ranasinghe, D. C.: Networked RFID Systems and
Lightweight Cryptography. Berlin: Springer pp. 33-44.
Thiesse, F. (2007): "RFID, privacy and the perception of risk: A strategic framework“. In: The
Journal of Strategic Information Systems. 16 (2), pp. 214-232.
Thiesse, F.; Flörkemeier, C.; Harrison, M. et al. (2009): "Technology, Standards, and Real-
World Deployments of the EPC Network“. In: IEEE Internet Computing. 13 (2), pp. 36-
43.
Thiesse, F.; Al-Kassab, J.; Fleisch, E. (2009): "Understanding the value of integrated RFID
systems: a case study from apparel retail“. In: European Journal of Information
Systems. 18 (6).
van der Heijden, H. (2006): "Mobile decision support for in-store purchase decisions“. In:
Decision Support Systems. 42 (2), pp. 656-663.
Vitzthum, S.; Konsynski, B. (2008): "CHEP: The Net of Things“. In: Communications of the
AIS. 22 , pp. 485-500.
Wark, T.; Corke, P.; Sikka, P. et al. (2007): "Transforming Agriculture through Pervasive
Wireless Sensor Networks“. In: IEEE Pervasive Computing. 6 (2), pp. 50-57.
Wright, S.; Steventon, A. (2004): "Intelligent Spaces The Vision, the Opportunities and
the Barriers“. In: BT Technology Journal. 22 (3), pp. 15-26.
von Reischach, F.; Guinard, D.; Michahelles, F.; Fleisch, E. (2009): "A mobile product
recommendation system interacting with tagged products“. In: IEEE International
Conference on Pervasive Computing and Communications.
... Intelligent transportation solutions such as asset tracking and mobile ticketing are examples, whereas intelligent health solutions such as patient monitoring and chronic disease management are examples of intelligent health solutions. In addition, smart city efforts are looking at technology such as live monitoring of vacant parking spots and intelligent street lighting, among other possibilities [5], [6], [7]. ...
... 5: Assume that each group has an average and re-allocate the cluster centers to their new positions. 6: Perform a repetition on Steps 3-5 until the center points do not move. 7: Identify and calculate the clustering SSE for each generated group, as well as the overall clustering SSE. ...
Preprint
Full-text available
Big data streams are possibly one of the most essential underlying notions. However, data streams are often challenging to handle owing to their rapid pace and limited information lifetime. It is difficult to collect and communicate stream samples while storing, transmitting and computing a function across the whole stream or even a large segment of it. In answer to this research issue, many streaming-specific solutions were developed. Stream techniques imply a limited capacity of one or more resources such as computing power and memory, as well as time or accuracy limits. Reservoir sampling algorithms choose and store results that are probabilistically significant. A weighted random sampling approach using a generalised sampling algorithmic framework to detect unique events is the key research goal of this work. Briefly, a gradually developed estimate of the joint stream distribution across all feasible components keeps k stream elements judged representative for the full stream. Once estimate confidence is high, k samples are chosen evenly. The complexity is O(min(k,n-k)), where n is the number of items inspected. Due to the fact that events are usually considered outliers, it is sufficient to extract element patterns and push them to an alternate version of k-means as proposed here. The suggested technique calculates the sum of squared errors (SSE) for each cluster, and this is utilised not only as a measure of convergence, but also as a quantification and an indirect assessment of the element distribution's approximation accuracy. This clustering enables for the detection of outliers in the stream based on their distance from the usual event centroids. The findings reveal that weighted sampling and res-means outperform typical approaches for stream event identification. Detected events are shown as knowledge graphs, along with typical clusters of events.
... This enables the labels to respond to a weaker flag from the readers. [4] In (figure 4) [7], dynamic labels are depicted. ...
Conference Paper
Full-text available
This research is an exploration into developing a system for enabling Radio Frequency Identification (RFID) labels to be connected to the Internet while taking into account their unique impediments. Additionally, this mechanism enables the tag to be extraordinarily distinct and spoken to as a communication material capable of communicating with other participants, which can facilitate and rearrange the use of the "Internet of Things" concept in the not-too-distant future. To build a mechanism capable of connecting RFID labels to the Internet. The methods taken by various researchers are investigated and dissected, enabling a better understanding of the difficulties and shortcomings associated with RFID labels connected to the Internet. The analysis and examination have resulted in the creation of another system that allows use of TCP/IP. The structure established in this paper is predicated on the capability of RFID labels to be used as procedures (TCP forms) within a host. As a result, each procedure has a procedure ID or port number, which enables various members to identify and communicate with the tag through the process ID. This is accomplished through a built-in interpretation portion that converts the RFID tag's authentic personality (ID) to a new ID that can be recognized as a TCP port number. The results of this paper show that the system worked effectively for the purpose for which it was designed. The results show that the actualized system enables RFID labels to be linked to the Internet and to be exceptionally distinct. Additionally, it enables labels to send and receive information and guidance outside of the RFID system, through the Internet, and from various members. The framework's success would provide several experts with opportunities to actualize the concept of "Internet of Things."
... The Internet of Things (IoT) is related to the connectivity of smart things on the Internet (Fleisch, 2010) and will also significantly affect enterprise systems. As the Internet connects computers, IoT connects a huge number of objects, devices, machines, etc., using sensors and providing data generation, storage, and analytics capabilities (Aryal et al., 2018). ...
Article
Enterprise Systems have become a critical feature of organisations not only for their integrative functions but also because they have become collectors and repositories of organisational data. In the past, this data was subject to relatively simplistic analysis, which could provide important but still basic information for management decision-making. With the emergence of new technologies, both data collection and analysis have become increasingly sophisticated and varied. This has led to significant improvements in the quality and timeliness of the information that can be used by management in decision-making and strategic planning. Even how managers use this information is being shaped by technologies that can augment human knowledge to further improve decision-making. With reviewing the literature, this editorial paper explains the scope of emerging technology, enterprise systems and the impacts on knowledge management and briefly introduces the articles included in this issue.
Article
Full-text available
This paper aims to probe the influence of innovation spillovers in the artificial intelligence (AI) and financial technology (Fin-tech) industries on the value of the internet of things (IoT) companies. Python was utilized to download public information from Yahoo Finance, and then the GARCH model was used to extract the fluctuations of cross-industry innovation spillovers. Next, the Fama–French three-factor model was used to explore the interactive changes between variables. The panel data regression analysis indicates that the more firms accept innovation spillovers from other industries, the better the excess return; however, this effect differs because of industrial attributes and the environmental changes induced by COVID-19. Additionally, this study finds that investing in large-cap growth stocks of IoT firms is more likely to yield excess returns. Finally, the study yields lessons for policy leverage to accelerate the upgrading and transformation of innovation-interactive industries by referring to the practices of Singapore and South Korea.
Article
Full-text available
Abstract Consumption habits are shaped by time and the conditions of the time. At this point, technological developments are very effective in changing consumer behavior. Traditional consumer identity has undergone great changes, especially in the present day of the fourth industrial revolution, and has gained a great dynamism and complexity day by day with the conveniences provided by technology. The fourth industrial revolution, or industry 4.0, is a period in which life becomes easier and more radical changes are experienced with an information-technology-oriented approach. In this period, there is a continuous development of informatics. Electronic devices, adapting these devices to life, and managing life with a rational adaptation process are in question. The internet of things, which is one of the important concepts of this period, has become one of the concepts that shape consumer behavior. Objects are the elements of the physical world (physical things) or the information world (virtual world) that can be defined and integrated into communication networks. The Internet of Things (IoT) encompasses all physical objects, i.e. "things", that connect to the Internet and other devices. The provision of fast, effective and interactive communication opportunities from health to education, from entertainment to many other fields, together with the internet of things, has enabled consumers to shape their consumption habits in this direction. This article focuses on consumer behavior shaped by the internet of things. Between 2017 and 2022, national and international literature samples were compiled on the internet of things and the change in consumer behavior, and a meta-analysis was made on them. Özet Tüketim alışkanlığı zamana ve zamanın şartlarına bağlı olarak şekillenmektedir. Bu noktada teknolojik gelişmeler tüketici davranışlarının değişmesinde oldukça etkilidir. Geleneksel tüketici kimliği, özellikle dördüncü sanayi devriminin yaşandığı günümüzde büyük değişimlere uğramış, teknolojinin sağladığı kolaylıklarla birlikte her geçen gün büyük bir dinamizm ve karmaşa kazanmıştır. Dördüncü sanayi devrimi ya da endüstri 4.0 dönemi, yaşamın giderek daha da kolaylaştığı ve bilişim-teknoloji odaklı olarak bu kolaylığın köklü değişimler yaşattığı bir dönemdir. Bu dönemde süreklilik arz eden bir bilişim gelişimi söz konusudur. Elektronik cihazlar, bu cihazların yaşama adapte edilmesi, rasyonel bir adaptasyon süreciyle yaşamın yönetilmesi söz konusudur. Bu dönemin önemli kavramlarından olan nesnelerin interneti tüketici davranışlarını şekillendiren kavramlar arasına girmiştir.Nesneler, tanımlanabilen ve iletişim ağlarına entegre edilebilen fiziksel dünyanın (fiziksel şeyler) veya bilgi dünyasının (sanal dünya) öğeleridir. Nesnelerin interneti (IoT) internete ve diğer cihazlara bağlanan tüm fiziksel nesneleri, yani "nesneleri" kapsamaktadır. Nesnelerin internetiyle birlikte sağlıktan eğitime, eğlenceden daha birçok alana kadar hızlı, etkili ve interaktif iletişim olanaklarının sağlanması tüketicilerin de bu yönde tüketim alışkanlıklarını şekillendirmelerini sağlamıştır. Bu makale, nesnelerin internetiyle birlikte şekillenen tüketici davranışlarına odaklanmaktadır. 2017-2022 yılları arasında nesnelerin interneti ve tüketici davranışlarının değişimine yönelik olarak ulusal ve uluslararası literatür örnekleri derlenerek, bunlar üzerinden bir meta analiz yapılmıştır.
Article
Over time, consumer awareness of environmental protection has grown. From big data in the Internet of things (IoT), green advertisement has indeed made a significant impact on consumers’ awareness of environmental protection. Therefore, this research studied the impact of the IoT and green advertisements on consumers’ behavior. An investigation was conducted based on focus group interviews, and the assumption was tested using statistical methods. The research results were as follows: 1) advertisement has a significant impact on consumer behavior; 2) consumers are willing to purchase advertised products that convey positive messages, regardless of their age, educational level, or average income level; 3) consumers with higher education levels have better environmental concepts; and 4) green advertisement has a significant impact on consumers’ environmental concepts and behaviors. In summary, the study found that green advertisement can improve consumers’ awareness of environmental protection and that consumers’ behavior can effectively influence the implementation of environmental protection.
Article
This study aims to investigate the applications of the Internet of Things (IoTs) and highlights the challenges faced to implement IoTs in Pakistani university libraries. This study opted exploratory sequential mixed-methods approach and was completed in two phases. In the first phase, purposefully selected university librarians were interviewed. In the second phase, quantitative data were collected from 122 university librarians across Pakistan. This study found that Pakistani university libraries implemented IoTs-based appliances including smart air conditioners, automatic fire alarms, smart hand sanitizer machines, and smart security doors. Moreover, the university libraries were using limited IoT-based services that are auto notification of check-out, check-in the reading materials, self-check-out and self-check-in system, recognition of user's card, and the use of Radio Frequency Identification (RFID) tags for security perspective. The major challenges to implementing IoTs applications were found lack of a highly networked and integrated environment, budget problems, unavailability of policy and strategic plan, and the lack of technical personnel. This novel study explains the extent of IoTs applications and the challenges to adopt and implement IoTs in Pakistani university libraries. The research findings would be useful for library and university administrators who are interested to implement IoTs applications in their libraries.
Chapter
The internet of things (IoT) is a global ecosystem of networked “things.” It is the subject of much research worldwide, although it still has many challenges to overcome before it can achieve its full potential. Many papers have been written on the IoT and related areas including big data analytics, smart cities, and industrial IoT (IIoT). These challenges have mostly been seen as technical, although the IoT's business and societal challenges are also important. Most authors of research papers discuss the research challenges with which they are most familiar, but a framework which identifies and classifies all the challenges and cross-references the publications describing them in detail, is much needed. In this chapter, the authors extend their earlier IoT classification scheme to include more recent papers, and business and societal challenges as well as technical ones. The nature of the classification scheme and research challenges are described; however, the other chapters of this book cover in more detail the individual challenges and proposed strategies to mitigate them.
Article
New technologies constantly change the paradigm of how businesses will be run in the future. The Internet of Things (IoT) enables new business opportunities for data-driven transformation in organisations. The emergence of the IoT concept has resulted in numerous definitions, with earlier references primarily focussed on the technological aspects. This has hindered the broader diffusion of the term IoT as, arguably, the definitions do not integrate other non-technological elements of IoT and focus more on business and service provisions. To resolve this, our research identifies the most significant building blocks required in designing an IoT system. This is accomplished through a methodological review of 122 definitions and their consolidation into a novel definitional framework. The definitional framework unifies the traditional technology focus of the earlier definitions and integrates additional elements that are likely to increase the adoption of a comprehensive definition to support the development of future business applications. Furthermore, the framework serves as a reference set for scholarly societies and standards organisations who, in the future, can be tasked with formulating a definition of IoT, as has been the case with the NIST definition of Cloud Computing, which was preceded by several academic studies on defining the term.
Chapter
Disaster management's objective is to minimise the potential damage caused by disasters, to provide victims with immediate and appropriate assistance, and to ensure an effective and rapid recovery. To accomplish these goals in the aftermath of a disaster, a coordinated and efficient rescue effort is required. As a result, breadth of information about the disaster's impact is required in order to plan an immediate and effective response. The internet of things (IoT) is poised to save lives in the event of a natural disaster. This chapter proposes an IoT-based solution for planning rescue operations in the aftermath of natural disasters. This chapter is further validated through an analysis of IoT technology adoption for disaster management using the task-technology fit (TTF) approach.
Article
Full-text available
R FID (Radio-Frequency Identification) technology has shown itself to be a promising technology to track movements of goods in a supply chain. As such, it can give unprecedented visibility to the supply chain. Such visibility can save labor cost, improve supply chain coordination, reduce inventory, and increase product availability. Industry reports and white papers are now filled with estimates and proclamations of the benefits and quantified values of RFID. Early adopters are now rallying more and more followers. However, most such claims are educated guesses at best and are not substantiated, that is, they are not based on detailed, model-based analysis. This paper argues that there is a huge credibility gap of the value of RFID, and that a void exists in showing how the proclaimed values are arrived at, and how those values can be realized. The paper shows that this credibility gap must be filled with solid model analysis, and therefore presents a great opportunity for the Production and Operations Management (POM) research community. The paper reviews some of the ongoing research efforts that attempt to close the credibility gap, and suggests additional directions for further strengthening the POM's contribution to help industry realize the full potentials of RFID.
Article
Full-text available
Monitoring and recording driving behavior has become technologically feasible recently which allows inference of drivers’ risk types. We examine the effects of such technologies in automobile insurance markets with adverse selection for both perfect competition and monopoly. Specifically, we assume that insurers can offer a contract with access to recorded information ex post, i.e., after an accident, in addition to the usual second-best contracts. We find that this leads to a Pareto-improvement of social welfare except when high risks initially received an information rent. Regulation can be used to establish Pareto-improvement also in these cases. Explicit consideration of privacy concerns of insurees does not alter our positive welfare results.
Article
The continuing progression of Moore's Law has enabled the miniaturisation and dramatic cost reduction in electronics over the last ten years. For a truly pervasive communications environment the challenges of hiding key hardware technologies from the user are rapidly being overcome. This paper reviews the status of these hardware technology developments in the pervasive space and briefly discusses other contributing factors that will enable the pervasive vision to be realised.
Article
Augmented reality (AR) is a novel human–machine interaction that overlays virtual computer-generated information on a real world environment. It has found good potential applications in many fields, such as military training, surgery, entertainment, maintenance, assembly, product design and other manufacturing operations in the last ten years. This paper provides a comprehensive survey of developed and demonstrated AR applications in manufacturing activities. The intention of this survey is to provide researchers, students, and engineers, who use or plan to use AR as a tool in manufacturing research, a useful insight on the state-of-the-art AR applications and developments.
Article
Over the last 30 years, we have seen the power of microprocessors double about every 18 months. An equally rapid increase applies to some other technological parameters such as stor- age capacity and communications bandwidth. This continuing trend means that computers will become considerably smaller, cheaper, and more abundant - they are becoming ubiquitous, and are even finding their way into everyday objects. This is resulting in the creation of "smart" things that can access the Internet and its varied resources, and maybe even cooperate with each other. Mobile phones are a forerunner in this technological field - they are now true computers equipped with a whole range of functionality and may well develop into control centers for a multitude of other personal auxiliary services. In the long term, ubiquitous computing will take on great economic significance. Industrial products will become "smart" due to their integrated information processing capacity, or take on an electronic identity that can be remotely queried, or be equipped with sensors for detecting their environment, enabling innovative products and totally new services to be developed. However, an "informatized" world full of objects that can detect aspects of their environment and communicate with each other also has serious societal implications. The social and political challenges of the ubiquitous computing era will be characterized by an increasing dependence on technology, control over the information to which everyday objects are linked, and the protection of privacy. 1. The trend towards invisible and ubiquitous computer technology
Purpose – This paper aims to report on research that examined the recent scholarly literature to identify the information quality attributes associated with radio frequency identification (RFID)-based benefits across sectors of the retail supply chain. Design/methodology/approach – Reflecting the recent interest in RFID technology, the literature review was limited to scholarly articles published since the late 1990s when there appears to have been a surge in research and publishing activity. Moreover, the paper uncouples RFID-focused technology findings that are a feature of many previous publications and reports on the decision-making attributes associated with the perceived benefits of adopting the technology. Findings – Many RFID-based benefits were found to be associated with the distribution and transportation sectors of the supply chain, however, an emerging number are also apparent in the retail and post retail domains. The improved information value associated with RFID-derived benefits was embodied in quality attributes that included timeliness, currency, accuracy and completeness. The paper proposes an RFID information value chain that maps benefits and information attributes across the supply chain. The paper is also one of the first that attempts to relate RFID-derived information with aspects of organisational decision making. Research limitations/implications – This study identified information attributes associated with RFID adoption within the retail supply chain that have led to enhanced organisational responsiveness through improved decision-making capabilities. As exploratory research in a nascent and emerging area, this research should be viewed as a starting point in the examination and identification of RFID-derived benefits and information, rather than a prescriptive and/or definitive type of classification system for RFID. Practical implications – Practical examples of RFID-derived benefits distilled from the literature tend to provide important retail supply chain lessons for organisations that are currently piloting or expecting to trial RFID. The study highlights operational and strategic implications of adopting RFID technology discussing them from an information value perspective. Originality/value – The paper is one of the first that examines the information value of RFID-derived benefits across the organisational supply chain. Moreover, both benefits and information attributes are mapped to specific sectors of the retail and distribution supply chain. http://www.emeraldinsight.com/journals.htm?articleid=1593225