A Systematic Mapping Study on Security in Conﬁgurable

Safety-Critical Systems Based on Product-Line Concepts

Richard May, Jyoti Gautam, Chetan Sharma, Christian Biermann and Thomas Leich

Harz University of Applied Sciences, Wernigerode, Germany

Keywords:

Security, Conﬁgurable Systems, Software Product Lines, Safety-Critical Systems, Systematic Mapping Study.

Abstract:

Safety-critical systems are becoming increasingly conﬁgurable. However, as the number of features and con-

ﬁgurations grows, the systems’ complexity also increases, making cyber attacks more likely. Nevertheless,

we miss an overview of security in conﬁgurable safety-critical systems which are based on product-line en-

gineering. Thus, we conducted a systematic mapping study in which we analyzed 44 papers (2008–2022)

to discuss relevant properties and to identify 8 research opportunities. Our key ﬁnding is that security in the

context of variability and safety-critical systems needs more consideration and research. We emphasize that

safety-critical systems, especially those with networking capabilities, cannot be safe if they do not provide

techniques to ensure security and do not consider the systems’ conﬁgurability. Our study is aimed to guide

both researchers and practitioners in understanding the importance of security for conﬁgurable safety-critical

systems, relevant properties, and open issues.

1 INTRODUCTION

Safety-critical systems (SCS), i.e., systems whose

execution can result in fatalities or extensive dam-

ages to a system, its users, or environment, are be-

coming increasingly predominant in various indus-

tries, e.g., manufacturing (Knight, 2002). SCS are

quite software-intensive (Ne

c et al., 2019), mak-

ing the software engineering process highly challeng-

ing as its main goal is to provide software that func-

tions reliably and predictably in the event of a system

breakdown, failure, regular system operation (Gut-

garts and Temin, 2010). As modern SCS usually rely

on networking technologies, addressing security be-

sides safety is more important than ever before. Ex-

ploiting system vulnerabilities or even software bugs

can not only lead to unauthorized data access but also

to fatal system errors, and thus a compromised system

safety (Mubeen et al., 2020). SCS are becoming in-

creasingly conﬁgurable (Lohm

uller and Bauer, 2019),

i.e., they are composed of diverse customizable com-

ponents to meet requirements, e.g., stakeholder de-

mands (Jamshidi et al., 2017). Precisely, they are

based on variability concepts comprising similar but

adapted variants built on the same assets but with

varying features, e.g., based on product-line engineer-

ing (PLE) (Fægri and Hallsteinsen, 2006). However,

as the conﬁgurability of software systems increases,

they also become more complex, e.g., due to conﬁgu-

ration options or feature interactions (Jamshidi et al.,

2017). This situation also leads to a growing attack

surface, making the overall engineering process even

more challenging (Wolschke et al., 2019). However,

there is currently no systematic review of recent re-

search focusing exclusively on the security of PLE-

based SCS, while covering a comparable body of

the literature landscape. By conducting a systematic

mapping study, we address this research gap to pro-

vide an understanding of the intersection of security,

PLE, and SCS. This way, our goal is to explore what

and to what extent security- and variability-related

properties have already been covered by existing re-

search and what topics still need more research. With

our study, we contribute the following:

• A systematic overview of recent research regarding

security of conﬁgurable SCS in PLE.

• A discussion of relevant properties of the investi-

gated topics and research opportunities.

• A replication package to ensure a higher compre-

hensibility and replicability of the study.

Our results can help researchers and practitioners in

selecting research topics and emphasize the impor-

tance of security in the development of variant-rich

software in safety-critical environments.

https://doi.org/10.5281/zenodo.7538781

May, R., Gautam, J., Sharma, C., Biermann, C. and Leich, T.

A Systematic Mapping Study on Security in Conﬁgurable Safety-Critical Systems Based on Product-Line Concepts.

DOI: 10.5220/0012006700003538

In Proceedings of the 18th International Conference on Software Technologies (ICSOFT 2023), pages 217-224

ISBN: 978-989-758-665-1; ISSN: 2184-2833

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

217

2 BACKGROUND

In this section, we provide relevant background infor-

mation on conﬁgurable systems, SCS, and security.

2.1 Conﬁgurable Systems

Today’s systems typically offer the ability to be cus-

tomized (i.e., including or excluding functionalities)

according to stakeholder demands, hardware limita-

tions, or legal regulations (Iqbal et al., 2022). In

this context, system variants comprising individual

features can be developed to meet relevant require-

ments. To ensure that all features have been success-

fully conﬁgured and that they are actually functional,

it is necessary to verify them — usually based on 3

different strategies, i.e., feature-, product-, or family-

based (Th

um et al., 2014). Conﬁgurable systems are

usually based on techniques and tools allowing the or-

ganization, modeling, documentation, and implemen-

tation of their conﬁguration options (i.e., features). A

well-known method to manage these variability me-

chanics is PLE (Schaefer et al., 2012). Conﬁgurable

systems are typically classiﬁed by their projection.

The problem space refers to the abstraction of the do-

main, while the solution space addresses a system’s

implementation. If both spaces are connected it is

called mapping (Apel et al., 2013a).

2.2 Safety-Critical Systems

The term safety-critical refers to the loss of human

lives, economic losses, or environmental damage,

e.g., in avionics (Knight, 2002). So, the main goal in

the development of SCS is to ensure safety, i.e., im-

plementing functions that protect the overall system,

parts of the system, or their users against undesirable

events that could cause harm (Rausand, 2014). One

of the most common safety standards is IEC 61508

(2010). According to this standard, there are 2 failure

categories, precisely hardware failures and systematic

failures. Both categories can be measured by the 4-

level software integrity level (SIL) which is the basic

measure that needs to be considered when develop-

ing safe SCS (IEC 61508, 2010). Nevertheless, due

to increasing software complexity, networking, and

growing data amount, ensuring the minimum possi-

ble safety risk for SCS is becoming quite challenging.

Precisely, safety is sensitive to system changes, e.g.,

system conﬁgurations (Debbech et al., 2019). Fur-

thermore, ensuring safety also implies ensuring secu-

rity of a system (Hatcliff et al., 2014), meaning the

presence of security issues has a major impact on the

safety of SCS (Ebnauf et al., 2019).

2.3 Security

Security is a software quality property, aiming to pro-

tect data against unauthorized access of persons or

systems which are not allowed to (Fægri and Hall-

steinsen, 2006). It is usually characterized by as-

sets (e.g, sensitive data), threats, risks, and mea-

sures (Myll

arniemi et al., 2015). A threat is deﬁned

as an unwanted but potential event that could harm

a system. A risk arises when a threat can poten-

tially be exploited, e.g., attacks, bugs, or environ-

mental errors. Overall, there are 6 security goals for

software systems, including conﬁdentiality, integrity,

availability (i.e., the CIA triad), accountability, au-

thenticity, and non-repudiation (i.e., information se-

curity goals) (ISO/IEC 27000, 2018).

3 METHODS

To accomplish our main goal, we conducted a system-

atic mapping study according to Petersen et al. (2015)

(cf. Figure 1).

IEEE Scopus

ACM

abstract selection

snowballing

title/

data extraction/synthesis

research opportunities

564

109

full-text selection

insights &

Figure 1: Methodological study overview (numbers indi-

cate the amount of selected papers).

3.1 Study Design

Search String. First, we created the following search

string for a search in IEEE XPLORE, SCOPUS, and

the ACM GUIDE TO COMPUTING LITERATURE,

consisting of relevant keywords of PLE and SCS:

(“product line*” OR “SPL” OR “SPLE”) AND

(“product famil*” OR “system famil*” OR “soft-

ware famil*” OR “conﬁg*” OR “variab*” OR

“variant*” OR “feature” OR “model*”) AND

(“safety-critical”)

The term security or related terms are intentionally

not included to be able to also ﬁnd papers only im-

plicitly dealing with security.

We are aware of a mapping study by Kenner et al.

(2021) focusing on safety and security in the context

of conﬁgurable software systems. However, they an-

alyzed safety and security separately (i.e., articles re-

trieved from Scopus that are either about safety or se-

curity) without focusing exclusively on SCS. Thus,

we cover a larger body of knowledge with another re-

search focus.

ICSOFT 2023 - 18th International Conference on Software Technologies

218

Selection Criteria. Papers were only selected if they

were published between 2008 and 2022 at a peer-

reviewed conference or journal provide with at least

4 pages and are concerned with security of conﬁg-

urable SCS. We intentionally excluded work focus-

ing on conﬁgurable SCS but not considering PLE, as

well as work describing safety-oriented approaches

but does not explicitly fall within the scope of SCS.

Extraction Criteria. To extract data, we deﬁned cri-

teria oriented towards PLE and security. We do not

focus on safety but only on security with regard to

our research objectives.

1. Publication

• Publication year of the paper.

• Contribution type as classiﬁcation of a paper,

i.e., open items, method, model, metric, or

tool (Engstr

om and Runeson, 2011).

• Domain of the SCS, e.g., automotive, avionics.

• Perspective indicates whether the paper’s focus

is on security of a system or on a security con-

cern based on PLE (Kenner et al., 2021).

2. Conﬁgurable SCS

• Variability type speciﬁes whether the software,

or both software and hardware are conﬁg-

urable (Queiroz and Braga, 2014).

• Projection reﬂects if the SCS addresses the prob-

lem space, solution space, or a mapping between

both (Apel et al., 2013a).

• Veriﬁcation, classiﬁes the SCS regarding veriﬁ-

cation strategies (Th

um et al., 2014).

• Evolution indicates whether system evolution

aspects are considered (Apel et al., 2013a).

3. Security

• Security standard addressed in the paper.

• Security goals of the software related to the SCS,

including goals of the CIA triad, information se-

curity, and further goals (e.g., reliability).

• Security threats and risks which are addressed

in the publication, including the description of

patterns.

• Security measures discussed or suggested to

tackle security risks.

3.2 Study Conduct

The automated search was conducted on November

, 2022 and yielded in 564 publications (534 af-

ter duplication removal). Then, papers were selected

by the ﬁrst author according to the selection criteria

(109 papers). However, we ﬁrst considered all papers

regarding PLE-related SCS even when they did not

mention security in the title or abstract. After read-

ing full-texts, we selected 37 papers which mention or

describe security to a certain degree. To increase the

number of publications, we did 1 iteration of forward

and backward snowballing. So, we found 7 more pa-

pers resulting in 44 papers in total (cf. Figure 1).

Next, the data extraction was carried out by the ﬁrst,

second, and third author. We used open-coding to la-

bel the identiﬁed data and open-card-sorting to ﬁt re-

curring data into common data categories. Finally, all

data was discussed by all authors to derive both well-

covered topics and future research opportunities.

4 RESULTS

In this section, we present the results of the literature

analysis (cf. Table 1).

4.1 Publication

Publication Years. We did not ﬁnd any work that ad-

dressed security concerns until 2010. In the following

4 years (2011–2014), the number of papers is consis-

tently low (average of 2 papers per year). From 2015–

2018, the number of papers increased to 4 papers per

year. After a peak of 8 papers (2019), the number of

papers decreased to an average of 4 papers per year

(2020–2022). Surprisingly, we identiﬁed only 2 pa-

pers in 2022.

Contribution Type. In most papers (17) a method

was introduced. 12 papers described open items. In

11 cases, models as problem solutions are described.

However, we did not identify an approach focusing

on metrics, especially in the context of security. In 4

cases, authors introduced tools.

Domain. The publications covered various domains.

The most common domains are cyber-physical sys-

tems (8), avionics (5), automotive (5), medical (4),

and production (3). 13 papers deal with general con-

ﬁgurable SCS.

Perspectives. Most publications (32) are concerned

with security for conﬁgurable systems, e.g., se-

curity as quality property for secure communica-

tion (Heikkil

a et al., 2016). In 6 papers, the authors

realized security functions based on variability tech-

niques, e.g., secure PLE in cloud environments (Kri-

eter et al., 2018). Moreover, we extracted 6 ap-

proaches focusing on both perspectives.

4.2 Conﬁgurable Safety-Critical

Systems

Variability Type. We identiﬁed that half of the pa-

pers (22) address only the conﬁgurability of SCS soft-

ware. The other half is focused on the conﬁgurability

of both software and hardware components.

A Systematic Mapping Study on Security in Conﬁgurable Safety-Critical Systems Based on Product-Line Concepts

219

Table 1: Overview of the extracted data related to security and PLE.

Security goals

CIA triad

Information

security

Projection Veriﬁcation

Reference

PLE for security

Security for PLE

Security standard

Conﬁdentiality

Integrity

Availability

Authorization

Accountability

Non-repudiation

Further goals

Threats and risks

Patterns

Measures

Problem space

Mapping

Solution space

Feature-based

Product-based

Family-based

Evolution

Trujillo et al. (2010)

Dordowsky et al. (2011)

Cichos et al. (2011)

Li and Yang (2012)

Apel et al. (2013b)

Ubayashi et al. (2013)

Andel et al. (2014)

Cleland-Huang et al. (2014)

Hatcliff et al. (2014)

Gallina and Fabre (2015)

Ayala et al. (2015)

Vogel-Heuser et al. (2015)

Arrieta et al. (2015)

Barron et al. (2016)

Etigowni et al. (2016)

Kuhrmann et al. (2016)

Heikkil

a et al. (2016)

Barner et al. (2017)

Carpenter et al. (2017)

Nicolas et al. (2017)

Pessoa et al. (2017)

Gannouni et al. (2018)

Islam and Azim (2018)

Krieter et al. (2018)

c and Nyberg (2018)

Bennaceur et al. (2019)

de Oliveira et al. (2019)

Lohm

uller and Bauer (2019)

Ebnauf et al. (2019)

Meixner et al. (2019)

Wolschke et al. (2019)

Shaaban et al. (2019)

Chumpitaz et al. (2019)

Burow et al. (2020)

Ghamizi et al. (2020)

Bressan et al. (2020)

Freitas et al. (2020)

Fischer et al. (2021)

Castro et al. (2021)

c et al. (2021)

White et al. (2021)

Bressan et al. (2021)

Zampetti et al. (2022)

Prikler and Wotawa (2022)

Fulﬁlled Not fulﬁlled

Projection. We found 15 papers in the problem space

and 14 paper in the solution space. 15 publications

covered a mapping.

Veriﬁcation. We discovered that 26 publications de-

scribed veriﬁcation. Usually, the authors presented

a feature-based veriﬁcation (15). In 8 cases, we

found information regarding product-based veriﬁca-

tion. Authors described family-based veriﬁcation

strategies in 3 cases.

Evolution. Software evolution is addressed in 16 pa-

pers. However, we note that evolution is usually only

mentioned and not described in detail. In 28 papers,

evolution was not mentioned at all.

4.3 Security

Standards. We found only 3 papers covering secu-

rity standards. These include AAMI TIR57 as a se-

curity management guidance document in the medi-

cal domain to meet the requirements of ISO 14971,

the general guidelines of the NIST cyber-physical

systems program, and the production-related security

standards IEC 62443 and IEEE 1686.

Goals. The goals of the CIA triad were only described

by 3 papers. Integrity (16) is the most common secu-

rity goal. However, note that integrity may also re-

fer to the SIL. Conﬁdentiality was described by a to-

tal of 12 papers. Only 1 paper referred to account-

ICSOFT 2023 - 18th International Conference on Software Technologies

220

ability (Hatcliff et al., 2014) and 2 papers to non-

repudiation (Shaaban et al., 2019; Burow et al., 2020).

Authorization was mentioned or described in 8 cases.

Threats and Risks. The publications showed diverse

security threats (30) or risks (11). Most common

threats are system complexity (9), trust (7), general

vulnerabilities (6), bugs (3), and conﬁgurability (3).

Mentioned risks include unauthorized access (4), cy-

ber attacks (2), or manipulation (2).

Measures. The identiﬁed security measures can be

organized in 3 groups: threat and risk prevention (11),

system architecture (11), and mitigation techniques

(10). Threat and risk prevention involves risk as-

sessment strategies (7), security policies (2), certiﬁ-

cation (1), and security testing (1). System architec-

ture refers to isolation techniques (4), decentraliza-

tion (2), program partitioning and diversity (2), dy-

namic variants (1), system hiding (1), and changing

library locations (1). The mitigation techniques cover

a wide range and are usually described in a more gen-

eral way, e.g., encryption (2).

5 DISCUSSION

Next, we discuss the study results to provide 8 rele-

vant research opportunities (RO).

5.1 Conﬁgurable Safety-Critical

Systems

Referring to the publication years, there is a trend to-

wards the number of papers related to conﬁgurable

SCS since 2015. Interestingly, until 2017, most pa-

pers focused on proposing methods, while since 2018,

there is a trend regarding open items. This may be due

to the evolving IoT technologies, which offer a vari-

ety of opportunities but also challenges related to sys-

tem architectures as well as security and safety risks.

Consequently, there is an increasing number of papers

which are more related to the problem space or the

mapping. So, we argue that (RO

) more research is

needed focusing on the actual solution of problems or

conﬁgurations (i.e., mappings) of conﬁgurable SCS.

The selected conﬁgurable SCS were not veri-

ﬁed in most cases. However, while general conﬁg-

urable software and storages refer more to product-

based veriﬁcation (Kenner et al., 2021), conﬁgurable

SCS focus much more on the veriﬁcation of fea-

tures. This fact may refer to the safety-critical prop-

erties of SCS, meaning their great dependence on

safety-related functionalities, e.g., the manipulation

of assets such as sensors. Nevertheless, the propor-

tion of papers referring only to safety-related veriﬁ-

Goal classification

CIA triad: 42

Information security: 11

Confidentiality: 12

Integrity: 16

Availability: 14

Authorization: 8

Accountability: 1

Non-repudiation: 2

PLE for Security: 6

Security for PLE: 32

Both: 6

Goal Perspective

Figure 2: Distribution of security goals and perspectives

(numbers indicate the amount of papers which can exceed

100% due to possibility of multiple goals per paper).

cation is most common. In this context, we empha-

size that (RO

) SCS should be veriﬁed regarding both

safety and security functionalities.

64% of the papers do not refer to evolution, e.g.,

feature updates. It is often not clear how such changes

are handled, how they inﬂuence each other, or how

they inﬂuence the security and thus safety of SCS.

So, we argue that it is essential to (RO

) explore the

dependencies of evolution processes, e.g., vulnerabil-

ities caused by evolutionary changes.

5.2 Security

Although security issues have an impact on the safety

of SCS (Ebnauf et al., 2019), no reference was found

to the ISO/IEC 27000 series as a major security stan-

dard. However, security requirements as well as

safety requirements (e.g., SIL) should be taken into

account according to standards. So, we recommend

to (RO

) consider both safety and security standards.

Not surprisingly, information on security, i.e.,

goals, risks, or measures, is usually missing or only

given in a superﬁcial way. However, threats are of-

ten named, albeit from a high-level perspective. For

instance, while complexity or trust are mentioned as

properties that may threaten the security of SCS, ac-

tions to exploit these threats are not described. So,

the existence of threats and their relevance is known

but (RO

) ideas of actual prevention strategies and

the relationships between the IoT as a major driver,

conﬁgurability, and safety seem quite under-explored.

Referring to goals, these are mainly oriented towards

A Systematic Mapping Study on Security in Conﬁgurable Safety-Critical Systems Based on Product-Line Concepts

221

requirements which are most relevant for applications

with IoT capabilities, i.e., the CIA triad and autho-

rization (cf. Figure 2).

5.3 Security in Conﬁgurable

Safety-Critical Systems

Research on SCS focuses rather on safety (52%) than

on security (21%) or both equally (27%). Although

this ﬁnding is not surprising, it highlights that the rel-

evance of security and its conﬁgurability is underesti-

mated. Since the number of attacks is constantly in-

creasing and the possible consequences of a success-

ful attack are even more fatal for SCS, the number of

papers addressing safety and security equally should

be signiﬁcantly higher. Thus, (RO

) both the conﬁg-

urability of security and the equal consideration of

safety and security for SCS needs more research.

Although 30% of the papers are designed for gen-

eral domains, 25% are developed for industrial appli-

cations, e.g., cyber-physical systems. Consequently,

there is a huge potential for SCS-related research

in this area, emphasizing the dependence of safety-

related industrial applications on security. The failure

of SCS can not only endanger people and the environ-

ment in this context but the economic damage, i.e.,

time, quality, and costs, may also be considerable.

However, only 2 of these papers refer to (domain-

related) security standards (e.g., IEEE 1686). Thus,

it is usually not clear, how these standards are applied

and which interactions or dependencies regarding po-

tential conﬁgurations may occur. So, it is essential

to (RO

) provide an understanding of security stan-

dards and implementation possibilities.

Moreover, new requirements based on potential

dependencies or conﬁguration options may occur

when considering both safety and security require-

ments in the SCS development process, e.g., cyber-

physical systems and the application of IoT technolo-

gies. Although the interplay of safety and security

is already discussed in research (Lyu et al., 2019),

the consideration of variability aspects inﬂuencing

the implementation of these requirements is missing.

Thus, (RO

) there is a need for analyzing emerging

dependencies between safety and security in the con-

text of conﬁgurability.

5.4 Threats to Validity

We lack comprehensive information about the accu-

racy of results (e.g., evaluations) as well as term un-

derstandings (e.g., integrity regarding safety or secu-

rity). Some authors explain their work in great detail

while others describe their approach from a high-level

perspective. Another issue arises regarding the fulﬁll-

ment of the extraction criteria. Speciﬁcally, the papers

usually provide only a few details on security. Both

issues regarding the internal validity caused several

assumptions on our side and might have led to misin-

terpretations. We are aware that our automated search

and manual ﬁltering strategy may have resulted in the

unintended exclusion of publications. This issue may

pose threats to the external validity. We aimed to miti-

gate these threats by systematically extracting data by

several researchers relying on established deﬁnitions

and criteria. All results were discussed until consen-

sus was achieved, especially in the context of data in-

terpretation and assignments. Overall, we argue that

our results are valuable and provide highly interesting

insights as we covered 15 years of research relying on

3 well-established literature databases.

6 RELATED WORK

We found several papers dealing with the analy-

sis of security in the context of conﬁgurable sys-

tems. For instance, Hammani (2014) analyzed non-

functional requirements in the context of PLE. The

review of Queiroz and Braga (2014) is concerned with

approaches dealing with PLE and SCS (2006–2013),

however, without considering security. Geraldi et al.

(2020) analyzed papers which are concerned with the

IoT as well as PLE (2006–2018). SCS are men-

tioned as domain while security is considered as a

non-functional requirement.

Overall, most related work covers rather software

engineering in general, is focused either on security,

safety, or SCS, is not related to PLE, or is not as com-

prehensive as our study (i.e., number of analyzed pa-

pers or literature databases). In contrast to these stud-

ies, we provide a comprehensive review with a differ-

ent focus than any related work, i.e., security of PLE-

related conﬁgurable SCS. The closest work related to

ours is a mapping study by Kenner et al. (2021) which

is focused on security and safety of conﬁgurable soft-

ware systems (2011–2020). However, they did not

exclusively focus on security of SCS.

7 CONCLUSION

In this paper, we conducted a systematic mapping

study on security concerns in conﬁgurable SCS

(2008–2022). We identiﬁed relevant insights and

presented 8 research opportunities in the context of

secure, conﬁgurable SCS. Overall, the intersection

of security, conﬁgurability, and SCS needs more

ICSOFT 2023 - 18th International Conference on Software Technologies

222

research to create an understanding of how security

is related to conﬁgurability and which dependencies

exist between security and safety in this context.

We emphasize that SCS cannot be safe if they do

not provide techniques to ensure security which

also takes into account variable features. So, when

modeling or conﬁguring SCS, security features and

their strategies must be as dynamic as the system fea-

tures are. Further research is strongly recommended,

e.g., analyzing relevant security requirements in

accordance to current security and safety standards.

REFERENCES

Andel, T. R., Whitehurst, L. N., and McDonald, J. T. (2014).

Software security and randomization through program

partitioning and circuit variation. In MTD. ACM.

Apel, S., Batory, D., K

astner, C., and Saake, G. (2013a).

Feature-oriented software product lines. Springer.

Apel, S., Von Rhein, A., Wendler, P., Gr

oßlinger, A., and

Beyer, D. (2013b). Strategies for product-line veriﬁ-

cation: Case studies and experiments. In ICSE. IEEE.

Arrieta, A., Sagardui, G., and Etxeberria, L. (2015). Cyber-

physical systems product lines: Variability analysis

and challenges. Jornadas de Computaci

on Empo-

trada.

Ayala, I., Amor, M., Fuentes, L., and Troya, J. M. (2015).

A software product line process to develop agents for

the iot. Sensors, 15(7).

Barner, S., Diewald, A., Migge, J., Syed, A., Fohler,

G., Faugere, M., and P

erez, D. G. (2017). Dreams

toolchain: Model-driven engineering of mixed-

criticality systems. In MODELS. IEEE.

Barron, S., Cho, Y. M., Hua, A., Norcross, W., Voigt, J.,

and Haimes, Y. (2016). Systems-based cyber security

in the supply chain. In SIEDS. IEEE.

Bennaceur, A., Ghezzi, C., Tei, K., Kehrer, T., Weyns, D.,

Calinescu, R., Dustdar, S., Hu, Z., Honiden, S., and

Ishikawa, F. (2019). Modelling and analysing resilient

cyber-physical systems. In SEAMS. IEEE.

Bressan, L., de Oliveira, A. L., and Campos, F. (2020).

An approach to support variant management on safety

analysis using chess error models. In EDCC. IEEE.

Bressan, L., de Oliveira, A. L., Campos, F., and Capilla,

R. (2021). A variability modeling and transformation

approach for safety-critical systems. In VaMoS. ACM.

Burow, N., Burrow, R., Khazan, R., Shrobe, H., and Ward,

B. C. (2020). Moving target defense considerations

in real-time safety-and mission-critical systems. In

MTD. ACM.

Carpenter, T., Hatcliff, J., and Vasserman, E. Y. (2017). A

reference separation architecture for mixed-criticality

medical and iot devices. In SafeThings. ACM.

Castro, T., Teixeira, L., Alves, V., Apel, S., Cordy, M., and

Gheyi, R. (2021). A formal framework of software

product line analyses. TOSEM, 30(3).

Chumpitaz, L., Furda, A., and Loke, S. (2019). Evolving

variability requirements of iot systems. In Software

Engineering for Variability Intensive Systems. Auer-

bach Publications.

Cichos, H., Oster, S., Lochau, M., and Sch

urr, A. (2011).

Model-based coverage-driven test suite generation for

software product lines. In MODELS. Springer.

Cleland-Huang, J., Gotel, O. C. Z., Huffman Hayes, J.,

ader, P., and Zisman, A. (2014). Software trace-

ability: Trends and future directions. In ICSE.

de Oliveira, A. L., Braga, R., Masiero, P., Parker, D., Pa-

padopoulos, Y., Habli, I., and Kelly, T. (2019). Vari-

ability management in safety-critical systems design

and dependability analysis. Journal of Software: Evo-

lution and Process, 31(8).

Debbech, S., Bon, P., and Dutilleul, S. C. (2019). Con-

ceptual modelling of the dynamic goal-oriented safety

management for safety critical systems. In ICSOFT.

ACM.

Dordowsky, F., Bridges, R., and Tschope, H. (2011). Imple-

menting a software product line for a complex avion-

ics system. In SPLC. IEEE.

Ebnauf, M., Abdelmoez, W., Ammar, H. H., Hassan, A.,

and Abdelhamid, M. (2019). State-driven architecture

design for safety-critical software product lines. In

ICOM. IEEE.

Engstr

om, E. and Runeson, P. (2011). Software product line

testing–a systematic mapping study. Information and

Software Technology, 53(1).

Etigowni, S., Tian, D., Hernandez, G., Zonouz, S., and But-

ler, K. (2016). Cpac: Securing critical infrastructure

with cyber-physical access control. In ACSAC. ACM.

Fægri, T. E. and Hallsteinsen, S. (2006). A software product

line reference architecture for security. In Software

Product Lines. Springer.

Fischer, S., Ramler, R., Klammer, C., and Rabiser, R.

(2021). Testing of highly conﬁgurable cyber-physical

systems–a multiple case study. In VaMoS. ACM.

Freitas, L., Scott III, W. E., and Degenaar, P. (2020).

Medicine-by-wire: Practical considerations on formal

techniques for dependable medical systems. Science

of Computer Programming, 200.

Gallina, B. and Fabre, L. (2015). Beneﬁts of security-

informed safety-oriented process line engineering. In

DASC. IEEE.

Gannouni, W., Doumbia, M. L., and Badri, A. (2018). Sys-

tematic approach furthering conﬁrmation measures of

safety critical automotive systems. Transactions on

The Built Environment, 174.

Geraldi, R. T., Reinehr, S., and Malucelli, A. (2020). Soft-

ware product line applied to the internet of things: A

systematic literature review. Information and Software

Technology, 124.

Ghamizi, S., Cordy, M., Papadakis, M., and Traon, Y. L.

(2020). Featurenet: Diversity-driven generation of

deep learning models. In ICSE. ACM.

Gutgarts, P. B. and Temin, A. (2010). Security-critical ver-

sus safety-critical software. In HST. IEEE.

Hammani, F. Z. (2014). Survey of non-functional require-

ments modeling and veriﬁcation of software product

lines. In RCIS. IEEE.

A Systematic Mapping Study on Security in Conﬁgurable Safety-Critical Systems Based on Product-Line Concepts

223

Hatcliff, J., Wassyng, A., Kelly, T., Comar, C., and Jones, P.

(2014). Certiﬁably safe software-dependent systems:

Challenges and directions. FOSE.

Heikkil

a, T., Dobrowiecki, T., and Dalgaard, L. (2016).

Dealing with conﬁgurability in robot systems. In

MESA. IEEE.

IEC 61508 (2010). Functional safety. Standard, IEC.

Iqbal, M. S., Krishna, R., Javidian, M. A., Ray, B., and

Jamshidi, P. (2022). Unicorn: Reasoning about con-

ﬁgurable system performance through the lens of

causality. In EuroSys. ACM.

Islam, N. and Azim, A. (2018). Assuring the runtime be-

havior of self-adaptive cyber-physical systems using

feature modeling. In CSSE. ACM.

ISO/IEC 27000 (2018). Information technology – security

techniques – information security management sys-

tems. Standard, ISO.

Jamshidi, P., Velez, M., K

astner, C., Siegmund, N., and

Kawthekar, P. (2017). Transfer learning for improving

model predictions in highly conﬁgurable software. In

SEAMS. ACM.

Kenner, A., May, R., Kr

uger, J., Saake, G., and Leich, T.

(2021). Safety, security, and conﬁgurable software

systems: a systematic mapping study. In SPLC.

Knight, J. C. (2002). Safety critical systems: Challenges

and directions. In ICSE.

Krieter, S., Kr

uger, J., Weichbrodt, N., Sartakov, V. A.,

Kapitza, R., and Leich, T. (2018). Towards secure

dynamic product lines in the cloud. In ICSE. IEEE.

Kuhrmann, M., Ternit

e, T., Friedrich, J., Rausch, A., and

Broy, M. (2016). Flexible software process lines in

practice: A metamodel-based approach to effectively

construct and manage families of software process

models. JSS, 121.

Li, D. and Yang, Y. (2012). Enhance value by building trust-

worthy software-reliant system of systems from soft-

ware product lines. In PLEASE. IEEE.

Lohm

uller, P. and Bauer, B. (2019). Software product line

engineering for safety-critical systems. In MODEL-

SWARD. ACM.

Lyu, X., Ding, Y., and Yang, S.-H. (2019). Safety and secu-

rity risk assessment in cyber-physical systems. Cyber-

Physical Systems: Theory & Applications, 4(3).

Meixner, K., Rabiser, R., and Bifﬂ, S. (2019). Towards

modeling variability of products, processes and re-

sources in cyber-physical production systems engi-

neering. In SPLC. ACM.

Mubeen, S., Lisova, E., and Vulgarakis Feljan, A. (2020).

Timing predictability and security in safety-critical in-

dustrial cyber-physical systems: A position paper. Ap-

plied Sciences, 10(9).

Myll

arniemi, V., Raatikainen, M., and M

annist

o, T. (2015).

Representing and conﬁguring security variability in

software product lines. In QoSA. ACM.

c, D. and Nyberg, M. (2018). Verifying contract-based

speciﬁcations of product lines using description logic.

In DL, volume 2211. CEUR-WS.

c, D., Nyberg, M., and Gallina, B. (2019). Constructing

product-line safety cases from contract-based speciﬁ-

cations. In SAC. ACM.

c, D., Nyberg, M., and Gallina, B. (2021). Product-line

assurance cases from contract-based design. JSS, 176.

Nicolas, C.-F., Eizaguirre, F., Larrucea, A., Barner, S.,

Chauvel, F., Sagardui, G., and Perez, J. (2017). Gsn

support of mixed-criticality systems certiﬁcation. In

SafeComp. Springer.

Pessoa, L., Fernandes, P., Castro, T., Alves, V., Rodrigues,

G. N., and Carvalho, H. (2017). Building reliable and

maintainable dynamic software product lines: An in-

vestigation in the body sensor network domain. Infor-

mation and Software Technology, 86.

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.

Prikler, L. M. and Wotawa, F. (2022). Challenges of testing

self-adaptive systems. In SPLC. ACM.

Queiroz, P. G. G. and Braga, R. T. V. (2014). Development

of critical embedded systems using model-driven and

product lines techniques: A systematic review. In SB-

CARS. IEEE.

Rausand, M. (2014). Reliability of Safety-Critical Systems:

Theory and Applications. Wiley.

Schaefer, I., Rabiser, R., Clarke, D., Bettini, L., Benavides,

D., Botterweck, G., Pathak, A., Trujillo, S., and Vil-

lela, K. (2012). Software diversity: State of the art and

perspectives. STTT, 14(5).

Shaaban, A. M., Gruber, T., and Schmittner, C. (2019).

Ontology-based security tool for critical cyber-

physical systems. In SPLC. ACM.

um, T., Apel, S., K

astner, C., Schaefer, I., and Saake, G.

(2014). A classiﬁcation and survey of analysis strate-

gies for software product lines. CSUR, 47(1).

Trujillo, S., Perez, A., Gonzalez, D., and Hamid, B.

(2010). Towards the integration of advanced engi-

neering paradigms into rces: Raising the issues for

the safety-critical model-driven product-line case. In

SD4RCES. ACM.

Ubayashi, N., Nakajima, S., and Hirayama, M. (2013).

Context-dependent product line engineering with

lightweight formal approaches. Science of Computer

Programming, 78(12).

Vogel-Heuser, B., Fay, A., Schaefer, I., and Tichy, M.

(2015). Evolution of software in automated produc-

tion systems: Challenges and research directions. JSS,

110.

White, D., Sahlab, N., Jazdi, N., and Weyrich, M. (2021).

Environment modeling for evaluating system variants

in model-based systems engineering. Procedia CIRP,

104.

Wolschke, C., Becker, M., Schneickert, S., Adler, R., and

MacGregor, J. (2019). Industrial perspective on reuse

of safety artifacts in software product lines. In SPLC.

ACM.

Zampetti, F., Tamburri, D. A., Panichella, S., Panichella,

A., Canfora, G., and Penta, M. D. (2022). Continuous

integration and delivery practices for cyber-physical

systems: An interview-based study. Transactions on

Software Engineering and Methodology.

ICSOFT 2023 - 18th International Conference on Software Technologies

224