![Usability Engineering process (© IEC 62366:2007, Fig. D.1)[27]](https://www.researchgate.net/profile/Francesco-Flammini/publication/330643330/figure/fig1/AS:786298361102338@1564479546333/Usability-Engineering-process-C-IEC-623662007-Fig-D127_Q320.jpg)
![Hexeract [41] (six-dimensional hypercube) -Petrie polygon Orthographic projections -The representation of the sixth dimension as the Smart Cyberspace It is quite difficult to illustrate a four-dimensional space in two dimensions. In case of a six-dimensional space, however, it is almost impossible. Figure 8 shows one way of representing a six-dimensional space in two dimensions. The Smartness Quotient will be defined in this space. The Smartness Quotient can be illustrated with the help of the Hexeract. The six dimensions are the following: Dim. 1: Autonomy, Dim. 2: Human-Machine Interaction, Dim. 3: Controllability For Complicated Dynamics, Dim. 4: BioInspired Behavior, Dim. 5: Usability Index Of Machine, Dim. 6: Environmental Performance Index of Machine. The values of these dimensions determine a point in the special six-dimensional Smart Cyberspace. The six-dimensional hypercube represents the space in which the Smartness Quotient is a point, like in the case of the MIQ. Why should spaces be discussed in such detail? The world and its higher dimensions have always interested humans, and it could not be possible to imagine the existence of smart devices if we did not know how they can become smart. The above described theory can show us how human mind can create from smart machines and devices a space that is no longer cognitive or fictive, but a real Smart Cyberspace.](https://www.researchgate.net/profile/Francesco-Flammini/publication/330643330/figure/fig3/AS:786298361102339@1564479546407/Hexeract-41-six-dimensional-hypercube-Petrie-polygon-Orthographic-projections-The_Q320.jpg)
Content uploaded by Francesco Flammini
Author content
All content in this area was uploaded by Francesco Flammini on Jul 30, 2019
Content may be subject to copyright.
Complex, Resilient and Smart Systems
Dániel Tokody1, a, József Papp2, b,
László Barna Iantovics3, c, Francesco Flammini4,d
1,2Óbuda University, Doctoral School on Safety and Security Sciences, Budapest, Hungary
atokody.daniel@dosz.hu, bpapp.jozsef@kvk.uni-obuda.hu
3Petru Maior University, Tirgu Mures, Romania
cibarna@science.upm.ro
4Linnaeus University, Växjö, Sweden
dfrancesco.flammini@lnu.se
Abstract “Cyber-Physical Systems” are co-engineered interacting entities
made of physical and computational components. Among cyber-physical sytems,
complex and “smart” systems increasingly provide the foundation of national
critical infrastructures, form the basis of emerging and future services, and improve
our quality of life in many areas. The concept of “smartness” has been often used
as a marketing catchphrase. This chapter explains smartness as a serious indicator
which can help to describe the machine intelligence level of different devices,
systems or networks also weighted by usability indices. The study included in this
chapter aims to describe the implementation of complex, resilient and smart system
on the level of devices, systems and complex system networks. The research
considers a smart device as a single agent, the system as a multi-agent system, and
the network of complex systems has been envisaged as an ad hoc multi-agent system
organised in a network. The physical incarnations of the latter could be, for
example, the subsystems constituting a smart city. In order to determine the
smartness of a certain system, the Machine Intelligence Quotient (MIQ), Usability
Index (UI) and Usability Index of Machine (UIoM), Environmental Performance
Index of Machine (EPIoM) indexes will be considered. The quality of human life is
directly influenced by the intelligence of smart machines. Finally, smartness also
play an essential in enabling the overall resilience of cyber-physical systems
through paradigms like pro-active dependability and self-healing.
2
1. Introduction: From Automated Machine to Autonomous
Smart Machine
There is a global demand for maintaining or improving the quality of life of citizens
and increased efforts are being made at all levels in order to fulfil such a demand.
Some regions are becoming overpopulated while others suffer from depopulation,
and, despite contrary efforts, there is a growing inequality in the distribution of
resources. Technological development and automation have had a complex effect
on the whole mankind. Automation has a significant importance for the society, as
it can help to reduce human factors in different processes to such an extent which
ensures greater resilience, usability and cost-efficiency in various sectors of both
industrial production and everyday life. Automation is not simply a technical
question; it is also an economic and social issue which concerns the whole society.
A fully-automated society saves humans a considerable amount of time which can
be used for scientific and cultural purposes or other useful activities. The primary
aim of this chapter is to discuss the present and future role of automation in human
life. It will highlight how to reach early automation systems to smart autonomous
cyber-physical systems.
What are the technical and economic advantages of automation? Improved living
standards, workforce efficiency, energy saving, cost reduction, better working
conditions, health protection, quality improvement, reduced amount of rejects,
increased operational safety and reliability, etc.
How long has automation been present and what does it mean exactly? The roots
of automation can be found in the word “automata”. This word originates from the
Greek language and means self-operating or self-moving. No one knows exactly
who made the first “automata” and when it was made. It is probable that the first
structure of this type was built in ancient times as a result of much-feared sorcery
and parallel scientific developments. In this context, medieval legendary mentions
the name of Albertus Magnus (later: Saint Albert the Great). As a major scientist of
his age, he was referred to as “Magnus in magia, major in philosophia, maxmus in
theologia ” or ”Doctor Universalis”. Legend says that he was working for many
decades on the construction of a mechanic servant. The fate of the construction,
however, was doomed when one of his apprentices, who was received by the
“servant” and heard it say to wait for the master, claimed that the machine must be
the devil’s invention. Other legends have a different ending. According to these
records Magnus’ apprentice, Thomas Aquino (later Saint Thomas Aguinas)
destroyed the work of his master. It is also interesting to mention that Albertus
Magnus is, among others, the patron saint of scientists. [12]
The path leading to full automation or smart autonomous systems has been lined
with many milestones in the course of time. Such milestones were the first steam
engines and centrifugal governors, then the emergence of controllable electric
machines and the appearance of computers, whose future role in automation is
difficult to predict as yet. In other words, these human-made devices have been
aimed to tame first the power of steam, then electricity and finally information.
3
The use of machines and the application of automation eliminate inappropriate or
inadequate human factors in the processes where their use has been justified. Due
to the fact that machines lack self-awareness, they can perform the tasks allocated
to them at their best knowledge, the fastest and most efficient way. They will not be
affected by ignorance, tiredness or any other human or physiological needs. As long
as the conditions necessary for their functioning are ensured, they will accomplish
their tasks. Full automation can be achieved first by partial and then by complex
mechanisation.
From the viewpoint of automation, different levels can be distinguished. In case
of partial automation, the system cannot function without human presence in the
measuring, supervising and controlling roles. Without human contribution they are
unable to operate. In case of complex automation, these roles are completely
overtaken from humans by the automated systems. With fully self-operating
systems only the tasks of supervision, development, checking, repair and
maintenance will be performed by humans. [12]
In the 1970s, the importance of automation was recognised from the point of
industrial use in the increase of productivity. Today its significance has far exceeded
its industrial role, as automation has been introduced in all fields of life. For
example, by the development of nanorobots [13] and their use in the medical
treatment of animals and humans, automation provides great opportunities for the
society. Not to mention another exciting area of automation represented by robots
(e.g. UAVs [14] or industrial robots [15]) or human-like robots, in other words,
androids.
From the point of humanity, the significance of automation can be compared to
the significance of the industrial revolution, or rather, mechanisation and
automation can be regarded as the stages of development that finally determined the
industrial revolution. [12]
Wolfgang Wahlster, Director of the German Research Centre for Artificial
Intelligence, already talked about Industry 4.0 in a presentation in 2013. According
to his view we live in the age of the 4th industrial revolution that in terms of
industrial automation, means the implementation of cyber-physical systems. [16]
Cyber-physical systems will play an increasingly important role in automation,
but not only in industrial automation. This new industrial revolution will affect
many other fields of science.
Our life has been automated to such level which, for example, allows for
autonomously operating vehicles to get from point A to point B safely (e.g. sensor-
based intelligent mobile robot navigation [17]). Of course, these systems also
require specific conditions and have their own limitations, but they are able to
operate in an increasingly automatic way to perform a growing number of tasks.
[18]
But what is the exactly a smart machine? “A smart machine is a device embedded
with machine-to-machine (M2M) and/or cognitive computing technologies such as
artificial intelligence (AI), machine learning or deep learning, all of which it uses to
reason, problem-solve, make decisions and even, ultimately, take action.”[19]
4
2. Usability, Environmental Performance, Smartness and
Resilience
What is Smart Machine design? At the design stage of the system, it is necessary to
quantify the system parameters and check the main system features as early as
possible. In general, it can be said that in order to develop a practicable, flexible and
resilient system, the aspects of scalability, modularity and extensions must be taken
into consideration. There is a growing tendency of considering such subjective
parameters as elegance or the comfort level ensured by the system. Ideally at the
planning phase the possibilities of creating a future-proof design which allows the
system to be used for future purposes should also be considered. In the course of
such planning it is important to bear in mind the rapidly changing technological
developments as well as the compliance with various international standards. With
regard to economy, energy efficiency and other requirements for such systems, it is
necessary to identify the bottlenecks hidden in the system. A system will always
have some bottlenecks, either because of economic conditions or the limited
availability of resources. If such bottlenecks are planned, however, the optimality
of the system can be ensured and a harmonious system can be developed. [20] (see
Fig. 1. 5C architecture for implementation and design [21])
Fig. 1 5C architecture for design of Smart Cyber-Physical Systems [21] [22]
What is resilience? According to the Cambridge English Dictionary, the resilience
is the quality of being able to return quickly to a previous good condition after
problems. How can smartness support future-proof resilience? Smartness is the road
to future-proof resilience since it enables novel paradigms like pro-active
dependability and self-healing. Those paradigms are completely different from the
current implementations of safety-critical and dependable systems where threats,
vulnerabilities and consequences are supposed to be known in advance during risk
assessment or somehow updated and uploaded later (e.g. threat signatures/patterns,
vulnerability information, troubleshooting instructions and repair workflow, etc.).
5
Resilience in future smart-sytems will increasingly leverage on embedded
intelligence in order to anticipate and detect unknown threats and automatically
compute the most appropriate and safe solutions by using approaches based on
machine learning, heuristics, fuzzy logic, bayesian inference and artificial neural
networks driving real-time co-simulation and online model-checking.
Smart machines represent a new field which is still to be thoroughly researched.
Recent research shows that the creation of smart machines is bringing to a higher
level of automation in many fields and that trend will continue in the future. We
mentioned that automation can be a tool to improve resilience. In terms of
functionalities, this automation level is close to robotisation, although there are
some differences. To make things more complex, the usage of smart systems in
critical infrastructures is actually generating the new paradigm of “Smart Critical
Infrastructures”, whose resilience eveluation becomes even more challenging. “The
resilience of an infrastructure is the ability to anticipate possible adverse
scenarios/events (including the new/emerging ones) representing threats and
leading to possible disruptions in operations/functionality of the infrastructure,
prepare for them, withstand/absorb their impacts, recover from disruptions caused
by them and adopt ti the changing conditions.” [23] There are five main resilience
engineering phases in protecting smart critical infrastructures: understand risk,
anticipate or prepare, absorb/withstand, respond/recover, adapt/learn. [24]
Adaptation/learning must leverage on machine intelligence.
Machine intelligence was first mentioned probably in the 1940’s. With the origin
of cybernetics, the possibilities to create thinking machines and robots were
reconsidered. The creation of smart machines is based on the development of
machine intelligence. Among other trends, connectionism has also had a great
influence on the development of smart machines. [25]
How could it be possible to create machines which would solve the problems of
humanity? The planning and development of smart machines are often limited by
human abilities. As it is the case, for example, with the way of thinking software
developers acquire at school. In fact, today software development is based on the
well-established methods learnt at school according to widely accepted
conventional paradigms. Different paradigms are not combined until software
developers form their own style. One of the primary aims of software development
is to make these programs error-free. It is obvious that errors cannot be completely
eliminated; still software developers will always aim to find new improved
solutions for the given tasks. Would all this guarantee problem-free operation? Even
if it was possible to develop a perfect program, one question would still remain to
answer. How could such a rigid system fit into a world that is continuously
changing? How human or natural processes adapt to new or unknown situations?
Could this be implemented in the system of machines? If humans meet new
problems while performing their tasks, they will try to find new solutions to solve
them. Today’s machines are rarely capable of doing something similar. It is enough
to consider the example of autonomous self-driving vehicles, which represent one
of the most advanced and discussed modern technologies. If a self-driving vehicle
6
equipped with state-of-the-art artificial vision approaches a lorry, the back of which
is covered with a picture of a group of cyclists, the vehicle will interpret this image
as a different traffic situation. For humans, it is not a problem to recognise the
situation, as they are able to perceive more details and make various decisions based
on formerly learned patterns, as they will know which pattern would work in this
situation. Similar considerations make today’s most advanced smart-systems still
seem rather “stupid” when compared with basic human intelligence, adaptation and
problem-solving capabilities.
2.1 Usability Index and the Usability Index of Machine - UIoM
“What is Usability?
The word usability means that the people that use a product can do so quickly
and easily to accomplish their own tasks. This definition comes up from four points,
which are essential for understanding what usability really are.
1. Focusing on the user
2. People use products to be productive
3. Users are busy people trying to accomplish tasks
4. The user decides when a product is easy to use
This means that to develop a usable product it’s important that the developer
know, or understand, how the potential user works. No one can fully replace the
potential user, hence the first point.” [9]
In order to determine the Usability Index, it is necessary to consider aspects of
multiple research fields at the same time. One of these fields is ergonomics, which
aims to understand human capacities and abilities. The other field is Human-
Computer Interaction, which deals with the planning, implementation and
evaluation of computer systems used by humans. The third aspect is User
Experience, which refers to the expectations and the real experience humans have
while using a product, a device or a machine from the date of purchase to their
disposal. This will include the shopping experience, the helpdesk support provided
during the use of the product, the convenience, luxury and the feeling that is
associated with the product. The fourth field is User-centred design, a planning
process which integrates the user’s expectations and demands. In many cases this
also includes the involvement of the user in the planning process. The final aspect
to be considered is Usability, which measures the effectiveness, efficiency and
satisfaction the user experiences while performing a specific task with the help of
the product in a given environment. [26][27]
The criteria referring to the general usability design of machines (see Fig. 2
Usability Engineering process) can be determined on the basis of the information
provided by a much researched field. This is the field of the usability design of
medical devices. The process of the usability design of machines consist of ten
steps. [26][27]
7
Fig. 2 Usability Engineering process (© IEC 62366:2007, Fig. D.1)[27]
Step 1 (User research): the Application specification contains the concept of
usage, the users and operators, the connections of the device to other devices, the
conditions of its use and the basic principles of operation. Such a specification must
be based on a user survey and market research carried out in advance. The
developers’ team must have a coherent vision of the device to be developed and its
long-term purposes in order to understand the basic requirements for the developed
product. [26][27]
Step 2 (Conceptual design): the Frequently used functions are specified in the
course of the theoretical design, in order to define the most frequently used and most
important functions of the device. [26][27]
Step 3 (Conceptual design): the conceptual design deals with the Identification
of hazards and hazardous situations related to usability. User activities and the use
of the device – with the exception of fully autonomous devices – can generate errors.
The prior identification and treatment of errors is done as part of the risk-
management procedure which also affects this phase of the usability process.
Therefore, it is necessary to define the criteria for the safe use of the device and to
identify the hazards which could be detected in advance. [26][27]
8
Step 4 (Requirement): furthermore, the Primary operating functions and
requirements must be determined. Among the frequently used functions, safety
critical functions are those functions which are the most critical from the point of
hazardousness.
Step 5 (Requirement): the second phase of the elaboration of requirements is
Usability specification, which summarises the information collected about the
device in the previous phases, with special attention made to primary operational
functions. This is the basic document of the certification and validation of usability.
[26][27]
Step 6 (Requirement): the third phase of the elaboration of requirements is the
development of the Usability validation plan. The Usability validation plan must be
prepared prior to the validation procedure, as this document specifies the validation
methods, validation criteria and the examinations carried out with the involvement
of representative users. [26][27]
Step 7 (Detailed design and specification): in this phase a detailed design must
be provided which includes the User interface design and implementation. It should
also include software development, the making of prototypes and the simultaneous
evaluation of usability. [26][27]
Step 8 (Usability verification): The evaluation of the created device starts with
the Usability verification. This means the comparison of the device with its
specifications and purposes. It ensures the product’s compliance with its
requirements. [26][27]
Step 9 (Usability validation): Usability validation is the second phase of the
evaluation, in which the compliance with user requirements is analysed according
to the validation plan. The methods used in the usability evaluation process include
analytic, empiric, formative or summative methods, which help to reveal any
usability issues. [26][27]
Step 10 (Monitoring): Usability issues can prevent the device from performing
its task or fulfilling its purposes. It can cause uncertainty in the users, who might
make a mistake as a consequence (e.g. they might fail to notice something, make
incorrect assumptions, perform inappropriate actions or misinterpret some
information). The surveillance and monitoring of the device can be done after it has
been introduced into the market. Any feedback from the users may help to correct
the errors. [26][27]
The applied main components are listed in Table 1 with reference to the Usability
Index of Machine (UIoM). From the discussion in this section, it should be clear
that usability is strictly related to the interference of human-factors with attributes
like performability and resilience, since e.g. bad design of Human-Machine
interfaces can facilitate human faults. Human faults can generate errors and then
failures when those errors activate. Human mistakes and response delays can also
significanlty impact on the threat/crisis management workflow in business
continuity, disaster recovery and reaction to emergencies. Generally speaking, since
human actions are error-prone, wherever total automation cannot be achieved yet
9
and human intervention is still required, those aspects are of paramount importance
in cyber-physical systems design.
2.2 Environmental Performance Index of Machines - EPIoM
Environmental Performance Index of Machines was developed as a result of the
co-operation of Yale University, Columbia University, the World Economic Forum
and the European Commission and it was first published in 2002. [10] This study
refers to the EPI version which is related to machines. In case of machines it is
especially important to reduce the use of non-renewable resources and to benefit
from renewable energy. They must be designed in a way that considers the
optimisation of the use of raw materials and the recycling of future (electronic)
waste, in order to ensure the environmentally conscious use of the device in its
whole lifecycle. [28] [29] The applied main components are listed in Table 1 with
reference to EPIoM.
3 The Implementation of Complex, Resilient and Smart Systems
3.1 Smart Device – Smart Agent
What is a smart agent used for? And what makes it smart? A smart agent (see Fig.
3a) has artificial intelligence, and it stores the information necessary for its
operation in ontologies by means of knowledge representation and semantic
models. It can also gain knowledge by learning. Supported by this knowledge, it
can recognize and even change the surrounding environment. It is capable of
independent operation, but it can also work as part of a system with cooperation and
self-organization capabilities [30][3][31][32].
Fig. 3 (a) A proposed operational architecture for Holonic Smart Agent [35]
[30]; (b) General CPS architecture [33] [34]
10
Smart systems are based on some forms of machine intelligence; therefore, in order
to create smart machines, it is necessary to know how depelop machine intelligence.
Wechsler defined the concept of intelligence as the following: “Intelligence is the
aggregate or global capacity of the individual to act purposefully, to think rationally
and to deal effectively with his environment” [42]. This definition has been given
with regards to human intelligence. However, it is also relevant for machines, as the
basis of machine intelligence also includes environment detection, decision making
and intervention control. At a higher level, intelligent machines are capable of
recognising objects and events, representing knowledge or determining their future
operation. In order to ensure intelligent operation, smart machines also need to be
capable of learning. This learning feature requires data and the information
generated from this data, which are collected by detection. A smart system will not
only use the data of its own detection, but it will also leverage on co-operation as a
multi-component system. Therefore, a device or machine which is capable of co-
operation can be defined as a smart device. In other words, different levels of
smartness can be distinguished even in case of machines. As an illustration, Figure
3 shows the architecture of a smart agent. Regarding resilience related aspects, the
aforementioned paradigm of pro-active dependability also refers to smart agents
that warn other cooperative agents about possible dangers, exactly like humans
working together warn each other when they realize other peers are in danger.
3.2 Smart Systems – Smart Multi-Agent Systems
Why should individual agents be connected? How could they be connected? There
are several reasons for this: autonomy versus team work, strength in unity, self-
defence or error-tolerant group coherence.
According to Csermely [36], humans are communal beings, and as a result, our
brain has developed in a way that it can list, revise and activate our relationships.
The network of relationships is the key to human survival. Social-psychological
surveys show that humans divide the world into groups of 5, 15, 35, 80 and 150
individuals, and these groups represent our family, close friends, further
acquaintances, and other smaller groups. We are closed into our cognitive space
determined by our cognitive characteristics, and it is difficult for us to think and
work in a bigger perspective. We can be successful and we can communicate and
exist in this small world restricted by our cognitive characteristics. Random
networks, where relationships between the elements have been made by chance, and
these relationships are easy to make, also have these small-world characteristics. It
is easy to make relationships, because neighbours know each other. In these
networks, clustering is fast and frequently occurring. [36]
The smart operation of machines is essentially in the realisation of the natural
analogue. That is the reason for working with self-organising elements. It is easy to
11
see that one agent is not enough to perform a complicated task. Therefore, it is
necessary to build a smart multi-agent system of several co-operating agents.
Intelligence depends on the learning capacity of machine systems, which is based
on the finding, using, processing, connecting and fusioning of data, or, in other
words, on generating information. Knowledge Discovery and Data Mining is a
research field which aims to create knowledge from this immense amount of data.
This can be automated by using smart agents. With the help of these co-operative
systems, which are capable of learning, a complex adaptive system can be built. In
nature, for example, in case of floods, fire ants are able to link their bodies together
and float on the surface of water saving their own lives and their colony. A swarm
intelligent system can also be made by the co-operative operation of machines, and
this system can be recognised by the interactions between the above-mentioned
smart agents and the agents in their immediate environment or between the agents
and the environment. In this system, cognitive science can help to realise machine
learning. “Complex systems network theory provides techniques for analysing
structure in a system of interacting agents, represented as a network”. (see Fig 4)
[37]
Fig. 4 Smart Multi-Agent System architecture - random temporary cluster
12
3.3. Smart Complex systems network – Ad-hoc networked Smart Multi-Agent
System sociograms
Following the analogy with nature, the example of cells can be mentioned in the
way they work and perform their tasks in the human body. A complex artificial
world can be created by computers and the programs running in their memories, as
today there is a tiny computer in almost every device. In order to perform their tasks,
machines need an executable program. The more precisely we would like to instruct
machines, the more detailed programs are needed, and this can be expensive, time-
consuming and requiring considerable resources. The solution to this problem lies
in the autonomy of machine systems. By creating a network in which smart agents
are able to autonomously co-operate, coordinate and communicate with each other
for a common purpose – a temporarily-organised network to perform specific tasks
- and to make any related decisions and interventions (actions), an ad-hoc networked
Smart Multi-Agent System could be built (see Fig. 5).
The reason for organising such networks or clusters can be, for example, that in
case of single smart agents, the embedded functions are not always sufficient to
perform the given tasks. Furthermore, they can only address the challenges of
today’s fast-changing world promptly and find the solution to a special problem if
they work together. Regarding resilience, that also holds when smart-agents need to
counteract threats or restore from failures requiring a joint effort of multiple
cooperating agents.
Fig. 5 Ad-hoc networked Smart Multi-Agent System sociograms
3. From Machine Intelligence Theory to Smart Machine Theory
According to Bein et al. [38] Machine Intelligence Quotient (MIQ) consists of four
key attributes, such as Autonomy, Human–Machine Interaction, Controllability for
Complicated Dynamics and Bio-inspired Behavior. The Machine Intelligence
Quotient is a measure to assess the intelligence of a Machine. Each attribute has a
number of major components. In case of Autonomy, these components are self-
calibration, self-diagnostics, self-tuning and fault-tolerance. Human–Machine
13
Interaction includes human-like understanding communication, emergence of
artificial emotion and ergonomic design, while Controllability for Complicated
Dynamics covers non-conventional, model-based, adaptation, motion planning and
non-linearity. Bio-inspired Behavior involves Neuro-science, Biologically
motivated behavior, Cognitive-based. According to the theory, in the model space
Autonomy and Human–Machine Interaction are constant factors, while
Controllability for Complicated Dynamics and Bio-inspired Behavior are
application specific. It is important to define the conditions of use for the model
environment, which is done according to the original theory of dynamic,
unstructured, and uncertain characteristics of Environment. It suggests three
methods to determine MIQ: Fuzzy Logic, Neural Networks and Genetic Algorithms
(e.g. neuro-fuzzy-genetic controller for robot manipulators [39]). According to Bein
et al., any intelligent system with those features can lead to improved safety,
enhanced reliability, higher efficiency and sustainability. [38]
4.1 Smart Cyberspace Theory – the Intelligent Cyberspace and the Smart
Cyberspace
Figure 6 shows those spaces which are created by human activities and thinking. In
the centre of the space, the human can be found, as an individual who is physically
present only in real space, but whose mind creates the virtual space, and can extend
his physical activities into this virtual space. Physical space represents the
geosphere of the Earth. Biosphere is only a part of this space. This is the biological
space where the conditions of life are ensured for humans and other living creatures
on Earth. Anthropogenic space refers to the world created by humans. Cities,
infrastructures and man-made facilities can be found in this space. Individual space
makes only a small part of anthropogenic space use by the individual. This is where
the individual lives and moves, and where, by his personal activities, he creates the
smallest space surrounding him. With regards to humans, the above-mentioned
spaces depend on the development level of the society, or, among others, the age,
the income, etc. of the individual. For example, the individual space is much smaller
for a child than for a professor at the top of his career or for a constantly travelling
businessman who is able to influence a more extended space. [40]
Virtual spaces are built from the real space and they are shaped by the thoughts
of the individual (cognitive space). This cognitive space includes the individual’s
view of the world, which is the mind’s mapping of the physical world in a subjective
way. Cyberspace is the artificial world of man-made devices, systems and networks
beyond the physical space. A typical example of this is the cyber world of the
Internet. Cyberspace is the mapping of the physical space, for example, in case of
the Internet, the servers, optical cables, routers and nodes of which the real physical
infrastructure is made. Cyberspace inherits the flexibility of the cognitive space,
therefore its construction, transformation or use may only be limited by human
imagination. Fictional space stands the furthest from reality in comparison with
14
other virtual spaces. At the same time, fictional space can also include real elements.
Quite obviously, these imaginary spaces are not in the scope of this study. A smart
agent is capable of the interactions shown in Figure 6. [40]
From the viepoint of this study, there is a growing number of applications where
the two MIQ factors (Controllability for Complicated Dynamics, and Bio-inspired
Behavior) have equal importance. Therefore, we created an improved version of the
above model, and we can state that there are certain cases when these two factors
co-exist in space and time.
The geometry of more than three space dimensions can be mathematically
described. The fourth space dimension is perpendicular to the other three (X, Y, Z)
dimensions, which are commonly used in everyday life. The space defined by four
dimensions (X, Y, Z, W) is called a four-dimensional space. The fourth dimentions
is represented by the W axis. This concept can be difficult to demonstrate in three
dimensions, but Figure 7 shows a generally accepted illustration of a four-
dimensional space. The Intelligent Cyberspace can be imagined with the help of the
four-dimensional space theory. This space, which is illustrated by Figure 7 as a
hypercube, contains the Machine Intelligence Index (marked in red in Fig. 7)
defined by the four attributes/dimensions which characterise Machines. For each
device, this point can be found in a different part of space depending on the values
of the attributes. The major components of the main attributes have already been
defined, but after 15 years we felt it was necessary to redefine them. Table 1
contains these revised definitions.
Fig.6 A Model of Human Perception versus Human-like Machine Perception
[35] [40]
15
Fig.7 Forth dimensions as the Intelligent Cyberspace, visualization of Machine
Intelligence Index [own]
16
Table 1 The origin of Smartness Quotient, the revised MIQ [own]
17
Fig. 8 Hexeract [41] (six-dimensional hypercube) - Petrie polygon
Orthographic projections – The representation of the sixth dimension as the Smart
Cyberspace
It is quite difficult to illustrate a four-dimensional space in two dimensions. In
case of a six-dimensional space, however, it is almost impossible. Figure 8 shows
one way of representing a six-dimensional space in two dimensions. The Smartness
Quotient will be defined in this space.
The Smartness Quotient can be illustrated with the help of the Hexeract. The six
dimensions are the following: Dim. 1: Autonomy, Dim. 2: Human–Machine
Interaction, Dim. 3: Controllability For Complicated Dynamics, Dim. 4: Bio-
Inspired Behavior, Dim. 5: Usability Index Of Machine, Dim. 6: Environmental
Performance Index of Machine. The values of these dimensions determine a point
in the special six-dimensional Smart Cyberspace. The six-dimensional hypercube
represents the space in which the Smartness Quotient is a point, like in the case of
the MIQ.
Why should spaces be discussed in such detail? The world and its higher
dimensions have always interested humans, and it could not be possible to imagine
the existence of smart devices if we did not know how they can become smart. The
above described theory can show us how human mind can create from smart
machines and devices a space that is no longer cognitive or fictive, but a real Smart
Cyberspace.
4. Smartness Theory, Smartness Quotient
What does Smartness Quotient mean exactly, what is it used for and what does it
describe? It is the measure of the success factor of the people-centred planning,
implementation, use, operation, disassembling and material recycling of machines
in relation to environmental effects and from the point of user experience. Its use is
18
based on the view of holistic systems, from the planning, through operation to
disassembling and material recycling of the machines. It shows how efficiently a
machine can work as an artificial form of life in interaction with humans, the
environment and with other machines. With the help of Smartness Quotient, it is
possible to compare machines in an objectively quantitative way, in terms of how
they are capable of ensuring a safe, sustainable, efficient and convenient life for
humans. It can also help to determine the direction of machine development and to
define Machine Service Quality on the basis of empiric research. In case of
traditional MIQ, the operation of “simple” machines can be examined in a four-
dimensional space with a functionality-based approach, while SQ ensures a
knowledge and ability-based examination in a six-dimensional space, which
supports the development of machines with artificial intelligence with the user in
focus.
5. Conclusion
This chapter has presented the basics of developing smart machines in a micro and
macro perspective, from machine intelligence, through the agents capable of
machine learning, to the theory of networks. Such a view is based on a multi-
dimensional space including MIQ and SQ attributes. We have pointed out the
difficulties of the path leading from machines to smart machines, and discussed the
previously defined and designed characteristics and major components which turn
a machine into a an artificial form of life, in the way Albertus Magnus had imagined.
One of the further objectives of research is to make the SQ value of machines
calculable by quantifying the main components in order to measure and categorise
the beneficiary effects to the society. Besides human failure analysis and usability
planning, further research will need to to deal with hidden errors in technical
systems and automatic corrections by smart-troubleshooting and self-healing. A
further refinement of the major components of EPIoM is possible by defining the
parameters strongly related to smart machines for the more general components.
The development of smart systems is part of the endeavour to ensure human
development, survival and well-being. Machine intelligence and smart systems are
transforming digital societies. In the Internet of Things era, smart-agents embedded
in cyber-physical systems enable new paradigms for Intelligent Transport Systems,
Smart Cities and Smart Factories, where the complexity of infrastructural networks
and its components is growing exponentially. The use of machine intelligence will
also result in increased resilience (including reliability, safety, security,
maintainability, self-healing, etc.), efficiency and sustainability in future generation
cyber-physical systems.
19
REFERENCES
[1] National Institute of Standards and Technology, “Cyber-Physical Systems,” 2017. [Online].
Available: https://www.nist.gov/el/cyber-physical-systems. [Accessed: 31-Dec-2017].
[2] A. M. Farid, “Designing multi-agent systems for resilient engineering systems,” in Lecture
Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and
Lecture Notes in Bioinformatics), 2015, vol. 9266, pp. 3–8.
[3] L. B. Iantovics, A. Gligor, and V. Georgieva, “Detecting Outlier Intelligence in the behavior
of intelligent coalitions of agents,” in 2017 IEEE Congress on Evolutionary Computation
(CEC), 2017, pp. 241–248.
[4] H.-J. Park, B. K. Kim, and K. Y. Lim, “Measuring the machine intelligence quotient (MIQ) of
human-machine cooperative systems,” IEEE Trans. Syst. Man, Cybern. - Part A Syst. Humans,
vol. 31, no. 2, pp. 89–96, 2001.
[5] H.-J. Park, B. K. Kim, and G.-Y. Lim, “Measuring machine intelligence for human-machine
cooperative systems using intelligence task graph,” in Proceedings 1999 IEEE/RSJ
International Conference on Intelligent Robots and Systems. Human and Environment
Friendly Robots with High Intelligence and Emotional Quotients (Cat. No.99CH36289), 1999,
vol. 2, pp. 689–694 vol.2.
[6] T. Ozkul, “Cost-benefit analyses of man-machine cooperative systems by assesment of
Machine Intelligence Quotient (MIQ) gain,” in 2009 6th International Symposium on
Mechatronics and its Applications, 2009, pp. 1–6.
[7] C. Li, Z. Ji, Z. Pang, S. Chu, Y. Jin, J. Tong, H. Xu, and Y. Chen, “On Usability Evaluation of
Human--Machine Interactive Interface Based on Eye Movement,” in Man-Machine-
Environment System Engineering: Proceedings of the 16th International Conference on
MMESE, S. Long and B. S. Dhillon, Eds. Singapore: Springer Singapore, 2016, pp. 347–354.
[8] G. Szabó, “Usability of Machinery,” in Advances in Safety Management and Human Factors:
Proceedings of the AHFE 2017 International Conference on Safety Management and Human
Factors, July 17--21, 2017, The Westin Bonaventure Hotel, Los Angeles, California, USA, P.
Arezes, Ed. Cham: Springer International Publishing, 2018, pp. 161–168.
[9] N. Aykin, Ed., Usability and internationalization of information technology. New Jersey:
Lawrence Erlbaum Associates, Inc., Publishers, 2005.
[10] A. et al. Hsu, Global Metrics For The Environment, The Environmental Performance Index
ranks countries ’ performance on high-priority environmental issues. New Haven: Yale
University, 2016.
[11] Z. Liouane, T. Lemlouma, P. Roose, F. Weis, Z. Liouane, T. Lemlouma, P. Roose, F. Weis,
and H. M. A. G. Neu-, “A Genetic Neural Network Approach for Unusual Behavior Prediction
in Smart Home,” in Advances in Intelligent Systems and Computing, A. M. Madureira, A.
Abraham, D. Gamboa, and P. Novais, Eds. Porto: Springer International Publishing AG 2017,
2016, pp. 738–748.
[12] J. Hollós, Az ipari elektronika alapfogalmai, 3. Budapest: Műszaki Könyvkiadó, 1979.
[13] A. Katsnelson, “DNA robot could kill cancer cells,” Nature, Feb. 2012.
[14] J. Cosic, P. Curkovic, J. Kasac, and J. Stepanic, “Interpreting Development of Unmanned
Aerial Vehicles using Systems Thinking,” Interdiscip. Descr. Complex Syst., vol. 11, no. 1,
pp. 143–152, 2013.
[15] G. Mester, S. Plet, G. Pajor, and Z. Jeges, “Flexible planetary gear drives in robotics,” in
Proceedings of the 1992 International Conference on Industrial Electronics, Control,
Instrumentation, and Automation, 1992, pp. 646–649 vol.2.
[16] W. Wahlster, “The Semantic Product Memory: An Interactive Black Box for Smart Objects,”
Springer, Berlin, Heidelberg, 2013, pp. 3–21.
[17] G. Mester and R. Aleksandar, “Sensor-Based Intelligent Mobile Robot Navigation in
Unknown Environments,” Int. J. Electr. Comput. Eng. Syst., vol. 1, no. 2, pp. 1–8, 2010.
[18] D. Tokody, “Digitising the European industry - holonic systems approach,” Procedia Manuf.,
vol. 22, pp. 1015–1022, 2018.
[19] M. Rouse, “Smart machines.” [Online]. Available:
https://searchcio.techtarget.com/definition/smart-machines. [Accessed: 28-Apr-2018].
[20] J. Crowcroft, “System Design: An Engineering Approach to Computer Networking.”
Cambridge University, 2011.
20
[21] J. Lee, B. Bagheri, and H. A. Kao, “A Cyber-Physical Systems architecture for Industry 4.0-
based manufacturing systems,” Manuf. Lett., vol. 3, pp. 18–23, 2015.
[22] D. Tokody and F. Flammini, “The intelligent railway system theory,” Int. Transp. is a Spec.
Ed. Int. Verkehrswesen, ISSN 0020-9511, vol. 69, no. 1, pp. 38–40, 2017.
[23] K. Øien, A. Jovanović, T. O. Grøtan, A. Choudhary, A. Øren, K. Tetlak, L. Bodsberg, and M.
Jelic, “Assessing resilience of Smart Critical Infrastructures SmartResilience based on
indicators.” Stuttgart, p. 87, 2017.
[24] A. Jovanovi, K. Ø, and A. Choudhary, “An Indicator-Based Approach to Assessing Resilience
of Smart Critical Infrastructures,” in Urban Disaster Resilience and Security, A. Fekete and F.
Fiedrich, Eds. Springer International Publishing, 2018, pp. 285–311.
[25] R. Schank, “Chapter 8 ‘SMART MACHINES.’” [Online]. Available:
https://www.edge.org/conversation/marvin_minsky-chapter-8-smart-machines. [Accessed:
21-Aug-2017].
[26] Z. Béky, “Az orvosi eszközök használhatósági tervezése (Usability Engineering).” B. Braun
Medical Kft., 2013.
[27] ElectrotechnicalCommissionInternational, “IEC 62366:2007, Medical devices — Application
of usability engineering to medical devices.” International Electrotechnical Commission,
2007.
[28] Z. Wang, B. Zhang, and D. Guan, “Take responsibility for electronic-waste disposal,” Nature,
vol. 536, no. 7614, pp. 23–25, Aug. 2016.
[29] D. Necula, N. Vasile, and M. F. Stan, “The impact of the electrical machines on the
environment,” in 2013 8th International Symposium on Advanced Topics in Electrical
Engineering (ATEE), 2013, pp. 1–4.
[30] D. Tokody and F. Flammini, “Smart Systems for the Protection of Individuals,” Key Eng.
Mater., vol. 755, p. pp 190-197, 2017.
[31] L. B. Iantovics, C. Rotar, and E. Nechita, “A Novel Robust Metric for Comparing the
Intelligence of Two Cooperative Multiagent Systems,” in Procedia Computer Science, 2016,
vol. 96, pp. 637–644.
[32] B. L. Iantovics and B. Crainicu, “Complex Mobile Multiagent Systems,” in 2008 First
International Conference on Complexity and Intelligence of the Artificial and Natural
Complex Systems. Medical Applications of the Complex Systems. Biomedical Computing,
2008, pp. 21–30.
[33] A. Esfahani, G. Mantas, D. Yang, A. Nascimento, J. Rodriguez, and J. Neves, “Towards Secure
Network Coding–Enabled Wireless Sensor Networks in Cyber-Physical Systems,” in Cyber-
Physical Systems, no. October, CRC Press, 2015, pp. 395–414.
[34] A. Zanni, “Cyber-physical systems and smart cities Learn how smart devices , sensors , and
actuators are advancing Internet of Things implementations,” IBM developerWorks, no. April.
pp. 1–8, 2015.
[35] R. Velik, “A Bionic Model for Human-like Machine Perception,” Vienna University of
Technology, 2008.
[36] P. Csermely, “Why Do We Like Networks?,” in Weak Links: The Universal Key to the Stability
of Networks and Complex Systems, Berlin, Heidelberg: Springer Berlin Heidelberg, 2009, pp.
5–52.
[37] J. Crowcroft, “Network/Graph Network/Graph Theory Theory.” University of Cambridge,
2011.
[38] Z. Bien, W.-C. Bang, D.-Y. Kim, and J.-S. Han, “Machine intelligence quotient: its
measurements and applications,” Fuzzy Sets Syst., vol. 127, no. 1, pp. 3–16, 2002.
[39] G. Mester, “Neuro-fuzzy-genetic controller design for robot manipulators,” in Industrial
Electronics, Control, and Instrumentation, 1995., Proceedings of the 1995 IEEE IECON 21st
International Conference on, 1995, vol. 1, pp. 87–92.
[40] G. Pirisi and A. Trócsányi, Általános társadalom - és gazdaságföldrajz. Pécs: Pécsi
Tudományegyetem, 2011.
[41] “6-cube.” [Online]. Available: https://en.wikipedia.org/wiki/6-cube. [Accessed: 21-Aug-
2017].
[42] D. Wechsler, Non-intellective factors in general intelligence. Psychological Bulletin, 37, 1940,
pp. 444-445
