MENTORS: Monitoring Environment for System of Systems
Antonello Calabr
`
o
a
, Said Daoudagh
b
and Eda Marchetti
c
ISTI-CNR, Pisa, Italy
Keywords:
Ontology, Run-time Monitoring, Vulnerability Detection.
Abstract:
Context: Systems Of Systems (SoSs) are becoming a widespread emerging architecture, and they are used in
several daily life contexts. Therefore, when a new device is integrated into an existing SoS, facilities able to
efficaciously assess and prevent anomalous and dangerous situations are necessary. Objective: The aim is to
define a reference environment conceived for monitoring and assessing the behavior of SoS when a new device
is added. Method: In this paper, we present MENTORS, a monitoring environment for SoS. MENTORS is
based on semantic web technologies to formally represent SoS and Monitoring knowledge through a core
ontology, called MONTOLOGY. Results and Conclusion: We defined the conceptual model of MENTORS,
which is composed of two phases: Off-line and On-line, supported by a reference architecture that allows its
(semi-)automation. Validation of the proposal with real use-cases is part of future activities.
1 INTRODUCTION
Nowadays, Systems of Systems (SoSs) are becoming
an emerging widespread architecture. SoSs are used
in several daily life contexts: from our homes through
to the industrial assembly lines. Often the develop-
ment of SoS requires the integration and collabora-
tion of different ICT components or devices typically
developed by several third parties. If from the one
hand this process assures high productivity and com-
petitiveness, from the one other it exposes the SoS
to important vulnerabilities and risks. A typical con-
sequence is that each of the integrated components
could be affected by, and involuntarily propagate, the
vulnerabilities of the others included in the SoS, with
potential drastically consequences. In this situation,
when a new device or component is integrated into
an existing SoS, facilities able to efficaciously assess
and prevent anomalous and dangerous situations are
therefore necessary. Among them, one commonly
adopted is the use of a monitoring system (Ullo and
Sinha, 2020), i.e., a means for the on-line analysis of
functional and non-functional properties.
Indeed, a monitor engine collects events from dif-
ferent levels from the different system components (or
sensors) and uses these data to infer complex patterns
that could represent a specific functional or non func-
a
https://orcid.org/0000-0001-5502-303X
b
https://orcid.org/0000-0002-3073-6217
c
https://orcid.org/0000-0003-4223-8036
tional property. The derived complex patterns repre-
sent the observed normal (or abnormal) behavior of
the monitored system (or its components). There-
fore, the monitor is able to: i) collect and analyze data
coming from the different SoS sources (e.g., sensors,
components or devices); ii) assess the run time SoS
(components or devices) behaviour; iii) promptly rise
up alarms in case of violations; and, iv) put in place
countermeasures if necessary.
However, even if notably efficient and effective,
the design, implementation, and management of a
monitoring system inside an SoS environment may
involve the participation of different stakeholders
such as SoS domain experts, device developers, or
monitoring experts. Thus, due to the intrinsic com-
plexity of the scenario, monitoring the integration of
new devices (or components) in different SoS envi-
ronments can result in a complex and costly activ-
ity (Rastogi et al., 2020). One of its main critical-
ities is the lack of a common understanding: ex-
perts of SoS domain could ignore the peculiarities
of monitoring activity and vice versa. Thus, this pa-
per wants to move a step ahead in this direction, by
joining the different definitions of the SoS and moni-
toring context into a unique structural representation.
For this, we conceived a smart and efficient moni-
toring processes for SoS, called MENTORS - Mon-
itoring ENviromenT fOR Sos. MENTORS can pro-
vide a support for all the stakeholders involved in the
monitoring process with the purpose of: i) lowering
Calabrò, A., Daoudagh, S. and Marchetti, E.
MENTORS: Monitoring Environment for System of Systems.
DOI: 10.5220/0010658900003058
In Proceedings of the 17th International Conference on Web Information Systems and Technologies (WEBIST 2021), pages 291-298
ISBN: 978-989-758-536-4; ISSN: 2184-3252
Copyright
c
2021 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
291
down costs of developing and setting up the moni-
toring environment, i.e., allowing the use of available
monitor engines (e.g., GLIMPSE (Bertolino et al.,
2011; Barsocchi et al., 2018)) without the necessity
to implement ad hoc solutions; ii) improving qual-
ity control by smart and effective rules specification
and encoding; iii) increasing flexibility and produc-
tivity by automating the monitoring process through
a reference architecture so as to make the moni-
tor designers agnostic of the domain specific chal-
lenges (see for instance (Palumbo, 2018)); and iv) in-
crease the interoperability by using standardized and
domain independent specification technologies (e.g.,
OWL (Motik et al., 2012) for the Ontology descrip-
tion, and RuleML (Boley et al., 2001) or Drools
1
for
instantiating rules).
Outline: Section 2 reports related work, whereas
the conceived monitoring process for the SoS, com-
posed of two main phases (off-line and On-line) is in-
troduced in Section 3. Section 4 illustrates the defini-
tion of standardized and domain independent specifi-
cation of the SoS and monitoring concepts. A prelim-
inary structure of the reference architecture allowing a
(semi-)automation of the proposed process is reported
in Section 5. Finally, Section 6 concludes the paper
and reports future work.
2 RELATED WORK
At state of the practice, different run time solutions
are currently available in many domains: traffic mon-
itoring systems (Won, 2020), indicators on the control
panel of a car (Fotescu et al., 2020), airplane or elec-
tronic device (Hidayanti, 2020), monitors for phys-
iological conditions of patients in a hospital (San-
tos et al., 2020) or systems for controlling complex
large-scale systems such as airports, railways or in-
dustrial plants (Bhamare et al., 2020). Usually, all the
monitoring solutions address two main challenges:
i) finding very powerful, concise and unambiguous
specification languages able to express the properties
to be validated (Khan et al., 2021); ii) defining ef-
ficient mechanisms to assess conformity of the sys-
tem against these properties or contracts (Burns et al.,
2018). In line with the current research, the aim of this
work is to advance the state of practice of the monitor-
ing process, by conceiving an easy to use and effective
solution for the application of the monitoring activ-
ity inside SoS. Indeed, MENTORS can enhances the
existing monitoring approaches for: 1) promptly ris-
1
https://www.drools.org/
ing warnings and detecting failures, and for enabling
system reconfiguration, reliable and trustworthy exe-
cution; 2) capturing, inferring and analysing complex
events, so to allow the detection of critical problems,
failures and security vulnerabilities; and 3) validat-
ing the trust and security level of run-time behavior
of SoS and its devices.
3 CONCEPTUAL MODEL AND
PROCESS
From a conceptual point of view, the monitoring pro-
cess presented in this paper is based on a structural
representation of the concepts and knowledge about
the SoS, devices and monitoring environments.
As schematized in Figure 1, the MENTORS mon-
itoring process is structured into two different phases:
i) Off-line phase, in which the concepts and the
knowledge about the different SoS and the available
devices are organized and integrated in a structural
specification, and monitoring rules defined; ii) On-
line phase, in which the monitoring activities are ex-
ecuted, results collected and inferences about the ob-
served behavior and possible rules improvements de-
fined.
Specifically, as reported in Figure 1, during the
former phase the conceptual steps are:
SoS Specification & Monitoring Specification:
SoSs are widespread adopted in several applica-
tion domain, each having its specific needs and
challenges (Raman and D’Souza, 2019). How-
ever, at the state of the practice there is not
a common knowledge-base for collecting into a
unique environment the SoS and monitoring re-
quirements, specifications, architectural design,
and the set of existing or conceivable monitoring
rules useful for the assessment or quality improve-
ment (Vierhauser et al., 2016). This kind of infor-
mation is still on the hands of a few experts and it
is not easily shareable between the different stake-
holders. As reported Figure 1, MENTORS pro-
vides a solution for this lack, i.e., a continuously
updatable structural representation of the research
and practical knowledge about SoS and the moni-
toring anemometers.
SoS & Monitoring Ontology: At the state of
the practice, one of the recognized means for
knowledge representation and environmental data
modeling is the use of ontologies (Motik et al.,
2012). They provide a common vocabulary use-
ful for modeling and understanding specific do-
mains by capturing knowledge in a structured and
WEBIST 2021 - 17th International Conference on Web Information Systems and Technologies
292
Figure 1: MENTORS: Conceptual Model.
formal way. As in Figure 1, MENTORS orga-
nizes, revises and structures the available knowl-
edge by means of an ontology (Berners-Lee et al.,
2001) called MONTOLOGY - MONitoring on-
TOLOGY (see Section 4).
Abstract Monitoring Rules: In the context of
SoS, and considering in particular the rules def-
inition, even if traditional performance indicators
could be still relevant, the dynamic evolution and
complexity of SoS require the availability of suit-
able rules, able to take into account also events
risen by the SoS context in which the device (or
its components) is included. Consequently, suc-
cessful rules definition requires a deep knowledge
of the SoS domain and its environment as well
as monitoring behavior and management. By ex-
ploiting the MONTOLOGY structure, the pro-
posal of this paper provides both: i) a common
evolving rules definition, able to improving qual-
ity control and increasing flexibility and produc-
tivity of the monitoring process; ii) means for
easily instrumenting the selected set of rules ac-
cording to the selected SoS domain needs and at-
tributes.
Crossing from Off-line phase to On-line phase, the
Abstract Monitoring Rules, generated through
MONTOLOGY, will be instantiated for a specific rule
language understandable by the selected Monitoring
component, and enacted for monitoring the SoS and
New Device during the On-line phase. Therefore, dur-
ing the On-line phase, the conceptual steps are:
SoS & New Device Monitoring: The instantiated
rules are used to monitor the new device behavior in-
side a target SoS. During this activity, possible rule
violations are detected and possibly managed.
Results & Analysis: The monitored data are then an-
alyzed so as to infer new knowledge about the SoS
or Device behavior, or to enrich the MONTOLOGY
contents.
In the remainder of this paper, details about MON-
TOLOGY and the MENTORS reference architecture
will be provided in Section 4 and Section 5.3, respec-
tively. It is out of the scope of the paper discussing the
validation of the proposal, which is part of our future
work.
4 MONTOLOGY: MONITORING
ONTOLOGY
In this section, we introduce our core ontology (called
MONTOLOGY) for supporting the realization of
MENTORS. We firstly present the development pro-
cess in Section 4.1, and then we briefly describe its
main modules in Section 4.2.
4.1 Process for Developing
MONTOLOGY
The process adopted for developing MONTOLOGY
is based on (Gharib et al., 2021), and it follows the
MENTORS: Monitoring Environment for System of Systems
293
Figure 2: Process for Developing MONTOLOGY (adapted from (Gharib et al., 2021)).
five principles proposed in (Gruber, 1995) (i.e., clar-
ity, coherence, extendibility, minimal encoding bias,
and minimal ontological commitment).
As depicted in Figure 2, the process is composed
of five main steps:
1. Scope & Objective Identification (Step 1 ) where
the scope and the purposes of the ontology and
its intended end-users are identified. In our case,
the purpose is to allow any possible stakeholder
to define, for every specific SoS domain, a set of
effective and well-defined monitoring rules able
to reveal the vulnerabilities and or problems that
a new device integration could arise.
2. Knowledge Acquisition (Step 2 ) where the
knowledge needed for the realization of the in-
tended ontology is collected. During this step,
monitoring and SoS experts have been involved
both for the identification of the concepts and their
relationships, as well as for capturing monitor-
ing and SoSs requirements. Analysis of the avail-
able literature concerning SoS and monitoring has
been also performed (de Almeida et al., 2020;
Silva et al., 2015).
3. Conceptualization (Step 3 ) where the core ontol-
ogy (i.e., MONTOLOGY) for supporting moni-
toring process is derived. MONTOLOGY is com-
posed of 13 concepts: 8 coming from SoS do-
main, and 5 coming from monitoring domain (see
Section 4.2 for more details).
4. Implementation (Step 4 ) where the conceptual-
ized ontology is codified into a formal language.
This requires an environment that guarantees (1)
checking lexical and syntactic errors so as to im-
plement an error-free ontology, and (2) an auto-
mated reasoner for detecting inconsistencies and
redundant knowledge. For MONTOLOGY im-
plementation, Prot
´
eg
´
e
2
has been selected be-
cause: it is an open-source environment allowing
creating, modifying, visualizing and checking the
consistency; and it relies on SPARQL
3
(Protocol
and RDF Query Language) for knowledge extrac-
tion.
5. Validation (Step 5 ) where the validation of
MONTOLOGY is performed to assure that it
2
https://protege.stanford.edu/
3
https://www.w3.org/TR/rdf-sparql-query/
meets the needs of its usage. For this purpose,
specific SPARQL queries are being defined to
evaluate the MONTOLOGY level of detail.
4.2 MONTOLOGY Modules
MONTOLOGY (MONitoring onTOLOGY) is com-
posed of two modules: System of Systems (SoS)
module (yellow classes in Figure 3) and Monitoring
module (green classes in Figure 3). Due to space lim-
itation, we just give some details about the two mod-
ules.
System of Systems (SoS) Module. It models SoS
and its components, as well as the relationship among
them. As in Figure 3, a SystemOfSystems is a col-
lection of Devices that represent the object of the
monitoring activities. Each Device is composed of
a specific set of Components. SoS and Components
are related to as set Attributes: SoS is connected to
SoSAttribute through the hasSoSAtribute relationship,
whereas Component is connected to ComponentAt-
tribute through the haCompAttribute relationship. As
in the figure, Attribute is connected to both Measure
and Metric concepts, and it is a super-class of SoSAt-
tribute and ComponentAttribute. Through this hier-
archy, these sub-classes can be connected to Measure
and Metric concepts as well.
Monitoring Module. It aims at modelling moni-
toring concepts and relationship between them. The
core class of Monitoring module is the Rule, which
is the atomic observable concept within MENTORS.
Indeed, the Monitor observes rules organized in Cal-
endar, which is an ordered set of rules. Each Calen-
dar is able to validate a specific Skill at run-time, i.e.,
during the on-line phase depicted in Figure 1. Skill
is specified as a set of Requirements, each verified
through a specific Rule (see the verifiesRequirement
relationship).
The two main modules are then connected each
other as following: (1) each SystemOfSystems de-
fines a set of Skill, each related to a specific Compo-
nent; (2) the core element of Monitoring module, i.e.,
Rule concept, is defined through a set of Attributes.
WEBIST 2021 - 17th International Conference on Web Information Systems and Technologies
294
Figure 3: Overview of MONTOLOGY.
5 MENTORS: ACTORS, USE
CASES AND ARCHITECTURE
In this section, we describe MENTORS, a monitor-
ing environment for System of Systems (SoS). We
firstly illustrate the main actors involved in MEN-
TORS; then, we describe the supported uses cases; fi-
nally, we present the conceived MENTORS reference
Architecture.
5.1 Actors
As anticipated in the introduction, the monitoring pro-
cess may involves several stakeholders as schema-
tized in Figure 4 Box A . In particular, the main actor
is MENTORS user who is the generic actor interact-
ing with MENTORS framework. This actor can be
in turn specialized into different actors. The Business
Manager, who is the user of the monitoring system.
She or he is in charge of the specification of the mon-
itoring needs (e.g., rules) as well as the administra-
tion and management of the monitoring data analy-
sis and alerts. The Learning System that represents
an autonomous system able to analyze the collected
monitoring data, so as to infer new knowledge (e.g.,
new rules, concepts or relationships) for improving
MONTOLOGY. The CEP Programmer who is an ex-
pert of a specific monitoring system, and she or he is
in charge of instantiating the rules subset into a pre-
cise rules language. The SoS Expert who is expert
of the SoS domain, its constituent parts and their at-
tributes. The Ontology User who has the knowledge
of the ontology concepts and management, and she
or he is the responsible to conceptualize, to imple-
ment and to maintain MONTOLOGY. This actor can
be specialized into Device Expert. They are experts
of a specific device, and he or she knows the Device
constituent parts, their needs and their attributes.
5.2 Use Cases
The use cases were guided by the conceptual model
introduced in Section 3. To better clarify the roles
of each actor, in Figure 4 Box
B the main function-
alities provided by MENTORS are reported. In par-
ticular, Ontology User and Learning System use the
Populate Ontology feature for specifying his or her
knowledge of the SoS or the device. More precisely,
Ontology User can populate MONTOLOGY with in-
MENTORS: Monitoring Environment for System of Systems
295
Figure 4: MENTORS: Actors and Use Cases.
dividuals and assertions about the SoS or the device.
This feature includes also the possibility to modify the
content of MONTOLOGY or extend it with already
existing ontologies. The Ontology User and the Busi-
ness Manager have then the possibility to Navigate
ontology, so as to explore the already existing knowl-
edge and to perform specific queries. Additionally,
Business Manager, through Select rules subset can
navigate the MONTOLOGY and can specify the most
suitable rules to be monitored. The selected rules will
then be instantiated by the CEP programmer through
Instantiate rule feature so as to be observed by se-
lected monitoring infrastructure. In particular, Moni-
toring feature is in charge of checking if the instanti-
ated rules are respected or not. This functionality is
also in charge of the storage of the monitoring data so
as to let the Business Manager to analyze them.
5.3 Proposed Architecture
Our reference architecture is depicted in Figure 5,
which is an abstract architecture that can be cus-
tomized with different real tools. MENTORS refer-
ence architecture is composed of the following com-
ponents:
(1) GUI provides a user-centric & user-friendly
graphical interface that allows MENTORS Users
to interact with MENTORS for performing all
the available functionalities previously introduced
and reported in Figure 4 Box B .
(2) Monitoring component is the core of the analysis
of the system execution implementing the Mon-
itoring and Instantiate rules features described
in Section 5.2. It includes the Complex Event
Processor (CEP) (de Almeida et al., 2020) sub-
component that is in charge of inferring simple
and complex pattern from the events captured dur-
ing the run time. An event, is a snapshot of a
change of the status within a system. Events are
captured by Probes, which are piece of code in-
jected within the device to monitor. Probes are n
charge to put on an envelope the change of sta-
tus occurred within the component monitored and
sent it to the monitoring. The received data will
be matched with the instantiated rules and results
will be stored into the Executed Data DB for fur-
ther analysis.
(3) Instantiated Rules DB contains the rules selected
by Business Manager through the GUI, and in-
stantiated by the CEP Programmer. The rules are
translated into the language of the CEP. In our
case, the Drools Language
4
has been used for the
scope.
(4) Executed Data DB stores data derived from sys-
tem execution, and it makes those data available
for further analysis.
(5) KB Manager implements Populate ontology,
Navigate Ontology and Select rules subset fea-
tures (see Section 5.2). Specifically, users can
manage the ontology through the GUI, while
the Learning System can access through specific
API. As in Figure 5, KB Manager interacts with
the Knowledge Base component for updating and
managing MONTOLOGY. KB Manager can be
considered as a middle-ware between: the known
Knowledge Base, the observed Monitoring and
4
Details about Drools Language can be found at: https:
//www.drools.org/
WEBIST 2021 - 17th International Conference on Web Information Systems and Technologies
296
Figure 5: MENTORS: Supporting Predictive Monitoring Architecture.
the inferred Learning System.
(6) Knowledge Base allows the MONTOLOGY
specification and update. In particular, MON-
TOLOGY is represented as an RDF (Manola and
Miller, 2004) graph, which is saved in a triple
store on which it is possible to perform queries.
(7) Learning System implements the Infer new
knowledge and contributes to the Populate On-
tology feature (see Section 5.2). Specifically,
this component, through a reasoner, analyzes the
Knowledge Base data and the information col-
lected in the Executed Data DB, for both perform-
ing consistency checks and inferring new emerg-
ing knowledge. The results of this analysis are
provided to Knowledge Base component through
KB Manager. This component can be instantiated
with different reasoners available in literature
5
,
such as Openllet
6
.
6 CONCLUSIONS
In this paper, we have introduced MENTORS, a refer-
ence monitoring environment specifically conceived
for SoS. In particular, MENTORS is targeting the sit-
uation in which a new device or component is inte-
5
An updated list of the currently
available reasoners can be found in:
https://www.w3.org/2001/sw/wiki/OWL/Implementations
6
https://github.com/Galigator/openllet
grated into an existing SoS, so as to assess and pre-
vent anomalous and dangerous behaviour of both the
device and the SoS. We firstly conceptualized MEN-
TORS as process consisting of two main parts, Off-
line and On-line. The former is supported by an on-
tological representation of the SoS and Monitoring
domains through MONTOLOGY ontology. This is
a core ontology useful for developing and deriving
meaningful rules that can be monitored, and it can
be exploited also for inferring new knowledge about
the SoS. The latter consists of monitoring the con-
ceived rules so as to promptly identify misbehaviour
in terms of run-time violations. To this purpose, we
have also presented a reference architecture that can
be customized with different tools. As future work,
we are planning to thoroughly validate MENTORS
within an ongoing European Project. This validation
could improve both MONTOLOGY and MENTORS
reference architecture for supporting new functional-
ities.
ACKNOWLEDGEMENTS
This work was partially supported by Cyber-
Sec4Europe H2020 Grant Agreement No. 830929
and the EU H2020 BIECO project Grant Agreement
No. 952702.
MENTORS: Monitoring Environment for System of Systems
297
REFERENCES
Barsocchi, P., Calabr
`
o, A., Ferro, E., Gennaro, C.,
Marchetti, E., and Vairo, C. (2018). Boosting a low-
cost smart home environment with usage and access
control rules. Sensors, 18(6):1886.
Berners-Lee, T., Hendler, J., and Lassila, O. (2001). The
Semantic Web. Scientific American, 284(5):28–37.
Bertolino, A., Calabr
`
o, A., Lonetti, F., and Sabetta, A.
(2011). GLIMPSE: a generic and flexible monitoring
infrastructure. In Giandomenico, F. D., editor, Pro-
ceedings of the 13th European Workshop on Depend-
able Computing, EWDC ’11, Pisa, Italy, May 11-12,
2011, pages 73–78. ACM.
Bhamare, D., Zolanvari, M., Erbad, A., Jain, R., Khan, K.,
and Meskin, N. (2020). Cybersecurity for industrial
control systems: A survey. computers & security,
89:101677.
Boley, H., Tabet, S., and Wagner, G. (2001). Design ra-
tionale for ruleml: A markup language for seman-
tic web rules. In Cruz, I. F., Decker, S., Euzenat,
J., and McGuinness, D. L., editors, Proceedings of
SWWS’01, The first Semantic Web Working Sympo-
sium, Stanford University, California, USA, July 30
- August 1, 2001, pages 381–401.
Burns, M., Griffor, E., Balduccini, M., Vishik, C., Huth,
M., and Wollman, D. (2018). Reasoning about smart
city. In 2018 IEEE International Conference on Smart
Computing (SMARTCOMP), pages 381–386.
de Almeida, V. P., Bhowmik, S., Lima, G., Endler, M., and
Rothermel, K. (2020). Dscep: An infrastructure for
decentralized semantic complex event processing. In
2020 IEEE International Conference on Big Data (Big
Data), pages 391–398. IEEE.
Fotescu, R.-P., Constantinescu, R., Alexandrescu, B., and
Burciu, L.-M. (2020). System for monitoring the
parameters of vehicle. In Advanced Topics in Op-
toelectronics, Microelectronics and Nanotechnologies
X, volume 11718, page 117180A. International Soci-
ety for Optics and Photonics.
Gharib, M., Giorgini, P., and Mylopoulos, J. (2021). Copri
v.2 — a core ontology for privacy requirements. Data
& Knowledge Engineering, 133:101888.
Gruber, T. R. (1995). Toward principles for the design of
ontologies used for knowledge sharing? International
Journal of Human-Computer Studies, 43(5):907–928.
Hidayanti, F. (2020). Design and application of monitor-
ing system for electrical energy based-on internet of
things. Helix, 10(01):18–26.
Khan, S., Nazir, S., Garc
´
ıa-Magari
˜
no, I., and Hussain, A.
(2021). Deep learning-based urban big data fusion
in smart cities: Towards traffic monitoring and flow-
preserving fusion. Computers & Electrical Engineer-
ing, 89:106906.
Manola, F. and Miller, E. (2004). RDF Primer. W3C Rec-
ommendation, WWW Consortium. http://www.w3.
org/TR/rdf-primer/.
Motik, B., Patel-Schneider, P. F., and Parsia, B. (2012).
OWL 2 Web Ontology Language structural specifi-
cation and functional-style syntax (second edition).
W3C recommendation, World Wide Web Consortium.
Palumbo, F. (2018). Leveraging smart environments for
runtime resources management. In Software Quality:
Methods and Tools for Better Software and Systems:
10th International Conference, SWQD 2018, Vienna,
Austria, January 16–19, 2018, Proceedings, volume
302, page 171. Springer.
Raman, R. and D’Souza, M. (2019). Decision learning
framework for architecture design decisions of com-
plex systems and system-of-systems. Systems Engi-
neering, 22(6):538–560.
Rastogi, V., Srivastava, S., Mishra, M., and Thukral, R.
(2020). Predictive maintenance for sme in indus-
try 4.0. In 2020 Global Smart Industry Conference
(GloSIC), pages 382–390.
Santos, M. A., Munoz, R., Olivares, R., Rebouc¸as Filho,
P. P., Del Ser, J., and de Albuquerque, V. H. C. (2020).
Online heart monitoring systems on the internet of
health things environments: A survey, a reference
model and an outlook. Information Fusion, 53:222–
239.
Silva, E., Batista, T., and Oquendo, F. (2015). A mission-
oriented approach for designing system-of-systems.
In 2015 10th System of Systems Engineering Confer-
ence (SoSE), pages 346–351. IEEE.
Ullo, S. L. and Sinha, G. R. (2020). Advances in smart en-
vironment monitoring systems using iot and sensors.
Sensors, 20(11).
Vierhauser, M., Rabiser, R., Gr
¨
unbacher, P., Seyerlehner,
K., Wallner, S., and Zeisel, H. (2016). Reminds : A
flexible runtime monitoring framework for systems of
systems. Journal of Systems and Software, 112:123–
136.
Won, M. (2020). Intelligent traffic monitoring systems
for vehicle classification: A survey. IEEE Access,
8:73340–73358.
WEBIST 2021 - 17th International Conference on Web Information Systems and Technologies
298