IoT-based Systems Actuation Conflicts Management Towards DevOps:
A Systematic Mapping Study
St
´
ephane Lavirotte
a
, G
´
erald Rocher
b
, Jean-Yves Tigli and Thibaut Gonnin
Universit
´
e C
ˆ
ote d’Azur, CNRS, Laboratoire I3S, 06903 Sophia-Antipolis, France
Keywords:
Internet of Things, Cyber-Physical Systems, Conflict Identification, Conflict Resolution, DevOps.
Abstract:
The Internet of Things (IoT) has long been understood as an infrastructure layer allowing to gather environ-
mental data through sensors. However, it also provides means to physically interact with our living environ-
ments through actuators. To the extent that actuation effects are not without risks on safety and trustworthiness,
providing the IoT infrastructure layer with merely sensors access control mechanisms is no longer sufficient.
It is also required to prevent conflicting (and possibly unsafe) actuation effects to occur in the physical en-
vironment and deploy means to resolve them. In this paper, we consider actuation conflicts management as
part of the DevOps approach, which aims to harmonize tools and objectives of actors involved in IoT-based
systems life cycle from their design to their deployment. In this context, a systematic mapping study (SMS) is
conducted to better understand the actuation conflicts management approaches and to what extent they could
be integrated into the DevOps life cycle.
1 INTRODUCTION
Internet of Things (IoT) based systems generally fol-
low a layered architecture: (1) a shared IoT infras-
tructure layer consisting of a set of connected and
distributed resources (e.g., processing units, memory,
sensors, actuators, etc.) likely to be embedded in ev-
eryday objects (chair, lamp, etc.) and/or things (room,
building, etc.), (2) a top layer where applications are
deployed and, (3) one or more intermediate layers
managing communications and ensuring the overall
coherency between applications and the shared infras-
tructure. Thereby, the literature refers to three-layer
architecture, middleware architecture, service-based
architecture, ve-layer architecture etc. (Kumar and
Mallick, 2018). The notion of coherence, here, has a
strong connection to that of conflict, “[...] ...a con-
text change that leads to a state of the environment
which is considered inadmissible by the application
or user’” (Tuttlies et al., 2007).
Following these architectural schemes, IoT-based
systems have long been limited to collecting field in-
formation from sensors; in this context, the problem
the intermediate layers must solve is merely tech-
nological; it aims to provide the infrastructure layer
with a sensors access control mechanism (Cecchinel
a
https://orcid.org/0000-0002-3341-6577
b
https://orcid.org/0000-0002-3874-6276
et al., 2014). The notion of coherence then refers to
the problem of managing direct conflicts (i.e., con-
current accesses). Managing actuation conflicts is a
much more challenging task. It is not only a ques-
tion of managing direct conflicts but also indirect con-
flicts that arise from the concurrent interactions of
actuators with a common physical system and that
can lead to an undesirable evolution of some of its
properties (Teixeira et al., 2011) (e.g., simultaneously
heating and cooling a room). In this context, the
problem the intermediate layers must solve, beyond
being technological, lies in the semantic interpreta-
tion of the effects produced in the environment and
their consequences for humans in terms of safety and
trustworthiness. From that viewpoint, by making ex-
plicit the interactions with the physical environment
through the IoT shared infrastructure layer, IoT-based
Cyber-Physical Systems (CPS/IoT) (Shih et al., 2016;
Damjanovic-Behrendt et al., 2018), an “integration of
computation with physical processes, intersection of
the physical and the cyber (Lee and Seshia, 2016),
are most likely to face with this problem.
Actuation conflicts management is a first class
concern in the realm of thrustworthy and safe IoT-
based systems, justifying the efforts put by the Eu-
ropean Union on this topic (e.g. ENACT project
1
,
1
https://www.enact-project.eu
Lavirotte, S., Rocher, G., Tigli, J. and Gonnin, T.
IoT-based Systems Actuation Conflicts Management Towards DevOps: A Systematic Mapping Study.
DOI: 10.5220/0009355102270234
In Proceedings of the 5th International Conference on Internet of Things, Big Data and Security (IoTBDS 2020), pages 227-234
ISBN: 978-989-758-426-8
Copyright
c
2020 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
227
Brain-IoT project
2
and SecureIoT
3
). However, to
date, only few surveys, systematic mappings and lit-
erature reviews have been conducted on the actuation
conflicts management problem in the context of IoT-
based systems. In (Resendes et al., 2014), the authors
conducted a Systematic Literature Review (SLR) on
the conflict detection and resolution problem in Home
and Building Automation Systems (HBAS). They
propose a taxonomy that classifies conflicts accord-
ing to four different dimensions: (1) source, (2) inter-
venients, (3) time of detection and (4) solvability. Al-
though it provides an overview of the research on con-
flicts detection and resolution, the study is restricted
to HBAS and does not provide recent information on
how far these topics are covered in research.
Actuation conflict management is crucial and calls
for a methodological break in the development pro-
cess of IoT-based systems. In this context, the De-
vOps approach, an agile and incremental develop-
ment approach that aims to harmonize the practices of
actors involved in all stages of a system life cycle (i.e.
from development to deployment and maintenance),
is promising.
The goal of this paper is to clarify the interest
and scope of recent research on the management of
actuation conflicts in the area of IoT-based systems.
In particular, we are interested in analyzing actua-
tion conflict management in the perspective of the De-
vOps approach. To this end, we rely on the system-
atic mapping approach, well established in evidence
based medicine and dedicated to provide researchers
with the ability to build a classification scheme and
structure a field of interest from which specific re-
search questions can be answered (Heinz, 2014; Sny-
der, 2019).
2 BACKGROUND
In this section, we provide an overview of Internet
of Things, Cyber Physical Systems and DevOps ap-
proach underlying the scope of this study.
2.1 Internet of Things (IoT)
Thanks to constant innovations in electronics and
communication technologies, the use of our surround-
ing physical entities (chair, lamp, houses, cities, etc.)
is transcended. Being connected to the Internet, new
forms of interaction are emerging from these physi-
cal entities thanks to sensors and actuators, embodied
2
http://www.brain-iot.eu/
3
https://secureiot.eu
in remotely accessible software services and resulting
in the fusion of the cyber and physical dimensions of
our environments. In this context, IoT is “the infras-
tructure enabling advanced services by interconnect-
ing (physical and virtual) things based on existing and
evolving interoperable information and communica-
tion technologies” (International Telecommunication
Union, ITU).
From an architectural point of view, IoT-based
systems are generally multi-layered. The Fig. 1
includes some examples from (Kumar and Mallick,
2018). Here, it is interesting to note the purpose given
to the shared IoT infrastructure layer:
Perception layer
Network layer
Application layer
Perception layer
Transport layer
Processing layer
Application layer
Business layer
Security layer
Storage layer
Pre-processing layer
Monitoring layer
Physical layer
Transport layer
Three layer Five layer
Six layer (FOG)
Figure 1: IoT architectures layers (Kumar and Mallick,
2018).
Perception Layer. Provides the ability to detect, col-
lect and gather information about the physical envi-
ronment and the connected objects within it.
Environmental Layer. Provides the ability to detect
objects or places that are under observation. This in-
cludes the ability to observe moving physical entities,
such as humans, cars, etc. and environmental proper-
ties such as temperature or humidity.
This semantics is somehow representative of the
prevailing idea that consists in associating IoT to envi-
ronmental sensing/monitoring capabilities. Although
this vision has led to tremendous improvements in
human well-being and assistance, optimization of re-
sources, etc., it is important not to forget the action ca-
pabilities offered by actuators. Only recently has this
capacity, and the associated risks to both humans and
their environments, been seriously considered. This is
witnessed by some on-going European projects. For
instance, ENACT project acknowledges that [...] IoT
system innovations have until now mainly been con-
cerned with sensors, device management and connec-
tivity, with the mission to gather data for process-
IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security
228
ing and analysis in the cloud” and consider actuation
as a first class concern in IoT [...] The next gen-
eration IoT systems need to [..] manage the closed
loop from sensing to actuation (Ferry et al., 2018).
BRAIN-IoT project specifically aims to support the
integration into an IoT environment of devices and
subsystems with actuation features that could possi-
ble give rise to mixed-critically situations (Conzon
et al., 2019).
2.2 Cyber-Physical Systems (CPS)
Cyber-Physical Systems (CPS) are a generalization of
the concept of embedded systems to that of connected
things with the objective of making them collaborate
for the control of physical processes (Rajkumar et al.,
2010). CPS find applications in the optimization of
resources, their means of supply, etc. at the heart
of the Industry4.0 and Industrial IoT (IIoT) revolu-
tions. However, by controlling physical processes,
these systems are not without risks for humans and
their environment, as evidenced by the SecureIoT Eu-
ropean project. This project aims to develop secure
services targeting the areas of digital automation in
manufacturing (Industry 4.0), social assistance robots
for coaching and health and connected cars and au-
tonomous driving. Such mechanisms are highly de-
manded by the industry in order to secure a whole new
range of IoT applications that transcend the bound-
aries of multiple IoT platforms, while involving au-
tonomous interactions between intelligent CPS sys-
tems and networks of smart objects”.
2.3 DevOps Approach
The DevOps (Sharma and Coyne, 2017) approach
aims to harmonize the practices of software devel-
opment (development, integration and testing) and
systems administration (deployment, operation and
maintenance) stakeholders (Fig. 2). This harmoniza-
tion is justified by the conflicting objectives of these
actors; on the one hand, software developers are con-
strained by cost and time, with the negative impacts
that this can have on the quality of the software de-
livered. On the other hand, IT administration actors
seek to achieve stability and quality objectives, at the
expense of costs and deadlines.
This approach is based on agile and lean manage-
ment methods and results in the collaboration of busi-
ness managers, developers, operations and quality
stakeholders in order to continuously deploy differ-
ent software versions. In this context, this approach
aims to pool the tools for implementing software ap-
plications, from their design to their deployment. The
Release
Deploy
Monitor
Operate
Code
Test
Build
Plan
Dev
Ops
Figure 2: DevOps life cycle.
underlying interest of this harmonization lies in the
repetability of the processes implemented throughout
the DevOps loop (Fig. 2) and paves the way for their
automation (Lwakatare et al., 2015). The latter is im-
portant in order to accelerate and maintain the conver-
gence towards a system that meets the requirements of
all actors involved. Automation is of particular impor-
tance in IoT-based systems where connected devices
are constantly providing feedback. Automation then
brings reactivity in updating these systems throughout
the DevOps loop as soon as they need to be.
In this paper, we seek to understand to what ex-
tent the tools for identifying and resolving actuation
conflicts can be integrated into IoT-based systems life
cycle in the context of DevOps. To this end, we rely
on the systematic mapping approach (Snyder, 2019)
consisting in establishing research questions and or-
ganizing the answers obtained from the study of sci-
entific publications. The methodology and results are
exposed in the sequel.
3 RESEARCH METHOD
This systematic mapping study was developed follow-
ing the guidelines proposed in (Petersen et al., 2015).
On the basis of the context and the motivations pre-
sented in §1 and 2, we define the research questions
(RQ) in §3.1. In order to define the scope of the study
and reduce possible biases in the selection process, we
explain the inclusion and exclusion criteria in §3.2.
3.1 Research Questions
This study aims to answer the set of research ques-
tions described in Table 1. The questions relating
to conflict management (RQ2) remain rather general,
the objective being to understand the limitations of
the current actuation conflicts management methods
towards their implementation within the DevOps ap-
proach.
IoT-based Systems Actuation Conflicts Management Towards DevOps: A Systematic Mapping Study
229
Table 1: Research questions.
RQ1 What are the primary studies statistics?
RQ1.1 What is the publication rate over years?
RQ1.2
In which types of venue (workshop, confer-
ence, journal) were the studies published?
RQ1.3
How studies are distributed in terms of aca-
demic and industrial affiliation and location?
RQ1.4 What application domains are concerned?
RQ2 How actuation conflicts are managed?
RQ2.1
What actuation conflicts are considered (di-
rect/indirect)?
RQ2.2
At what stage of the IoT-based systems life cy-
cle are they implemented?
RQ2.3 What is their automation level?
RQ2.4 What is their maturity level?
3.2 Search Strategy
The study was conducted using three different
databases: ACM-DL, IEEE Xplore and Scopus. Dur-
ing the selection process, the inclusion and exclusion
criteria defined in tables 2 and 3 were applied.
Table 2: Inclusion criteria.
Inclusion criteria
Primary peer-reviewed paper
The scope of the paper is fully related to the research
questions
Paper written in English language
Publication year 2008
Table 3: Exclusion criteria.
Exclusion criteria
White paper, technical report, thesis, book chapter,
patent and presentation
The content of the paper is not appropriate to answer the
research questions
Duplicate
The queries associated to each database are given
in table 4. The scope of the search is restricted to
papers related to IoT and CPS domains published in
2008 onward. 2008 was a pivotal year in the field
of IoT. This is the year from which the number of
IoT-related publications has started to increase signif-
icantly (Fig. 3). It was also the year in which the
first international conference on IoT was held (Flo-
erkemeier et al., 2008).
The second conjunctive part of the queries re-
stricts the scope of the search to papers dealing with
actuation, source of the direct and indirect conflicts.
While the notion of actuation conflict has a strong
meaning in the field of robotics (the physical side), in
the software engineering domain (the cyber side), it is
more about the one of Feature Interaction (FI) (Bruns,
2005), a feature “being a unit of functionality that can
2006
2008 2010 2012 2014
2016
2018
0.5
1
1.5
2
·10
4
#publications
Figure 3: Evolution of publications related to the Internet
of Things (IoT).
be developed and evolved independently (Bocovich
and Atlee, 2016). Therefore, it is worth taking into
account these different semantic interpretations in the
search, IoT-based systems developers having a strong
software engineering culture. Finally, the third con-
junctive part of the queries expresses the notion of
conflict through a set of synonyms.
On the basis of the papers found using the search
string, a backward snowballing technique was used
(Wohlin, 2014) in order to identify additional relevant
papers. By following this approach, three more pa-
pers were added. Taking the guidelines and applying
the exclusion criteria, an extensive review of the se-
lected papers was made by three researchers analyz-
ing the title, the abstract and the content of each ex-
tracted paper. Consensus on keeping or rejecting pa-
pers was found during meetings conducted through-
out the selection process. Finally, a total of 26 papers
(Table 10) were selected as a result of this classifi-
cation process while 2842 papers were excluded, as
shown in Table 5.
4 RESULTS HIGHLIGHTS
The following sections are devoted to providing an
analysis of the selected publications according to the
research questions identified in §3.1.
4.1 Overview
This section provides answers to the research question
RQ1 and associated sub-questions RQ1.1, RQ1.2,
RQ1.3 and RQ1.4.
Answering RQ1.1: Until 2011, there were only few
papers dealing with actuation conflict management in
IoT-based systems as depicted in Fig. 4.
IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security
230
Table 4: Search strings and filters for each database.
Library Advanced /Command search query string Added filters
ACM-DL
(”internet of things” OR iot OR ”cyber-physical system” OR cps)
AND (actuation OR actuator OR feature) AND (conflict OR inter-
action OR interference OR shared OR concurrency)
Published 2008 onward
IEEE Xplore
(”internet of things” OR iot OR ”cyber-physical system” OR cps)
AND (actuation OR actuator OR feature) AND (conflict OR inter-
action OR interference OR shared OR concurrency)
Year range: from 2008 to 2019
Conferences and Journals
Scopus
TITLE-ABS-KEY((”internet of things” OR iot OR ”cyber-physical
systems” OR cps) AND (actuation OR actuator OR feature) AND
(conflict OR interaction OR interference OR shared OR concur-
rency))
Year: limit to ”2008 to 2019”
Subject area: limit to Computer Science
Document type: limit to Conferences
and Journals
Table 5: The search and selection stages for primary studies.
Source SearchResults Removing duplicates Reviewing titles Reviewing abstract Scanning content
ACM-DL 866 834 112 29 12
IEEE Xplore 970 943 92 28 9
Scopus 1032 505 40 16 5
Total 2868 2107 269 78 26
Figure 4: Primary studies per year.
Based on the Fig.4, although limited, one can de-
note an increasing publication trend on the subject
from 2011 onward, demonstrating the growing in-
terest of the research community to the IoT-based
systems actuation conflict management problem.
We conducted our search process in November 2019
which can explain the low amount of papers published
for this year.
Answering RQ1.2: Fig. 5 depicts the distribution of
primary studies published over years and per venue
type. There is a constant number of publications dur-
ing the period 2014-2018. 24% are journal papers,
62% are conference papers and the remaining 14%
are workshop papers.
Answering RQ1.3: by focusing on authors’ affilia-
tion, one can denote in Fig. 6 that most of the authors
of the primary studies are researchers (83%).
The involvement of industry in this research area
is still limited (3%). There are also joint research be-
Figure 5: Primary studies per venue type.
Figure 6: Primary studies per affiliation type.
tween academic and industry (14%). This underline
the raising importance of actuation conflict manage-
ment concern in industrial use cases, still at an aca-
demic research level. These results tend to confirm
that IoT-based systems actuation conflict manage-
IoT-based Systems Actuation Conflicts Management Towards DevOps: A Systematic Mapping Study
231
ment problem is still in its infancy. The top most
countries are United States of America (8), Austria
(4), following by Canada, France and Japan (3). Re-
searchers from other countries are also involved in
this research for a total of 19 countries.
Answering RQ1.4: in terms of application areas, a
short predominance can be observed for Smart Home
relative use-cases (and, in the broad sense, smart-*
systems), ahead of the automotive use-cases. The ap-
plications specific to CPS as defined in §2.2 will be
found under the terms Automotive, Smart Factory and
Robotics. Bottom-line, IoT-based systems actua-
tion conflict management problem affects all socio-
economical layers ranging from humans (Smart
Health), houses (Smart Home) to cities (Smart
City) and industry (Smart Factory).
Figure 7: Primary studies per domains.
4.2 Analysis towards the DevOps
Perspective
Answering RQ2.1: Direct and indirect conflicts are
taken into account and are well balanced (Table 6).
Papers on IoT deal with direct conflicts more than
indirect conflicts. This trend is reversed for papers
on CPS that are more focused on indirect conflicts.
By making interactions with the physical environment
explicit, it is expected to observe such a trend for CPS.
Table 6: Actuation conflicts types considered in studies.
IoT CPS Total
Direct 34.48% 13.79% 48.27%
Indirect 20.69% 31.03% 51.72%
Total 55.17% 44.82% 100%
Answering RQ2.2 & RQ2.3: The table 7 shows the
Dev/Ops dichotomy in the identification of direct and
indirect conflicts. Whether at design time (Devs) and
operation (Ops), the analysis shows a well-balanced
Table 7: Actuation conflicts identification.
@design @runtime Total
Direct 27.58% 20.69% 48.27%
Indirect 24.14% 27.58% 51.72%
Total 51.72% 48.27% 100%
handling to the identification of direct and indirect
conflicts.
Concerning the resolution, the picture is not as
good (Table 8). Conflict resolution is far from being
addressed in all the papers, 20.69% of them discuss
actuation conflicts identification without proposing a
solution to resolve them. On the other hand, 51.71%
of the papers assume manual actuation conflict reso-
lution. In addition to the ad-hoc nature that this type
of management implies, this raises also the question
of the scalability of the proposed approaches. In-
deed, the number of the possible combination of inter-
actions and their effects increase exponentially with
the number of interactions considered. Scalability is
considered a first-class concern for only three papers
(S15, S19, S22).
Besides scalability, it is also worth noting that au-
tomation is at the heart of the DevOps approach by
allowing reactive and timely update of these systems
throughout their life cycle, all the more important as
IoT devices never stop providing feedback. Given
the large number of distributed actuators (and sen-
sors) likely to be involved, a non automated solution
is likely to fail in insuring these timely updates. In
this context, only 27% of the papers rely on a param-
eterized approach or synthesize a conflict manager to
resolve them, the latter type of approach representing
10% of the papers. These latter approaches therefore
seem to be the most relevant for addressing the scala-
bility and automation issues and deserve further work.
Table 8: Actuation conflicts resolution strategies.
None
Manual
Parameterized
Synthesis
Direct 6.89% 31.03% 3.45% 6.89%
Indirect 13.80% 20.68% 13.80% 3.44%
Total 20.69% 51.71% 17.25% 10.33%
Table 9: Actuation conflicts management maturity.
Theoretical
Use-case
In silico
In vitro
In vivo
Direct 3.45% 6.89% 17.24% 20.70% 0.00%
Indirect 0.00% 13.80% 31.03% 6.89% 0.00%
Total 3.45% 20.69% 48.27% 27.59% 0.00%
IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security
232
Table 10: List of the papers selected for the study (J: Journal, C: Conference, W: Workshop).
# Title Year v # Title Year v
1 Formal Verification of Cyber-physical Fea-
ture Coordination with Minimalist Qualitative
Models
2019 J 16 Towards a model-based verification method-
ology for Complex Swarm Systems
2016 C
2 IoTC2: A formal method approach for de-
tecting conflicts in large scale IoT systems
2019 C 17 An Application Conflict Detection and Reso-
lution System for Smart Homes
2015 W
3 A spatially aware policy conflict resolution
for information services
2018 J 18 Dependable control systems with Internet of
Things
2015 J
4 Automata-Based Generic Model for Interop-
erating Context-Aware Ad-Hoc Devices in
Internet of Things
2018 J 19 Safe Composition in Middleware for the In-
ternet of Things
2015 W
5 Context Aware Virtual Assistant with Case-
Based Conflict Resolution in Multi-User
Smart Home Environment
2018 C 20 Coordination of ECA rules by verification
and control
2014 C
6 IotSan: Fortifying the safety of IoT systems 2018 C 21 DepSys: Dependency aware integration of
cyber-physical systems for smart homes
2014 C
7 Taming and optimizing feature interaction in
software-intensive automotive systems
2018 C 22 Distributed programming framework for fast
iterative optimization in networked cyber-
physical systems
2014 J
8 Continuous variable-specific resolutions of
feature interactions
2017 C 23 ECA rules for IoT environment: A case
study in safe design
2014 C
9 Event management for simultaneous ac-
tions in the Internet of Things
2017 C 24 Harnessing evolutionary computation to en-
able dynamically adaptive systems to man-
age uncertainty
2013 W
10 Modeling architectures of cyber-physical
systems
2017 C 25 Component-oriented Interoperation of Real-
time DEVS Engines
2011 C
11 Synchronization abstractions and separa-
tion of concerns as key aspects to the in-
teroperability in IoT
2017 C 26 Programming support for distributed opti-
mization and control in cyber-physical sys-
tems
2011 C
12 Towards Model Checking of Network Appli-
cations for IoT System Development
2017 J 27 Semantic Web-based policy interaction de-
tection method with rules in smart home for
detecting interactions among user policies
2011 J
13 Detection of Runtime Conflicts among Ser-
vices in Smart Cities
2016 C 28 Toward a programming model for safer per-
vasive spaces
2010 W
14 Minimalist qualitative models for model
checking cyber-physical feature coordina-
tion
2016 C 29 Feature interaction detection in the automo-
tive
2008 C
15 SPIRE: Scalable and Unified Platform for
Real World IoT Services with Feature Inter-
action
2016 C
Answering RQ2.4: We can see here the maturity
problem already mentioned in the research/industry
dichotomy (Fig. 6). In particular, none of the pro-
posed approaches for identifying and resolving actua-
tion conflicts have been validated in vivo. This obser-
vation is aggravated for indirect conflicts whose val-
idation does not go beyond the in silico stage. This
confirms the difficulty of implementing the manage-
ment of indirect actuation conflicts, and therefore the
importance of evaluating them to reach the maturity
required for a transfer.
5 CONCLUSION
This SMS study highlighted recent work on the ac-
tuation conflict management in the field of IoT-based
systems. More specifically, we have been interested
in studying their applicability into the DevOps ap-
proach. While direct/indirect conflicts identification
methods applicability is well balanced from develop-
ment to execution stages, most of the current resolu-
tion methodologies lacks important properties to pre-
tend to their exploitation within the DevOps loop.
Indeed, most of them are not automated failing to
scale and ensure timely and reactive systems updates
throughout their life cycle (which is one of the main
reason for promoting the DevOps approach).
Furthermore, the maturity of the identification and
resolution of actuation conflict does not go beyond in
vitro and in silico stages respectively. No experimen-
tation has been conducted in-vivo.
This study will need to be complemented by a
Systematic Literature Review (SLR) to analyse the
IoT-based Systems Actuation Conflicts Management Towards DevOps: A Systematic Mapping Study
233
different methods used for conflict identification and
resolution.
ACKNOWLEDGMENTS
The research has received funding from the European
Commission’s H2020 Program under grant agreement
numbers 780351 (ENACT).
REFERENCES
Bocovich, C. and Atlee, J. M. (2016). Feature-oriented
modelling in bip: A case study. In ModComp@ MoD-
ELS, pages 6–11.
Bruns, G. (2005). Foundations for features. In FIW, pages
3–11. Citeseer.
Cecchinel, C., Mosser, S., and Collet, P. (2014). Soft-
ware Development Support for Shared Sensing In-
frastructures: A Generative and Dynamic Approach.
In Schaefer, I. and Stamelos, I., editors, Software
Reuse for Dynamic Systems in the Cloud and Beyond,
Lecture Notes in Computer Science, pages 221–236,
Cham. Springer International Publishing.
Conzon, D., Rashid, M. R. A., Tao, X., Soriano, A., Nichol-
son, R., and Ferrera, E. (2019). Brain-iot: Model-
based framework for dependable sensing and actua-
tion in intelligent decentralized iot systems. In 2019
4th International Conference on Computing, Commu-
nications and Security (ICCCS), pages 1–8. IEEE.
Damjanovic-Behrendt, V., M
¨
uhlberger, M., de Luca, C.,
Christos, T., and Arnautovic, E. (2018). Iot4cps–
trustworthy iot for cps.
Ferry, N., Solberg, A., Song, H., Lavirotte, S., Tigli, J.-Y.,
Winter, T., Munt
´
es-Mulero, V., Metzger, A., Velasco,
E. R., and Aguirre, A. C. (2018). Enact: Development,
operation, and quality assurance of trustworthy smart
iot systems. In International Workshop on Software
Engineering Aspects of Continuous Development and
New Paradigms of Software Production and Deploy-
ment, pages 112–127. Springer.
Floerkemeier, C., Langheinrich, M., Fleisch, E., Mattern,
F., and Sarma, S. E. (2008). The Internet of Things:
First International Conference, IOT 2008, Zurich,
Switzerland, March 26-28, 2008, Proceedings, vol-
ume 4952. springer.
Heinz, M. (2014). Systematic mapping studies. Mainz:
Universit
¨
at Koblenz-Landau.
Kumar, N. M. and Mallick, P. K. (2018). The internet of
things: Insights into the building blocks, component
interactions, and architecture layers. Procedia com-
puter science, 132:109–117.
Lee, E. A. and Seshia, S. A. (2016). Introduction to embed-
ded systems: A cyber-physical systems approach. Mit
Press.
Lwakatare, L. E., Kuvaja, P., and Oivo, M. (2015). Dimen-
sions of devops. In International conference on agile
software development, pages 212–217. Springer.
Petersen, K., Vakkalanka, S., and Kuzniarz, L. (2015).
Guidelines for conducting systematic mapping stud-
ies in software engineering: An update. Information
and Software Technology, 64:1–18.
Rajkumar, R., Lee, I., Sha, L., and Stankovic, J. (2010).
Cyber-physical systems: the next computing revolu-
tion. In Design Automation Conference, pages 731–
736. IEEE.
Resendes, S., Carreira, P., and Santos, A. C. (2014). Con-
flict detection and resolution in home and building
automation systems: a literature review. Journal
of Ambient Intelligence and Humanized Computing,
5(5):699–715.
Sharma, S. and Coyne, B. (2017). Devops for dummies. 3rd
limited ibm edn.
Shih, C.-S., Chou, J.-J., Reijers, N., and Kuo, T.-W. (2016).
Designing cps/iot applications for smart buildings and
cities. IET Cyber-Physical Systems: Theory & Appli-
cations, 1(1):3–12.
Snyder, H. (2019). Literature review as a research method-
ology: An overview and guidelines. Journal of Busi-
ness Research, 104:333–339.
Teixeira, T., Hachem, S., Issarny, V., and Georgantas, N.
(2011). Service Oriented Middleware for the In-
ternet of Things: A Perspective. In Abramowicz,
W., Llorente, I. M., Surridge, M., Zisman, A., and
Vayssi
`
ere, J., editors, Towards a Service-Based Inter-
net, Lecture Notes in Computer Science, pages 220–
229, Berlin, Heidelberg. Springer.
Tuttlies, V., Schiele, G., and Becker, C. (2007). Comity-
conflict avoidance in pervasive computing environ-
ments. In OTM Confederated International Confer-
ences” On the Move to Meaningful Internet Systems”,
pages 763–772. Springer.
Wohlin, C. (2014). Guidelines for snowballing in system-
atic literature studies and a replication in software en-
gineering. In Proceedings of the 18th international
conference on evaluation and assessment in software
engineering, page 38. Citeseer.
IoTBDS 2020 - 5th International Conference on Internet of Things, Big Data and Security
234