Received January 2, 2021, accepted January 14, 2021, date of publication January 18, 2021, date of current version

February 2, 2021.
Digital Object Identifier 10.1109/ACCESS.2021.3052311

Systematic Mapping Study on Security

Approaches in Secure Software Engineering
RAFIQ AHMAD KHAN 1 , SIFFAT ULLAH KHAN 1, HABIB ULLAH KHAN 2,

AND MUHAMMAD ILYAS 1

1 Software Engineering Research Group, Department of Computer Science and IT, University of Malakand, Chakdara 18800, Pakistan
2 Department of Accounting and Information Systems, College of Business and Economics, Qatar University, Doha, Qatar
Corresponding author: Habib Ullah Khan ([email protected])
This work was supported in part by the Qatar National Library, Doha, Qatar, and in part by the Qatar University under Grant
IRCC-2020-009.

ABSTRACT In the modern digital era, software systems are extensively adapted and have become an integral
component of human society. Such wide use of software systems consists of large and more critical data
that inevitably needs to be secured. It is imperative to make sure that these software systems not only
satisfy the users’ needs or functional requirements, but it is equally important to make sure the security
of these software systems. However, recent research shows that many software development methods do
not explicitly include software security measures during software development as they move from demand
engineering to their final losses. Integrating software security at each stage of the software development
life cycle (SDLC) has become an urgent need. Tackling software security, various methods, techniques,
and models have been suggested and developed, however, only a few of them provide strong evidence for
building secure software applications. The main purpose of this research is to study security measures in the
context of the development of secure software (SSD) during the study of systematic mapping (SMS). Based
on the inclusion and exclusion criteria, 116 studies were selected. After the data extraction from the selected
116 papers, these were classified based on the quality assessment, software security method, SDLC phases,
publication venue, and SWOT analysis. The results indicate that this domain is still immature and sufficient
research work needs to be carried out particularly on empirically evaluated solutions.

INDEX TERMS Software security, secure software development, secure software engineering, software
development life cycle, security approaches, systematic mapping study.

I. INTRODUCTION key concern, such as the Internet of Things (IOT) and the
Over the last two decades, the software industry observed Internet of Every Things, the advancement of Internet-based
phenomenal growth, and the same is continued at a rapid software systems, cloud computing, social networking, and
pace. Software is now an important aspect of our lives and location-based services. Therefore, software programs grow
it seems almost impossible to find a field that does not have in size, complexity, inclination, and connectivity.
the use of software in their day to day business. The world In addressing both the technological and human aspects
in every aspect has been modernized by an immense use of involved, there is a growing need for understanding Secure
software systems. Software Engineering (SSE) methods. SSE is about building
On the contrary, misuse of software can lead to heavy software that can deal with potentially aggressive attacks,
economic loss in the financial sector, sabotage in the com- maintaining basic security features: privacy, integrity, and
munication sector, critical data theft in databases, and misuse access to sensitive assets [1]–[3]. Also besides, new business
of software in the missile controlling system can endanger paradigms, versatile customers’ requirements, rapid advance-
human life. Rapid developments in information and commu- ment in ICTs, and new regulations are constantly making
nication technologies (ICTs) have made software security a a software application to evolve accordingly [4]. SSE rec-
ommends that software security is an important factor to
The associate editor coordinating the review of this manuscript and be taken into account during the start of the software life
approving it for publication was Sedat Akleylek . cycle (SDLC) [5]. To build and deploy a secure software

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 9, 2021 19139
 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

system, we need to integrate security features into our life RQ1.3: What is the SWOT analysis of security approaches
cycle of application development and align current SSE in secure software development?
methods [6], [7]. RQ1.4: What are the popular venues for secure software
Most businesses view security as a post-development development?
process [8]. Security isn’t considered at some point in the The rest of the paper is organized as follows: Section II
pre-development phase [9]. There is no approval for the presents the concepts of software security, section III presents
method to be used, we still have little understanding of the relevant work, and section IV describes the research
the need for secure software development. There are also methodology. The outcomes of the SMS are discussed in
few facts about the effectiveness of existing approaches to section V according to the research questions. Section VI rep-
dealing with real problems and a limited view of how they resents the implications for results and practice. Section VII
contribute to the assessment of safety concerns [10]. Threats gives the findings and future work. Finally, in section VIII,
put systems at greater risk for major losses that can be dif- the risks to validity are addressed.
ficult to recover [11]. The majority of software programs
are designed and deployed without attention to protection II. SOFTWARE SECURITY
desires [12], [13]. Hidden attacking risks within or outside This section introduces the concept of software security for
the organization are emerging day-by-day, results in huge future discussion that will serve as context building.
financial loss, as well as confidentiality and credibility losses
by putting the availability and integrity of organizational A. THE CONCEPTS OF SOFTWARE SECURITY
data at risk [14], [15]. The coding phase of SDLC is more Let’s take a look at some of the software security concepts as
prone to error, as the programmer leaves some errors unin- stated in the literature:
tentionally, which increases software vulnerability to more • ‘‘The idea of engineering software that continues to
attacks [11], [16]. Such vulnerabilities can be the denial function correctly under malicious attack [7], [20]’’.
of services, code execution, memory corruption/data loss, • ‘‘The process of designing, building, and testing soft-
cross-site scripting (XSS), improper access control, SQL ware for security [21], [22]’’.
injection, integer overflow, buffer overflow, and the format • ‘‘Software security is the process of discussing an
string, etc., [12], [17]. application to discover risks and vulnerabilities of the
Based on the above evidence, we can conclude that pro- application and its data [23]’’.
tecting software programs in the development stages isn’t • ‘‘Software security is a system-wide issue that takes
sufficient and there is a great need to locate higher approaches into account both insecurity mechanisms (such as access
and ways to protect software programs. This paper provides control) and design for security (such as a robust design
a systematic study of maps (SMS) based on our pre-defined that makes software attack difficult)’’ [24].
process and the proposed model [18], [19] to highlight • ‘‘Software security is about building secure software:
the existing security measures for building secure software. designing software to be secure, making sure that soft-
It is reported that the findings, in our study reported in ware is secure, and educating software developers, archi-
this paper, will have an impact on the body of information tects, and users about to build secure things [25]’’.
by providing a tax on the SSE-related research topic that • ‘‘Defends against software exploit by building software
may invite researchers to focus on further research in this to be secure in the first place mostly be getting by
field. the design right (which is hard) and avoiding common
We investigate the outcome of this mapping study to mistakes (which is easy) [26]’’.
provide a reference direction for interested and vigilant • ‘‘Software security is the ability of the software to
researchers to address and explore new research trends and resist, tolerate and recover from events that intentionally
gaps in the subject domain. The findings of the SMS may threaten its dependability [27]’’.
assist the practitioners by providing a deep insight into the Security in various terms has been described by vari-
subject domain about security approaches, security limita- ous researchers. In the above concepts, the important thing
tions, and unresolved specific and general issues. to remember is that most of the definitions speak about
To achieve the aforementioned objectives, this SMS ‘‘building secure software’’ instead of ‘‘securing software’’.
addresses the following research questions (RQs): Building secure software involves designing and securely
RQ1. What is the state-of-the-art in Secure Software implementing the software while securing software tries to
Engineering? develop the software first and then enforce security measures
To answer RQ1, we have analyzed the literature based on to make it secure.
the following sub-questions:
RQ1.1:What are the existing security methods, imple- B. SOFTWARE SECURITY REQUIREMENTS
mented by software development organizations that return the Software security has remained a neglected area, from the
largest competitive edge? earliest generations of software development. But that does
RQ1.2: Which particular SDLC phase has been most not mean that the problem has never been raised before;
discussed and addressed in the literature? however, it was misunderstood, taken lightly, misjudged, and

19140 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

not done as it should have been. During the entire software III. BACKGROUND
development life cycle, software security is an essential fac- Have you ever tried to find the cause of insecure software?
tor that needs to be addressed [18]. In general, security is Who is responsible for unsafe software? Service analysts do
characterized as a non-functional requirement, and, for this their best to find operational and non-operational require-
reason, security checks are usually carried out during the ments that meet customer needs. System designers are doing
final of SDLC [9], [28], [29]. It means that software secu- their best to find the most powerful design. Developers are
rity requires proper care even in the first stage of software doing their best to improve the app in a very efficient way.
development [9], [15], [28]. Testers do their best to detect software crashes. From demand
Today, Internet-enabled applications, the removal of bugs inquiries to software development, from software testing to
in the form of buffer overload, and incompatible error man- maintenance, the entire project team has put all its efforts into
agement are major issues in software security [22], [30]. the SDLC to ensure quality software but instead of all these
Millions of people do business by different means every day, efforts, the software is still produced with risks and numerous
such as the Internet, ATM, cell phone, email, etc. The soft- security errors.
ware is used by people who remember that it is reliable and In addition to the extensive efforts made to build secure
trustworthy and that the services they perform are secure. But software and everyone is playing their part to the best of their
if this includes security gaps, then how can they be considered ability; makes you wonder where these security errors came
secure? Security in software has become an important part from? To answer this query, we need to suppose carefully,
of daily life. Due to budget constraints and software release have we pointed out security to this point? Is security taken
time in the market, many developers consider security as a into consideration at any level of SDLC? I’m afraid the
subsequent thinking problem that may have poor software solution is nearly a large ‘NO’ at all. Engineers only broaden
quality [9]. Software security was considered part of software particular requirements (overall performance necessities) and
testing in the early days, but over time, it has been shown that ignore safety requirements. But it will be more beneficial
security is not a backward concern and it is very important if they take the security requirements in the earlier phases
to consider how software engineers can incorporate security and following the same in the requirement gathering phases;
into the early stage of SDLC [28]. such as design, development, and testing. And once they
To this end, this study aims to analyze the process of incorporate the security at all stages of SDLC, the software
software development from the perspective of each SDLC developed by adapting the security measures will produce
phase and to attempt to define key security measures to be more secure software.
used to make the most secure applications in all stages of the Security activities during the requirement phase serve three
SDLC. purposes [13], [23]: Initial security requirements are iden-
tified and implemented. Second, with the security require-
C. QUALITIES OF A SECURE SOFTWARE ments in hand, the project team understands and recognizes
Secure software is about building software that can with- the importance of security. Finally, with the needs of security
stand strong attacks, maintaining basic security structures: in the hands, budget, resources, and time of security activities
confidentiality, integrity, and access to sensitive assets [31]. in future stages can be better estimated.
These three security structures are called the CIA [32]. Any During the design phase, the project team focuses on iden-
software that enlists the CIA can be considered as secure tifying the attacker’s interests, potential access points, and
software [32]. Software security characteristics are defined as critical security areas, etc. [34], [35]. The next step is to iden-
‘‘the degree to which a product or system protects information tify the threats running on the software. All the security data
and data so that persons or other products or systems have collected in the design phase so far goes into the threatening
the degree of data access appropriate to their types and levels model. Threatening models can be considered an important
of authorization’’ [33]. Features of secure software are con- milestone in terms of secure software [36]. Does the security
fidentiality, integrity, denial, accountability, and authenticity. building function provide full details of how the software can
Other aspects of software security are as follows [33]: be attacked? What can be attacked? What areas of attack are
• Secure Data Transport attractive? What kind of threats work etc. [37]. According
• Protect Database Storage to this information, the security structure is continuously
• Authorized Data Access updated to include security.
• Secure Authorization The implementation phase plays a twofold role from a
• Internal / output authentication security perspective [38], [39]: First, it avoids security errors
• Power of Evidence entering the software, and secondly, detects existing software
• Complete Access errors. The first role is done by writing a secure code. The
• Alternative Identification second role of detecting security errors begins with static
• Accessibility Management capabilities analysis by automated tools. After automatic analysis, a man-
• Session Management Powers ual update is performed. After that, the software is fully
• Secure User Management functional and ready to go to the testing phase.

VOLUME 9, 2021 19141

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

After implementation, the software is sent to the testing approaches are McGraw’s Secure Software Development
team. The tests were performed mainly on test cases gen- Life Cycle (SSDLC) process [22], Microsoft Software Devel-
erated during test planning [40]. The testing team identifies opment Life Cycle (SDL) or Trustworthy Computing Secu-
security errors, reports to the development team, and the rity Development Life Cycle [53], Security Requirements
development team corrects them in this code [40], [41]. The Engineering Process (SREP) [54], Aprville and Pourzandi’s
testing phase ends when all test cases are conducted, and Secure Software Development Life Cycle process [55],
retrospective testing of all sensitive areas has taken place [42]. Core security requirements artifacts [56], Comprehensive,
Like any other form of testing, security testing involves Lightweight Application Security Process (CLASP) [57],
determining who should do it and what activities they should Haley and his colleagues’ framework [58], and Security Qual-
undertake. Because security testing involves two approaches, ity Requirements Engineering (SQUARE) [59].
the question of who should do it has two answers. Standard OWASP Security Verification Standard (ASVS) version
testing organizations using a traditional approach can perform 3.0 is a community effort to establish a framework of secu-
functional security testing [43]. For example, ensuring that rity requirements and controls that focus on normalizing
access control mechanisms work as advertised is a classic the functional and non-functional security controls required
functional testing exercise. On the other hand, traditional when designing, developing, and testing modern web applica-
QA staff will have more difficulty performing risk-based tions [60]. The ASVS is a list of application security require-
security testing. The problem is one of expertise. First, secu- ments or tests that can be used by architects, developers,
rity tests (especially those resulting in complete exploits) testers, security professionals, and even consumers to define
are difficult to craft because the designer must think like an what a secure application is [60].
attacker [43]. Second, security tests don’t often cause direct ISO/IEC 27001:2005 covers all types of organizations
security exploits and thus present an observability problem. (e.g. commercial enterprises, government agencies, not-for-
A security test could result in an unanticipated outcome that profit organizations) [61]. It specifies the requirements for
requires the tester to perform further sophisticated analysis. establishing, implementing, operating, monitoring, review-
Bottom line: risk-based security testing relies more on exper- ing, maintaining, and improving a documented Information
tise and experience than we would like [43]. Security Management System within the context of the orga-
Before the release of the software, a security review was nization’s overall business risks. It specifies requirements
performed [44]. The purpose of the review is to identify the for the implementation of security controls customized to
remaining security errors. The developing team corrects code the needs of individual organizations or parts thereof. It is
against security errors identified in the review report. After designed to ensure the selection of adequate and proportion-
a review, a security audit was conducted, and according to ate security controls that protect information assets and give
such an audit report, management decided to issue a soft- confidence to interested parties.
ware [45]. After such a release, the software is ready for Browser identity indicators, including URLs and
shipment. EV certificates, are supposed to help users identify phishing,
After release and distribution, the software is commercially social engineering, and other attacks, but prior lab studies
used. Later, a decision was made to rectify non-critical safety and surveys suggested that older browser identity UIs are
errors [46]. So another code is changed to remove these not effective security tools [62]. Modern browser identity
security errors in the form of a patch. The patch is then applied indicators are not effective. To design better identity indi-
to the software after testing and the patch is released [47]. cators, we recommend that browsers consider focusing on
Because of this, to address the software system security active negative indicators, explore using prominent UI as an
various models, practices, strategies, and methods have been opportunity for user education, and incorporate user research
proposed and developed to improve security procedures in the into the design phase [62].
stages of SDLC [9], [24], [32], [48]. To effectively address We conclude that the most relevant contributions with
security issues that exist during the application process, it is such goals are the work carried out by Nabil et al. [24],
necessary to consider secure considerations in all develop- Silva et al. [63], and Guiena et al. [64]. The authors
ment processes that minimize the threats of critical secu- in [24], identified and classified the available software secu-
rity requirements or to identify critical errors in software rity approaches in SDLC. To identify and mitigate the soft-
development [24]. ware security threats, Silva et al. [63], covered the current
Some security approaches aim to assist the software engi- technologies. For the ubiquitous system, Guiena et al. [64]
neers in evaluating security risks; such as Attack Trees [49], identified 132 approaches, address issues in various phases
combining goal-orientation and use-case modeling, which of the software engineering cycle. Most of the stud-
is an effective method of software requirement engineer- ies addressed maintenance/evolution, implementation, and
ing [50], Secure Tropos, a security-oriented extension to feedback phases.
the goal-driven requirements engineering methodology [51], The main focus of our study is to cover the most relevant
whereas others allow the software engineers to address SSE models, frameworks, methods, processes, metrics, and
these risks by reusing design decisions [52] or sustaining topics in the existing literature. The result can be used as a
the decision making process [36]. Other software security reference guide and direction for future research.

19142 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

C. RESEARCH QUESTIONS
In Table 1, the research questions (RQs) are discussed along
with their key motivations.

D. SEARCH STRATEGY
A search using a search string composed of keywords specific
to this study was carried out to address the RQs and applied
to many academic electronic libraries and search engines.

1) SEARCH STRING
The first step is to establish the search string, and the param-
eters for PICO (Population, Intervention, Comparison, and
Outcomes) have been reported in the literature [65], [71].
• Population: Secure software development
• Intervention: Software Security
• Comparison: Due to the exploratory study, the Compar-
FIGURE 1. Stages of SMS process. ison is not relevant and so excluded.
Outcomes: Security Approaches
The search string was constructed on connecting the fea-
IV. RESEARCH METHODOLOGY tures of PICO by Boolean AND and OR connectors:
The main purpose of the SMS is to provide formal ways to ((‘‘software security’’ OR ‘‘software privacy’’ OR ‘‘secure
integrate the information found in simple basic lessons with software’’ OR ‘‘software protection’’ OR ‘‘software safety’’)
a set of map questions [65]. SMS is a literature study that AND (‘‘Software Engineering’’ OR ‘‘Software Development
focuses on selecting and combining all high-quality research, lifecycle’’ OR ‘‘SDLC’’ OR ‘‘Software security Model’’))
related to a specific topic and provides a complete summary
of current texts applicable to specific map queries [66]. Com- 2) LITERATURE RESOURCES
pared to a systematic literature review (SLR), SMS were per- We run the search string in various online digital libraries
formed on a wide range of research questions to identify gaps such as IEEE Xplore, ACM, Springer Link, Science Direct,
in a particular research area. It also identifies gaps in existing and Wiley Online Library. In Google Scholar, we also run this
subjects and determines trends in future studies [67]–[69]. search string. Table 2 displays the number of search results
SMS, therefore, maintains a great imminent value in the field per search engine/database.
of software engineering by providing a general view of the
literature in this particular domain. E. STUDY SELECTION CRITERIA
According to Felderer and Carver [70], the method of The search strings were inspired by similar researches [24],
conducting SMS consists of three main stages: planning, [34], [69], [72]–[75] and also from the authors’ suggestions.
conducting, and reporting. The goal of these phases is Initially, the search string was added to the metadata of the
to classify, analyze and interpret, based on the strength selected libraries without affecting the title, abstract, and
of their evidence, all available studies relevant to a spe- keyword constraints. The first author retrieved each paper
cific subject, to draw conclusions, and finally provide and detailed information was documented about each relevant
recommendations [65], [68]–[70]. The various stages used in paper. The other authors were asked to review the articles
performing SMS are shown in Figure 1. using their title and abstract, to base the previous steps.
Based on SMS guidelines, the following inclusion and
A. RESEARCH QUESTIONS exclusion criteria have been established [65].
In Table 1, the research questions (RQs) are discussed along
with their key motivations. 1) INCLUSION CRITERIA
• IC1: Papers in the area of software security.
• IC2: Papers related to SDLC.
B. SEARCH STRATEGY
• IC3: The paper focuses on one or several security
A search using a search string composed of keywords specific
approaches from SSD perspectives.
to this study was carried out to address the RQs and applied
to many academic electronic libraries and search engines. 2) EXCLUSION CRITERIA
• EC1: Papers that are not published in journals, confer-
1) SEARCH STRING ences, workshops, or symposiums.
The first step is to establish the search string, and the param- • EC2: Papers that aren’t available in English.
eters for PICO (Population, Intervention, Comparison, and • EC3: A research that is not in the Software Engineering
Outcomes) have been reported in the literature [65], [71]. domain.

VOLUME 9, 2021 19143

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 1. Research questions and main motivation.

TABLE 2. Search string results per database.

questionnaire was therefore designed to assess the quality of

candidate papers.
Senior researchers at the Software Engineering Research
Group, University of Malakand (SERG_UOM), in Pakistan,
distribute and fill out the questionnaire. The scoring plan
applied in Appendix A (a, b, c, and d columns) is presented
below:
a) The paper was published in a recognized and stable
journal or conference. This question was ranked by consid-
ering the computer science conference ranking in Computing
Research Education (CORE)∗ , 2018, and Journal Citation
Reports (JCR)∗∗ , 2018 lists.
FIGURE 2. Search and selection process. Conferences, workshops, and symposiums were assigned
to one of the following categories:
• EC4: Papers with workshop summaries. • A1 - Flagship conference, a leading venue in a discipline
• EC5: The studies which occur several times in the final area. ‘‘Very relevant (+2)’’
set (duplicated studies) • A - Excellent conference and highly respected one in a
• EC6: Books, web pages, and magazine articles are discipline area. ‘‘Relevant (+1.5)’’
excluded. • B - Good conference and well regarded one in a disci-
The summary of the search process is shown in Figure 2. pline area. ‘‘Somewhat Relevant (+1)’’
In total, 556 papers relating to security approaches for SSD • C - Other ranked conferences that meet minimum stan-
were found after the implementation of the inclusion criteria. dards. ‘‘Not relevant (+0.5)’’
After the removal of duplicate papers from various sources • No ranking (+0), One not presenting CORE ranking.
of the same paper, the remaining papers were 476. Finally, The JCR2 published by Thomson Reuters3 offers annual
upon applying the exclusion criteria through which we further rankings of science and social science journals in the subject
removed 360 studies, we selected 116 papers as net studies. categories relevant to the journal, based on Impact Factor (IF)
The paper IDs and titles of the 116 papers finally selected are results. In each subcategory, quartile rankings are obtained
listed in Appendix-A.
1∗ http://portal.core.edu.au/conf−ranks/
F. QUALITY ASSESSMENT 2∗∗ https://www.scimagojr.com/journalrank.php?category=1710
Quality assessment is the main activity that increases the 3∗∗ http://mjl.clarivate.com/cgi-bin/jrnlst/jlsearch.cgi?PC=MASTER&
strength of a study. To enhance the quality of our work, a short Error=1

19144 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

from each journal by occupying the quartile of the IF distri- Systematic reviews, therefore, emphasize that evaluating the
bution in the same subject category. quality of the study selection is important. The quality assess-
Q1 - Denotes the top 25% of the IF distribution. ‘‘Very ment is an important step in gaining a general view of the
relevant (+2)’’ consequences of the paper on the subject [77].
• Q2 – Denotes middle and high positions (between top
50% and top 25%). ‘‘Relevant (+1.5)’’ A. WHAT ARE THE SECURITY METHODS IMPLEMENTED
• Q3 –Denotes Middle-low position (top 75% to BY SOFTWARE DEVELOPMENT ORGANIZATIONS THAT
top 50%). ‘‘Somewhat Relevant (+1)’’ RETURN THE LARGEST COMPETITIVE EDGE?
• Q4 –Denotes the lowest position (bottom 25% of the IF To answer RQ1.1, we extracted data by considering the pres-
distribution). ‘‘Not relevant (+0.5)’’ ence of each SSE method. The results of the collected data
• No ranking (+0), One not in the JCR list. according to this question are highlighted in Table 3.
a) The primary interest of the paper is the security In this SMS ‘‘Software security measurement and anal-
approach used in the SSD context. Yes (+1), Partially ysis’’ is the most cited (29%) category. CORAS [78]
(+0.5), and No (+0). is the most popular method in this category. CORAS
b) The paper presents and/or explicitly assesses an works as a model-driven method used towards the defen-
approach as a solution to deal with software security sive risk analysis. Another most cited (6%) SSE method
risks. The study obtains the full score (Yes (+1)) if in this category is ‘Appropriate-and-Effective-Guidance-for-
it presents a new approach, (Partially (+0.5)) if it Information-Security (AEGIS)’. AEGIS [79] also works as a
presents an existing approach, and (No (+0)) if it does software engineering method to create secure systems based
not present any solution to deal with software security on security requirements identification upon context used,
risks. modeling, and risk analysis.
c) The study is empirical and presents relevant data for The term ‘‘Privacy-preserving Technology’’ has the
our SMS. The results and conclusions of the study next highest frequency value (26%). In this category
are strengthened by empirical evidence and it pro- ‘‘SecureUML’’ and ‘‘UMLsec’’ is the most cited (17%)
vides important and reliable information about future SSE methods, where SecureUML is an approach aim-
research and practice [76]. Yes (+1), Partially (+0.5), ing to bridge the gap amongst the security-modeling-
and No (+0). languages and design-modeling-languages, whereas UMLsec
The classification scheme and its finding are presented in works as a UML extension allowing the characteristics of
Appendix A. security-relevant to access control and confidentiality [84].
The next highest frequency value, in the study, is ‘‘Soft-
G. DATA EXTRACTION AND SYNTHESIS ware testing’’ methods (24%). In this category ‘‘Software
We have extracted the following data from each article of final static analysis’’ with a frequency 7 is the most popular soft-
selection (116 papers): ware testing methodology. Another dominant category in this
• Paper-ids and Publication Title (presented in study is ‘‘Secure requirement engineering’’ (16%). Table 3
Appendix-A) shows that 10 studies in this SMS considered the ‘‘Secu-
• Security Methods (RQ1.1, presented in Table 3 ) rity Quality Requirement Engineering (SQUARE) method’’.
• SDLC Phase discussed and covered (RQ1.2, presented For eliciting and documenting the security requirements,
in Section V, Subsection B, Figure 3) the SQUARE method is used [80].
• SWOT Analysis of Security Approaches (RQ1.4,
presented in Section V, Subsection C) B. WHICH PARTICULAR SDLC PHASE HAS BEEN MOST
• Publication Channel (RQ1.4, presented in Section V, DISCUSSED AND ADDRESSED IN THE LITERATURE?
Subsection D, Table 4, 5) This research question focus to find those studies in which
• Quality Assessment of Papers (presented in a particular software development life cycle phase covered
Appendix-A) the security aspect. According to our findings, the phase
• Paper Titles (presented in Appendix-B) of SDLC where the security aspect is considered may vary
• Paper Publication Year (presented in Appendix-A) from study to study. Figure 3 shows that 32% of studies
• Research Method used in Paper (presented in in our SMS highlighted software security methods in the
Appendix-A) design phase of the SDLC. This is typically from the fact
• Security Approach Types discussed in Paper (presented that the vulnerabilities of design level works as the major
in Appendix-A) sources of security risks in software systems [86], [87]. 50%
of software defects are usually identified and detected in
V. RESULTS AND DISCUSSION the designing phase of SDLC [81]. For developing secure
To answer the research questions presented in Section IV, software systems, the design phase of SDLC works as the
we have shared the results of this study in these sections as: basis in this regard [87]. Reducing the risks at this phase may
Appendix A shows that 78.5 percent of studies ranked minimize the efforts in other phases. It is mandatory for a
either higher than or equal to the average score of 2.5 points. software designer to properly check the aspect of security

VOLUME 9, 2021 19145

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 3. Secure software engineering methods.

19146 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 3. (Continued.) Secure software engineering methods.

Figure 3 portrays that 22% of the selected studies consid-

ered the security of software in the testing phase of SDLC.
The most lengthy, complex, and expensive phase of SDLC
is software testing [83]. Software testing is a vital activity
for increasing the quality of software development projects.
It is a core phase for software development, but to test the
programs thoroughly is not always the core subject under
software engineering education [91]. Due to this shortfall,
the software developers frequently treat the software test-
ing as a liability which finally affects the overall quality of
FIGURE 3. Frequency of software security studies in SDLC.
software. One underlying reason is that standardized testing
mechanisms are frequently perceived as boring and challeng-
during this phase and immediately make the necessary ing as compared to the creative coding phase and design
changes that will ultimately improve the overall security. activities [91].
Software security attributes play a key role in security The security testing technique is one of the most sig-
design [37]. These attributes help to enhance software secu- nificant, effective, and commonly applied measures for the
rity and also help to ensure software quality [82]. improvement of software security, used to identify the vul-
In this study, we have noted that 26% of studies consid- nerabilities and to ensure the functionality of security. For
ered software security across the whole SDLC. This fact the identification of threats that might compromise security,
is rising due to the integration of security in all phases of threat modeling is a systematic way in this regard and it
SDLC, which is a challenging issue in most organizations. has been considered a well-known accepted practice by the
We argued that the consideration of security should be an software testing industry [92].
effective input towards the other phases of SDLC i.e. from Figure 3 shows that 12% of studies in this SMS high-
requirements gathering to designing, implementation, test- lighted security approaches in the coding phase of the
ing, and deployment. For designing, building, and deploying SDLC. Deployed software is continuously under the attack of
secure software, it is essential to adapt the current software hackers exploiting vulnerabilities for decades and an increase
engineering practices and methodologies for the inclusion of is being noticed in these attacks [39], [84]. The antiviruses,
specified security-related activities. intrusion detection mechanisms, and firewalls are not enough

VOLUME 9, 2021 19147

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

to solve this problem. Each phase of SDLC must include vari- C. WHAT IS THE SWOT ANALYSIS OF SECURITY
ous suitable security defenses, analysis, and countermeasures APPROACHES IN SECURE SOFTWARE DEVELOPMENT?
that result in further secure released code [85], [86]: Each of the above approaches contributes to its benefits and
• Encrypt the files and copies. Storing data in encrypted limitations in a specific domain. In the following subsections,
form helps all output and backup versions of databases we presented the SWOT (Strength, Weakness, Opportunity,
to be secured. Data Sunrise Data Encryption is the only Threat) analysis of security approaches in details:
way to do so.
• Auditing all the servers and copies. Using so helps you 1) MODELS
see who’s been trying to get access to confidential info. a: SSD-CMM: Systems Security Engineering Capability
• Using the Intrusion Detection System (IDS) network. Maturity Model
• It is recommended that a stringent access and privilege Strength: SSD-CMM covers all phases of the software devel-
management policy be enforced and maintained. opment process.
• Don’t give client staff excessive rights and revoke Weakness: It does not guarantee good results:
expired rights in time. • Appraisal similarities.
• Secure coding is the practice of developing computer • You need to properly understand the model and how to
software in a way that guards against the accidental use it.
introduction of security vulnerabilities. • It does not remove the need for testing/evaluation.
• Secure coding standards are rules and guidelines used to • Understanding how SSE-CMM contributes to the vali-
prevent security vulnerabilities. Used effectively, these dation.
security standards prevent, detect, and eliminate errors Opportunity: SSE-CMM looks for the following security
that could compromise software security. features:
• Validate input. Validate input from all untrusted data • Operational security
sources • Information security
• Heed compiler warnings • Network security
• Architect and design for security policies • Physical security
• Keep it simple • Staff security
• Default deny • Administrative security
• Adhere to the principle of least privilege • Communication security
• Sanitize data sent to other systems • Security of meetings.
• Practice defense in depth Threats:
• Use effective quality assurance techniques • Requires great Investment
It is therefore needed from requirement-to-design-to- • Requires commitment at all levels
implementation-to-testing and deployment, software security • The need to translate PAs into the context of the organi-
necessarily is integrated throughout SDLC to provide the zation
secure and best software to the user community.
We have noted that 12 out of 116 (10%) studies in the b: Building Security In Maturity Model (BSIMM) [89]
SMS have discussed security methods in the requirement Strength: The BSIMM project provides SSDL maturity and
phase of SDLC. Nowadays, where most software in the cloud control risk
and web-based, facing a diversity of stakeholders with entan- Weakness: Not secure for all uses.
gled requirements, the development of secure software is a Opportunity: It provides an opportunity concerned with
complex task. measuring a large number of vendors to manage security risks
During software development, security is often neglected. and threats
Nowadays, the researcher’s emphasis to include the Threats: IT does nothing to guarantee that any particular
security aspects in every phase of software develop- vendor product is secure
ment during SDLC, specifically at the early phases.
Goel et al. [95], suggests Security-Requirements-Elicitation- c: Misuse case modeling [90]
and-Assessment-Mechanism (SecREAM) holds the security Strength: It focuses on security from the beginning of the
issues that appear at the start of software development. design process and provides early design decisions.
Mead [87] provides a mechanism for measuring the security Weakness:
requirements engineering process, aligned with the method • Its main weakness is that it is simple.
of Security-Quality- Requirements-Engineering (SQUARE). • It needs to be combined with powerful tools to establish
Similarly, in the maintenance phase of SDLC, we aimed an adequate project implementation plan.
to find any security-related mechanism in the software. Opportunity: It allows us to provide equal weight for opera-
Song-Kyoo [88] suggested the stochastic type maintenance tional and non-operational needs (e.g. requirements, platform
method for the security of software through the use of a closed requirements, etc.), which may not be possible with other
Queuing-model of unreliable backups. tools.

19148 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

Threats: Lack of structure and semantics. • Security Risk Tolerance

• Scalability and Dynamism
d: Secure Tropos Methodology [91] • Expensive and
Strength: This approach incorporates the requirement • Difficult Implementation
engineering concepts of need such as character, purpose,
planning, and security engineering concepts such as threat, 2) FRAMEWORKS
security issues, and security approach under a cohesive a: Trustworthy Computing Security Development Life Cycle
security protection process. or Microsoft Software Development Life Cycle (MS SDL) [93]
Weakness: Modeling IS and Business assets are Strength: MS SDL is policy compliant system that:
confusing. • Assesses security policy compliance
Opportunity: It is a way to develop security-conscious • Meets security policy
software to support the analysis and development of secure • Provides code security training
and reliable software programs. • Establishes security requirements
Threats: Do not include modeling constructs to model the • Conducts a final security review, and
attacker’s intention and the threat impacts. • Executes an incident response plan
Weakness: The process is not perfect and is still evolving.
e: McGraw’s Secure Systems Development Life
Opportunity:
Cycle (SSDLC) model [22]
• MS SDL performing dynamic analysis
Strength: SSDLC is a step-by-step method of developing
• Establish design requirements
information systems. Typical tasks include:
• Using threat modeling, and
• Budget cutting
• Approved tools
• Collecting business needs
Threats: The process may either to reach perfection or to
• Modeling and
cease evolving in the foreseeable future.
• Writing user documentation
Weakness: Documentations are expensive and time- b: KAOS: Knowledge acquisition in automated
consuming to make. It’s also hard to keep right now. specification Framework [94]
Opportunity: SSDLC provides security in developing
Strength: It offers a comprehensive and sound way of arrang-
information systems.
ing security requirements and using a graphical way of resolv-
Threats: Secure software development (SSD) activities for
ing security risks.
implementation: None
Weakness:
• One of the problems associated with KAOS with the
f: RBAC: Role-Based Access Control Model [92]
transformations between models of different approaches
Strength: Restricting the access of the system to registered
is information loss.
users. Mandatory access control (MAC) or Discretionary
• There is no graphical concrete syntax in the statement
Access Control (DAC) can be enforced.
part of the acquisition language that would represent
Weakness:
the different concepts and links supported by the meta-
• It would be difficult if you try to build a complicated
model.
role for a large organization as there will be thousands
• Deontic logic extensions are not well incorporated into
of workers with few positions that can trigger a role
the statement sublanguage to support a deeper level of
explosion.
systematic reasoning regarding agent capabilities and
• Some restrictions can be made on accessing certain
assignment of responsibilities.
device activities using RBAC, but you cannot limit
• More explicit support should also be given to coopera-
access to certain data.
tion and communication among agents.
• You can delegate permissions and privileges to user
functions, but not to operations and objects. Opportunity: In different fields such as mechanics,
Opportunity: telecommunications, healthcare, KAOS finds its applica-
• Reduce administrative work and IT support. tion and can therefore be used for any kind of information
• With RBAC, when an employee is hired or changes system.
their job, you can decrease the need for paperwork and Threats: Lack of inbuilt mechanism for evaluating secu-
password changes. rity risks.
• Maximizing operational efficiency.
• RBAC offers a streamlined approach that is logical in c: SECTET: Model-Driven Security Engineering
the definition. framework [95]
• Improving compliance. Strength: This approach allows the automated generation of
Threats: security policies and workflow used on the target framework
• Role Explosion for the configuration of security services.

VOLUME 9, 2021 19149

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

Weakness: This method does not currently use as a user Opportunity: For the creation of stable and functional
interface to write high-level security policies. systems, AEGIS offers essential resources
Opportunity: Moreover, the main benefits of this Threats: There are no SSD security assurance activities
approach are the separation of security-enhanced business and Usage in the Industry is also not reported
modeling, security modeling, code creation, and system exe-
cution. e: OCTAVE (The Operationally Critical Threat, Asset, and
Threats: Lack of high-level security policies Vulnerability Evaluation) [100]
Strength: Decisions are focused on the risks of confidential-
3) METHODS ity, the credibility of sensitive data-related assets
a: SQUARE: Security Quality Requirements Engineering Weakness: It may still be useful outside of their bound-
method [96] aries, the result cannot be guaranteed.
Strength: It integrates security considerations into the early Opportunity: Organizational risks are calculated by
stages of SDLC. OCTAVE and focus on strategic and functional aspects.
Weakness: SQUARE did not achieve the following activ- Threats: Lack of trust.
ities:
• The identification of users of the software. f: Petri nets [36]
• Detection of possible device attackers. Strength: For modeling attacks, we can use the Petri net.
• Recognition of the attacker’s interest in a piece of soft- Weakness: Petri net model has weaknesses in its inability
ware’s resources/assets. to test an unbounded position for exactly a particular marking
• Detection of the capabilities of the attacker. and to take action on the result of the test.
Opportunity: SQUARE offers a way for information tech- Opportunity: It is one of several mathematical modeling
nology systems and applications to elicits, categorize, and languages for the description of a distributed system. The
prioritize security requirements. continuity of events has become increasingly common.
Threats: Lack of identification of an illegal access Threats: Due to its concurrency, it increases the complex-
ity of operations.
b: SecureUML, and UMLsec [97]
Strength: UMLsec makes it possible to model security fea- 4) PROCESSES
tures related to confidentiality and access control. a: Aprville and Pourzandi’s Secure Software Development
Weakness: UMLsec is guided by security criteria, how- Life Cycle Process [55]
ever, it does not mean to model them explicitly.
Strength: This process provides:
Opportunity: The goal of SecureUML is to bridge the gap
• Identification of high-level security
between languages for security modeling and languages for
• Modeling of threats
design modeling.
• Specification of security requirements, and
Threats: Only permitted activities were executed.
• Risks analysis
There is no required unique restriction language in
Weakness: Usage in the Industry: Not reported
UMLsec.
Opportunity:
• Avoiding vulnerabilities of buffer overflow and a format
c: CORAS: Combine methods for risk analysis and
string
semiformal description methods [98]
• Usage of established security algorithms
Strength: For modeling risks and unwanted actions,
• Usage of secure programming language
the CORAS profile provides advanced case diagrams for
Threats: This process mostly does not use the artifacts of
usages.
SSD activity In the later stage of software development.
Weakness: CORAS just draw threat scenarios/diagram.
Opportunity: It encourages engagement and cooperation
b: Software Security Assessment Instrument(SSAI) [101]
between various groups of stakeholders involved in a risk
assessment. Strength: SSAI detects vulnerabilities and also provide mit-
Threats: CORA’s threat modeling tool has no risks/threats igation practices for it.
analysis facilities. Weakness: No SSD activities for design, implementation.
Usage in the industry is none.
d: AEGIS: Appropriate and effective guidance for Opportunity:
information security [99] • SSAI mitigate vulnerabilities
Strength: Identification of properties, abuse cases, opera- • It provides a list of scanning tools for the security code.
tional context, and support for simulation of security require- • It offers potential security checklist items to guide soft-
ments. ware development.
Weakness: Secure software development (SSD) activities Threats: This process also mostly does not use the artifacts
for implementation: None of SSD activity In the later stage of software development.

19150 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 4. Articles by their journal and conference rank.

TABLE 5. Top publication venues of identified articles.

D. WHAT ARE THE POPULAR VENUES FOR SECURE VI. IMPLICATIONS FOR RESEARCH AND PRACTICE
SOFTWARE DEVELOPMENT? The findings of this research contribute to the SSD context
Table 4 presents information about the total score of from various perspectives. Initially, they provide a clearer
the selected articles based upon their considerations from understanding of the security practices and methods used
Computing Research Education (CORE)∗ 2018 and Journal in the SSD environment for the academic community and
Citation Reports (JCR)∗∗ 2018 lists. explain differences in areas where security approaches are
The table data shows that 64% of the selected papers in ineffective or inadequate. This knowledge will open opportu-
our study are published in Q1-Q4 journals and Core A∗ -C nities for SSD researchers to solve problems and design new
conferences. The remaining 36% of articles are published security techniques that can minimize the impact of security
in well-reputed conferences and journals; however, some of risks and threats in the development of secure software. Sec-
them have no ranking in the CORE 2018 and JCR lists. The ondly, the development of applications participating in highly
detailed score for each of the selected studies is available in distributed projects would provide an overview of current
Appendix B. approaches aimed at promoting their efforts and recognizing
Table 5 shows the distribution of selected studies based those most likely to meet their needs. The guidelines taken
on publication types. Journals, conferences, workshops, and from the conclusions of this SMS for researchers and practi-
symposiums are the four main publication types with 43.9% tioners are as follows:
(57 studies), 36.2% (42 studies), 7.7% (9 studies), and • Practitioners, such as software developers and project
6.9% (8 studies) of the selected studies respectively. Overall, managers, are eager to learn about the latest research
116 publications’ venues are identified that cover various on the subject, along with researchers interested in
areas of computer science; such as software engineering, SSD project management. They need to study the pub-
software security, and networking, etc. It means that such lished articles more from the ‘‘Information and Software
a domain has received large attention from the researchers. Technology’’, ‘‘Software System Model’’, ‘‘Journal of
We observed that ‘‘Information and Software Technology’’, System and Software’’, ‘‘Computer Standards and Inter-
‘‘Software System Model’’, ‘‘Journal of System and Soft- faces’’, and ‘‘Security and Communication Networks’’
ware’’, ‘‘Computer Standards and Interfaces’’, and ‘‘Security journals. Similarly, it will be more beneficial for them
and Communication Networks’’ are the most recurring publi- to read the published articles from the proceedings
cation venues for security approaches for SSD topic. We also of the ‘‘International Conference on Availability, Reli-
explored that ‘‘International Conference on Availability, Reli- ability, and Security’’ and ‘‘International Conference
ability, and Security’’ and ‘‘International Conference on Soft- on Software Engineering (ICSE)’’ and its affiliated
ware Engineering (ICSE)’’ is the most frequent source of workshops and symposiums. In short, we can say
worthy publications to the topic concerned. It demonstrates that the above venues are important sources to study
the importance of software security research in software engi- the security approaches in SSD. Similarly, researchers
neering and other related fields. are encouraged to submit their high-quality research

VOLUME 9, 2021 19151

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 6. Classification of selected studies based on data extraction.

19152 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 6. (Continued.) Classification of selected studies based on data extraction.

VOLUME 9, 2021 19153

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 6. (Continued.) Classification of selected studies based on data extraction.

papers to these journals and conferences in the SSD such approaches to care about the security in SSD. While
domain. working in this domain, we put a suggestion that practi-
• The outcome of this SMS suggests a lack of empirical tioners might cooperate with researchers in applying the
evidence of the different suggested solutions. In the security approaches in their practice.
study, only 30.12% of the security approaches have been • The SSD domain is still immature and sufficient
empirically validated. research work is required, as this area has not been
Researchers are encouraged to review their security meth- reached a solution-oriented stage, particularly on empir-
ods using experiments, case studies, and ordinary litera- ically evaluated solutions.
ture review on the first choice of research methods to
obtain stronger qualitative and quantitative outcomes and VII. THREATS TO VALIDITY
to build approaches that can fulfill the demands of SSD The validity of the study is concerned with its findings being
projects. trustworthy. The limitations are listed as follows for this SMS:
• As shown in Appendix A and B, many approaches are • Construct Validity
intended to enhance the security in SSD. Practitioners, The construct validity refers to the degree of the analysis
particularly software developers and designers may use to which the operational measures represent what the RQs

19154 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 7. Titles of the final selected papers.

VOLUME 9, 2021 19155

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 7. (Continued.) Titles of the final selected papers.

19156 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

TABLE 7. (Continued.) Titles of the final selected papers.

expect to respond. Two variables can be easily defined in this VIII. CONCLUSION AND FUTURE WORK
SMS as a challenge to the construct validity. The research At each level of SDLC, incorporating software protection is
string used in the study is one, while the digital libraries now a primary need for secure software development. SSE
checked are the other. We conducted a systematic search advocates that the security of software is an essential factor
using a wide variety of words in the sample to further expand that should be evaluated during the early stages of SDLC [5].
the research scope. For the validity to include the maximum We need to incorporate security features into our application
relevant literature, the keywords of the study were included development life cycle and adapt the latest SSE approaches
after detailed discussion and recommendations by the two to build and deploy a secure software system [6], [7].
authors. The choice of digital libraries for the studies was We conclude from the above discussion that it is not good
another risk to construct validity. The identifying of five enough to secure the software system in post-development
digital libraries as the key sources in such a domain mitigated phases and there is a dire need to figure out better ways and
this risk. means to secure the software system. It is our conviction
• Internal Validity that this paper will contribute by offering a taxonomy for the
In internal validity, the classification and decision to allo- SSE-related research subject and will provide a breakdown
cate a specific security approach to a particular category of papers in the same field on each topic. The taxonomical
and method of software development may be evaluated. The and demographical information on a particular research topic,
classification was suggested to minimize this effect and the presented in this paper, will invite more opportunities to
categorization process was performed by the two authors, promote further research in this field.
while the final results were checked by the third one. More- We investigate the outcome of this mapping study to
over, to allow the conclusions drawn from the results to provide a reference direction for interested and vigilant
be replicated, the steps and activities in this scheme have researchers to address and explore new research trends and
been clearly defined. The internal validity risks have been gaps and the subject domain. This study will provide a deep
minimized such that all interested readers are encouraged to insight into the subject domain that could enrich and more
freely see the data retrieved from the papers of the studies update the software practitioners about security approaches,
displayed. security limitations, and unresolved specific and general
• Conclusion Validity issues. After the final selection and analysis of 116 arti-
The validity of the conclusion is the degree to which the cles were selected and these were classified based on secu-
assumptions made about the relationships are rational and rity approaches according to the quality assessment, soft-
concerned with the potential to reproduce these results. The ware security method, SDLC phase has been most discussed
possibility of conclusion validity was a factor in such SMS and addressed, publication venue, and SWOT analysis. The
that could lead to the conclusion of incorrect findings of a results indicate that this domain is still immature and suffi-
relationship in the data observed. Each step of the data col- cient research work needs to be carried out particularly on
lection, extraction, and analysis was validated by a systematic empirically evaluated solutions.
process and periodic reviews carried out by the participating Also, we plan to develop a Software Security Assur-
researchers to reduce this threat. The explanation behind this ance Model (SSAM) [18] to assist vendor organizations
move was that for similar research, the same procedure was to assess their readiness for secure software development.
done in the literature. We will develop the SSAM model by using the output of
• External Validity future RQs, supervisor inputs, and guidance from existing
External validity includes how much it is possible to gen- studies [102]–[111]. The model will produce numerous eval-
eralize the outcomes of a study. To diminish this threat, uation reports, such as a list of security measurements and
the ratio of security models, frameworks, methods, processes, their solutions to be used in each phase of the SDLC by
guidelines, metrics, and tools have been included in this work. GSD vendor organizations. Our main aim is to answer the

VOLUME 9, 2021 19157

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

following research questions (RQs) in the future to achieve [13] D. Mellado, C. Blanco, L. E. Sánchez, and E. Fernández-Medina, ‘‘A sys-
the above-mentioned goals: tematic review of security requirements engineering,’’ Comput. Standards
Interfaces, vol. 32, no. 4, pp. 153–165, Jun. 2010.
RQ1: What are the security threats to the development of [14] I. Velásquez, A. Caro, and A. Rodríguez, ‘‘Authentication schemes and
secure software products, as described in the literature and methods: A systematic literature review,’’ Inf. Softw. Technol., vol. 94,
industrial survey, to be avoided by GSD vendor organiza- pp. 30–37, Feb. 2018.
[15] Y. Lee and G. Lee, ‘‘HW-CDI: Hard-wired control data integrity,’’ IEEE
tions? Access, vol. 7, pp. 10811–10822, 2019.
RQ2: What are the practices to be implemented by GSD [16] Z. A. Maher, H. Shaikh, M. S. Khan, A. Arbaaeen, and A. Shah, ‘‘Factors
vendor organizations, as defined in the literature and indus- affecting secure software development practices among developers—An
investigation,’’ in Proc. ICETAS, 2018, pp. 1–6.
trial survey, to build secure software products? [17] G. E. Rodríguez, J. G. Torres, P. Flores, and D. E. Benavides, ‘‘Cross-
RQ3: Is the proposed SSAM model practically robust in site scripting (XSS) attacks and mitigation: A survey,’’ Comput. Netw.,
vol. 166, Jan. 2020, Art. no. 106960.
helping GSD vendor organizations in assessing their readi- [18] R. A. Khan and S. U. Khan, ‘‘A preliminary structure of software
ness to build secure software? security assurance model,’’ in Proc. 13th Int. Conf. Global Softw. Eng.,
Gothenburg, Sweden, 2018, pp. 137–140.
[19] K. S. U. Khan, R. Ahmad, and I. M. Yazid, ‘‘Systematic mapping study
APPENDIX A protocol for secure software engineering,’’ in Proc. Asia Int. Multidisci-
See Table 6. plinary Conf. (AIMC), 2019, pp. 367–374.
[20] G. McGraw, ‘‘From the ground up: The DIMACS software security
workshop,’’ IEEE Secur. Privacy, vol. 1, no. 2, pp. 59–66, Mar. 2003.
APPENDIX B [21] G. Hatzivasilis, I. Papaefstathiou, and C. Manifavas, ‘‘Software security,
See Table 7. privacy, and dependability: Metrics and measurement,’’ IEEE Softw.,
vol. 33, no. 4, pp. 46–54, Jul. 2016.
[22] G. McGraw, ‘‘Software security,’’ IEEE Secur. Privacy, vol. 2, no. 2,
ACKNOWLEDGMENT pp. 80–83, Aug. 2004.
The findings achieved herein are solely the responsibility [23] R. M. Parizi, K. Qian, H. Shahriar, F. Wu, and L. Tao, ‘‘Benchmark
of the authors. In addition to this, all members of the Soft- requirements for assessing software security vulnerability testing tools,’’
in Proc. COMPSAC, 2018, pp. 825–826.
ware Engineering Research Group, University of Malakand [24] N. M. Mohammed, M. Niazi, M. Alshayeb, and S. Mahmood, ‘‘Explor-
(SERG_UOM) are obliged to do so in general and both ing software security approaches in software development lifecycle:
Dr. Nasir Rashid and Ghulam Murtaza in particular, for their A systematic mapping study,’’ Comput. Standards Interfaces, vol. 50,
pp. 107–115, Feb. 2017.
reviews and suggested improvements. [25] N. R. Mead and G. McGraw, ‘‘A portal for software security,’’ IEEE Secur.
Privacy Mag., vol. 3, no. 4, pp. 75–79, Jul. 2005.
REFERENCES [26] G. Hoglund, and G. McGraw, Exploiting Software. Boston, MA, USA:
Addison-Wesley, 2004, pp. 1–44.
[1] J. C. S. Núñez, A. C. Lindo, and P. G. Rodríguez, ‘‘A preventive secure [27] D. Verdon and G. McGraw, ‘‘Risk analysis in software design,’’ IEEE
software development model for a software factory: A case study,’’ IEEE Secur. Privacy Mag., vol. 2, no. 4, pp. 79–84, Jul. 2004.
Access, vol. 8, pp. 77653–77665, 2020. [28] N. S. A. Karim, A. Albuolayan, T. Saba, and A. Rehman, ‘‘The prac-
[2] S. V. Solms and L. A. Futcher, ‘‘Adaption of a secure software develop- tice of secure software development in SDLC: An investigation through
ment methodology for secure engineering design,’’ IEEE Access, vol. 8, existing model and a case study,’’ Secur. Commun. Netw., vol. 9, no. 18,
pp. 125630–125637, 2020. pp. 5333–5345, Dec. 2016.
[3] M. Z. Gunduz and R. Das, ‘‘Cyber-security on smart grid: Threats [29] Y. Mufti, M. Niazi, M. Alshayeb, and S. Mahmood, ‘‘A readiness
and potential solutions,’’ Comput. Netw., vol. 169, Mar. 2020, model for security requirements engineering,’’ IEEE Access, vol. 6,
Art. no. 107094. pp. 28611–28631, 2018.
[4] E. K. Szczepaniuk, H. Szczepaniuk, T. Rokicki, and B. Klepacki, ‘‘Infor- [30] M. Ammar, G. Russello, and B. Crispo, ‘‘Internet of Things: A survey on
mation security assessment in public administration,’’ Comput. Secur., the security of IoT frameworks,’’ J. Inf. Secur. Appl., vol. 38, pp. 8–27,
vol. 90, Mar. 2020, Art. no. 101709. Feb. 2018.
[5] L. Bracciale, P. Loreti, A. Detti, R. Paolillo, and N. B. Melazzi, [31] X. Hu, Y. Zhuang, Z. Cao, T. Ye, and M. Li, ‘‘Modeling and validation for
‘‘Lightweight named object: An ICN-based abstraction for IoT device embedded software confidentiality and integrity,’’ in Proc. 12th Int. Conf.
programming and management,’’ IEEE Internet Things J., vol. 6, no. 3, Intell. Syst. Knowl. Eng. (ISKE), Nanjing, China, Nov. 2017, pp. 1–6.
pp. 5029–5039, Jun. 2019. [32] M. Khari, Vaishali, and P. Kumar, ‘‘Embedding security in software
[6] M. Zhang, X. D. C. D. Carnavalet, L. Wang, and A. Ragab, ‘‘Large-scale development life cycle (SDLC),’’ in Proc. Int. Conf. Comput. Sustain.
empirical study of important features indicative of discovered vulnerabil- Global Develop. (INDIACom), 2016, pp. 2182–2186. [Online]. Available:
ities to assess application security,’’ IEEE Trans. Inf. Forensics Security, https://ieeexplore.ieee.org/document/7724651
vol. 14, no. 9, pp. 2315–2330, Sep. 2019. [33] H. Xu, J. Heijmans, and J. Visser, ‘‘A practical model for rating software
[7] G. McGraw, ‘‘Six tech trends impacting software security,’’ Computer, security,’’ in Proc. IEEE 7th Int. Conf. Softw. Secur. Rel. Companion,
vol. 50, no. 5, pp. 100–102, May 2017. Jun. 2013, pp. 231–232.
[8] J. Li, Y. Zhang, X. Chen, and Y. Xiang, ‘‘Secure attribute-based data [34] B. B. Mayvan, A. Rasoolzadegan, and Z. G. Yazdi, ‘‘The state of the art on
sharing for resource-limited users in cloud computing,’’ Comput. Secur., design patterns: A systematic mapping of the literature,’’ J. Syst. Softw.,
vol. 72, pp. 1–12, Jan. 2018. vol. 125, pp. 93–118, Mar. 2017.
[9] A. Sharma and M. P. Kumar, ‘‘Aspects of enhancing security in software [35] S. P. Kadam and S. Joshi, ‘‘Secure by design approach to improve security
development life cycle,’’ Adv. Comput. Sci. Technol., vol. 10, no. 2, of object oriented software,’’ in Proc. INDIACom, Mar. 2015, pp. 24–30.
pp. 203–210, 2017. [36] D. Xu and K. E. Nygard, ‘‘Threat-driven modeling and verification of
[10] M. Essafi, L. Labed, and H. B. Ghezala, ‘‘Towards a comprehensive view secure software using aspect-oriented Petri nets,’’ IEEE Trans. Softw.
of secure software engineering,’’ in Proc. Int. Conf. Emerg. Secur. Inf., Eng., vol. 32, no. 4, pp. 265–278, Apr. 2006.
Syst., Technol., 2007, pp. 181–186. [37] R. Kumar, S. A. Khan, and R. A. Khan, ‘‘Analytical network process for
[11] R. Syed, M. Rahafrooz, and J. M. Keisler, ‘‘What it takes to get retweeted: software security: A design perspective,’’ CSI Trans. ICT, vol. 4, nos. 2–4,
An analysis of software vulnerability messages,’’ Comput. Hum. Behav., pp. 255–258, Dec. 2016.
vol. 80, pp. 207–215, Mar. 2018. [38] S. Janakiraman, K. Thenmozhi, J. B. B. Rayappan, and R. Amirtharajan,
[12] A. K. Srivastava and S. Kumar, ‘‘An effective computational technique for ‘‘Lightweight chaotic image encryption algorithm for real-time embed-
taxonomic position of security vulnerability in software development,’’ ded system: Implementation and analysis on 32-bit microcontroller,’’
J. Comput. Sci., vol. 25, pp. 388–396, Mar. 2018. Microprocessors Microsyst., vol. 56, pp. 1–12, Feb. 2018.

19158 VOLUME 9, 2021

R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

[39] T. Diamantopoulos, K. Thomopoulos, and A. Symeonidis, ‘‘QualBoa: [63] P. Silva, R. Noël, M. Gallego, S. Matalonga, and A. Hernan, ‘‘Software
Reusability-aware recommendations of source code components,’’ in development initiatives to identify and mitigate security threats—A sys-
Proc. MSR, 2016, pp. 488–491. tematic mapping,’’ in Proc. CibSE, 2016, pp. 1–15.
[40] Y.-H. Tung, S.-C. Lo, J.-F. Shih, and H.-F. Lin, ‘‘An integrated security [64] A. S. Guinea, G. Nain, and Y. Le Traon, ‘‘A systematic review on the
testing framework for secure software development life cycle,’’ in Proc. engineering of software for ubiquitous systems,’’ J. Syst. Softw., vol. 118,
18th Asia–Pacific Netw. Oper. Manage. Symp. (APNOMS), Oct. 2016, pp. 251–276, Aug. 2016.
pp. 1–4. [65] K. Petersen, S. Vakkalanka, and L. Kuzniarz, ‘‘Guidelines for conducting
[41] B. S. Clegg, J. M. Rojas, and G. Fraser, ‘‘Teaching software testing systematic mapping studies in software engineering: An update,’’ Inf.
concepts using a mutation testing game,’’ in Proc. ICSE-SEET, 2017, Softw. Technol., vol. 64, pp. 1–18, Aug. 2015.
pp. 33–36. [66] A. K. Barbara, B. D. Budgen, and O. P. Brereton, ‘‘Using mapping studies
[42] M. Felderer and F. Elizabeta, ‘‘A systematic classification of security as the basis for further research—A participant-observer case study,’’ Inf.
regression testing approaches,’’ Int. J. Softw. Tools Technol. Transf., Softw. Technol., vol. 53, no. 6, pp. 638–651, 2011.
vol. 17, no. 3, pp. 305–319, Jun. 2015. [67] B. Kitchenham, O. P. Brereton, D. Budgen, M. Turner, J. Bailey, and
[43] B. Potter and G. McGraw, ‘‘Software security testing,’’ IEEE Secur. S. Linkman, ‘‘Systematic literature reviews in software engineering—A
Privacy, vol. 2, no. 5, pp. 81–85, Sep. 2004. systematic literature review,’’ Inf. Softw. Technol., vol. 51, no. 1, pp. 7–15,
[44] J.-P. Arcangeli, R. Boujbel, and S. Leriche, ‘‘Automatic deployment of 2009.
distributed software systems: Definitions and state of the art,’’ J. Syst. [68] J. Morán, C. Riva, and J. Tuya, ‘‘Testing MapReduce programs: A sys-
Softw., vol. 103, pp. 198–218, May 2015. tematic mapping study,’’ J. Softw. Evol. Process, vol. 31, no. 3, p. e2120,
[45] R. Colomo-Palacios, E. Fernandes, P. Soto-Acosta, and X. Larrucea, Mar. 2019.
‘‘A case analysis of enabling continuous software deployment through [69] R. E. Lopez-Herrejon, S. Illescas, and A. Egyed, ‘‘A systematic mapping
knowledge management,’’ Int. J. Inf. Manage., vol. 40, pp. 186–189, study of information visualization for software product line engineering,’’
Jun. 2018. J. Softw. Evol. Process, vol. 30, no. 2, p. e1912, Feb. 2018.
[46] S. Velmourougan, P. Dhavachelvan, R. Baskaran, and B. Ravikumar, [70] M. Felderer and J. Carver, ‘‘Guidelines for systematic mapping stud-
‘‘Software development life cycle model to improve maintainability of ies in security engineering,’’ in Empirical Research for Software Secu-
software applications,’’ in Proc. 4th Int. Conf. Adv. Comput. Commun., rity: Foundations and Experience. Boca Raton, FL, USA: CRC Press,
Aug. 2014, pp. 270–273. Jan. 2018.
[47] V. Midha and A. Bhattacherjee, ‘‘Governance practices and software [71] D. Budgen, M. Turner, P. Brereton, and B. Kitchenham, ‘‘Using mapping
maintenance: A study of open source projects,’’ Decis. Support Syst., studies in software engineering,’’ in Proc. PPIG, 2008, pp. 195–204.
vol. 54, no. 1, pp. 23–32, Dec. 2012. [72] S. Y. Chadli, A. Idri, J. N. Ros, J. L. Fernández-Alemán, J. M. C. de Gea,
[48] V. A. Uzunov, B. E. Fernández, and F. Katrina, ‘‘Engineering security into and A. Toval, ‘‘Software project management tools in global software
distributed systems: A survey of methodologies,’’ J. Universal Comput. development: A systematic mapping study,’’ SpringerPlus, vol. 5, no. 1,
Sci., vol. 18, no. 20, pp. 2920–3006, 2012. p. 2006, Nov. 2016.
[49] Z. Aslanyan, F. Nielson, and D. Parker, ‘‘Quantitative verification and [73] M. El Bajta, A. Idri, J. Nicolás, J. Fernández-Alemán, and A. Toval,
synthesis of attack-defence scenarios,’’ in Proc. IEEE 29th Comput. ‘‘Software project management approaches for global software devel-
Secur. Found. Symp. (CSF), Lisbon, Portugal, Jun. 2016, pp. 105–119. opment: A systematic mapping study,’’ Tsinghua Sci. Technol., vol. 23,
[50] T. H. Nguyen, J. Grundy, and M. Almorsy, ‘‘Integrating goal-oriented no. 6, pp. 690–714, Dec. 2018.
[74] C. Wang, M. Daneva, M. Sinderen, and P. Liang, ‘‘A systematic mapping
and use case-based requirements engineering: The missing link,’’ in Proc.
study on crowdsourced requirements engineering using user feedback,’’
ACM/IEEE 18th Int. Conf. Model Driven Eng. Lang. Syst. (MODELS),
J. Softw. Evol. Process, vol. 31, no. 10, p. e2199, Oct. 2019.
Ottawa, ON, Canada, Sep. 2015, pp. 328–337.
[75] A. Manjavacas, A. Vizcaíno, F. Ruiz, and M. Piattini, ‘‘Global software
[51] V. G. Vassilakis, H. Mouratidis, E. Panaousis, I. D. Moscholios, and
development governance: Challenges and solutions,’’ J. Softw. Evol. Pro-
M. D. Logothetis, ‘‘Security requirements modelling for virtualized 5G
cess, vol. 32, no. 10, p. e226, Oct. 2020.
small cell networks,’’ in Proc. 24th Int. Conf. Telecommun. (ICT),
[76] D. Šmite, C. Wohlin, T. Gorschek, and R. Feldt, ‘‘Empirical evidence
Limassol, Cyprus, May 2017, pp. 337–362.
in global software engineering: A systematic review,’’ Empirical Softw.
[52] F.-B. Eduardo, Security Patterns in Practice: Designing Secure Architec-
Eng., vol. 15, no. 1, pp. 91–118, Feb. 2010.
tures Using Software Patterns, 1st ed. Hoboken, NJ, USA: Wiley, 2013. [77] P. Brereton, B. A. Kitchenham, D. Budgen, M. Turner, and M. Khalil,
[53] S. Lipner and H. Michael. The Trustworthy Computing Security
‘‘Lessons from applying the systematic literature review process within
Development Lifecycle. Accessed: Mar. 3, 2019. [Online]. Available:
the software engineering domain,’’ J. Syst. Softw., vol. 80, no. 4,
https://msdn.microsoft.com/en-us/library/ms995349.aspx
pp. 571–583, Apr. 2007.
[54] D. Mellado, E. Fernandez-Medina, and M. A. Piattini, ‘‘Common criteria [78] M. S. Lund, B. Solhaug, and K. Stolen, Model-Driven Risk Analysis: The
based security requirements engineering process for the development CORAS Approach. Berlin, Germany: Springer, 2011, pp. 1–79.
of secure information systems,’’ Comput. Standards Interfaces, vol. 29, [79] I. Flechais, M. A. Sasse, and S. M. V. Hailes, ‘‘Bringing security home:
no. 2, pp. 244–253, 2007. A process for developing secure and usable systems,’’ in Proc. NSPW,
[55] A. Apvrille and M. Pourzandi, ‘‘Secure software development by exam- 2003, pp. 49–57.
ple,’’ IEEE Secur. Privacy Mag., vol. 3, no. 4, pp. 10–17, Jul. 2005. [80] N. R. Mead, Identifying Security Requirements Using the Security Quality
[56] J. D. M. Charles, B. Haley, and B. Nuseibeh, ‘‘Core security require- Requirements Engineering (SQUARE) Method (Integrating Security and
ments artifacts,’’ Dept. Comput., Fac. Math. Comput., Open Univ., Software Engineering). Pittsburgh, PA, USA: Carnegie Mellon Univ.,
Milton Keynes, U.K., Tech. Rep. 2004/23, 2004. [Online]. Available: Dec. 2005, pp. 1–21, doi: 10.4018/9781599041476.ch003.
http://computing.open.ac.uk [81] V. Maheshwari and M. Prasana, ‘‘Integrating risk assessment and threat
[57] G. Dan. Build Security in Introduction to the CLASP Process. [Online]. modeling within SDLC process,’’ in Proc. ICICT, 2016, pp. 1–5.
Available: http://buildsecurityin.us-cert.gov/daisy/bsi/articles/best- [82] S. Moyo and E. Mnkandla, ‘‘A novel lightweight solo software devel-
practices/requirements/548.html opment methodology with optimum security practices,’’ IEEE Access,
[58] C. B. Haley, R. Laney, J. Moffett, and B. Nuseibeh, ‘‘Security require- vol. 8, pp. 33735–33747, 2020.
ments engineering: A framework for representation and analysis,’’ IEEE [83] S. U. R. Khan, I. U. Rehman, and S. U. R. Malik, ‘‘The impact of test case
Trans. Softw. Eng., vol. 34, no. 1, pp. 133–152, Jan./Feb. 2008. reduction and prioritization on software testing effectiveness,’’ in Proc.
[59] N. R. Mead, E. D. Hough, and T. R. Stehney, II, ‘‘Security quality Int. Conf. Emerg. Technol., 2009, pp. 416–421.
requirements engineering (SQUARE) methodology,’’ Softw. Eng. Inst., [84] R. Kumar, S. K. Pandey, and S. I. Ahson, ‘‘Security in coding phase of
Carnegie Mellon Univ., Tech. Rep. CMU/SEI-2005-TR-009, 2005. SDLC,’’ in Proc. 3rd Int. Conf. Wireless Commun. Sensor Netw., 2007,
[60] J. Manico, OWASP, Application Security Verification Standard 3.0.1, pp. 118–120.
2016, pp. 1–70. [85] R. C. Seacord, Secure Coding in C and C++. Reading, MA, USA:
[61] Information Technology—Security Techniques—Information Security Addison-Wesley, 2013.
Management Systems—Requirements, document ISO/IEC 27001:2005, [86] A. Mousa, M. Karabatak, and T. Mustafa, ‘‘Database security threats and
International Organization for Standardization, 2005, pp. 1–34. challenges,’’ in Proc. ISDFS, 2020, pp. 1–5.
[62] C. Thompson, M. Shelton, E. Stark, M. Walker, E. Schechter, and [87] N. R. Mead, ‘‘Measuring the software security requirements engineering
A. P. Felt, ‘‘The Web’s identity crisis: Understanding the effectiveness of process,’’ in Proc. IEEE 36th Annu. Comput. Softw. Appl. Conf. Work-
website identity indicators,’’ in Proc. USENIX, 2019, pp. 1715–1731. shops, Jul. 2012, pp. 583–588.

VOLUME 9, 2021 19159

 R. A. Khan et al.: Systematic Mapping Study on Security Approaches in SSE

[88] K. Song-Kyoo, ‘‘Design of enhanced software protection architecture by RAFIQ AHMAD KHAN received the M.Phil.
using theory of inventive problem solving,’’ in Proc. IEEE Int. Conf. Ind. degree in computer science with a specialization
Eng. Eng. Manage., Dec. 2009, pp. 978–982. in software engineering from the University of
[89] BSIMM: Building Security In Maturity Model. Accessed: Mar. 5, 2019. Malakand, Pakistan, under the research supervi-
[Online]. Available: https://www.bsimm.com/ sion of Dr. S. U. Khan, where he is currently
[90] G. Sindre and A. L. Opdahl, ‘‘Eliciting security requirements with misuse
pursuing the Ph.D. degree.
cases,’’ Requirements Eng., vol. 10, no. 1, pp. 34–44, Jan. 2005.
[91] H. Mouratidis, P. Giorgini, and G. Manson, ‘‘When security meets soft- His research interests include software secu-
ware engineering: A case of modeling secure information systems,’’ J. Inf. rity, empirical software engineering, systematic
Syst., vol. 30, no. 8, pp. 609–629, 2005. literature review, requirements engineering, green
[92] D. Basin, J. Doser, and T. Lodderstedt, ‘‘Model driven security: From computing, software testing, agile software devel-
UML models to access control infrastructures,’’ ACM Trans. Softw. Eng. opment, and global software engineering. He has authored several papers in
Methodol., vol. 15, no. 1, pp. 39–91, 2006. well-reputed international conferences and journals, including ICGSE and
[93] S. Lipner, ‘‘The trustworthy computing security development lifecy- IEEE ACCESS.
cle,’’ in Proc. 20th Annu. Comput. Secur. Appl. Conf. (ACSAC), 2004,
pp. 1–12.
[94] A. van Lamsweerde and E. Letier, ‘‘Handling obstacles in goal-oriented
requirements engineering,’’ IEEE Trans. Softw. Eng., vol. 26, no. 10,
pp. 978–1005, 2000.
[95] M. Alam, J.-P. Seifert, and X. Zhang, ‘‘A model-driven framework for
trusted computing based systems,’’ in Proc. EDOC, 2007, p. 75.
[96] N. R. Mead and T. Stehney, ‘‘Security quality requirements engineering SIFFAT ULLAH KHAN received the Ph.D. degree
(SQUARE) methodology,’’ ACM SIGSOFT Softw. Eng. Notes, vol. 30, in computer science from Keele University, U.K.,
no. 4, pp. 1–7, Jul. 2005. in 2011.
[97] T. Lodderstedt, D. Basin, and J. Doser, ‘‘SecureUML: A UML-based He was the Head of the Department of Software
modeling language for model-driven security,’’ in Proc. Int. Conf. Unified Engineering, University of Malakand, Pakistan,
Modeling Lang., 2002 pp. 426–441. for three years, where he was also the Chairman of
[98] M. S. Lund, B. Solhaug, and K. Stølen, Model-Driven Risk Analysis—The the Department of Computer Science and IT and is
CORAS Approach. Berlin, Germany: Springer, 2011, pp. 1–358.
currently an Associate Professor in Computer Sci-
[99] I. Flechais, C. Mascolo, and M. A. Sasse, ‘‘Integrating security and
usability into the requirements and design process,’’ Int. J. Electron. ence. He is also the Founder and the Leader of the
Secur. Digit. Forensic, vol. 1, no. 1, pp. 12–26, 2007. Software Engineering Research Group, University
[100] C. J. Alberts, A. J. Dorofee, J. F. Stevens, and C. Woody, Introduction of Malakand. He has successfully supervised ten M.Phil. and four Ph.D.
to the OCTAVE Approach. Pittsburgh, PA, USA: Carnegie Mellon Univ., scholars. He has authored over 100 papers, so far, in well-reputed inter-
Software Engineering Institute, 2003, pp. 1–37. national conferences and journals. His research interests include software
[101] D. P. Gilliam, T. L. Wolfe, J. S. Sherif, and M. Bishop, ‘‘Software outsourcing, empirical software engineering, agile software development,
security checklist for the software life cycle,’’ in Proc. WET ICE, 2003, systematic literature review, software metrics, cloud computing, require-
pp. 243–248. ments engineering, and green computing/IT. He received the Gold Medal
[102] W. Ren, O. Ma, H. Ji, and X. Liu, ‘‘Human posture recognition using a (Dr. M. N. Azam Prize 2015) from the Pakistan Academy of Sciences in
hybrid of fuzzy logic and machine learning approaches,’’ IEEE Access,
recognition of his research achievements in the field of computer (software).
vol. 8, pp. 135628–135639, 2020.
[103] A. Agrawal, M. Alenezi, R. Kumar, and R. A. Khan, ‘‘Measuring
the sustainable-security of Web applications through a fuzzy-based
integrated approach of AHP and TOPSIS,’’ IEEE Access, vol. 7,
pp. 153936–153951, 2019.
[104] F. A. Al-Zahrani, ‘‘Evaluating the usable-security of healthcare software
through unified technique of fuzzy logic, ANP and TOPSIS,’’ IEEE
Access, vol. 8, pp. 109905–109916, 2020. HABIB ULLAH KHAN received the Ph.D. degree
[105] R. Kumar, A. I. Khan, Y. B. Abushark, M. M. Alam, A. Agrawal, and in management information systems from Leeds
R. A. Khan, ‘‘An integrated approach of fuzzy logic, AHP and TOPSIS Beckett University, U.K. He is an Associate
for estimating usable-security of Web applications,’’ IEEE Access, vol. 8, Professor of MIS with the Department of Account-
pp. 50944–50957, 2020. ing and Information Systems, College of Busi-
[106] R. Kumar, A. I. Khan, Y. B. Abushark, M. M. Alam, A. Agrawal, and
ness and Economics, Qatar University, Qatar.
R. A. Khan, ‘‘A knowledge-based integrated system of hesitant fuzzy
He has nearly 20 years of industry, teach-
set, AHP and TOPSIS for evaluating security-durability of Web appli-
cations,’’ IEEE Access, vol. 8, pp. 48870–48885, 2020. ing, and research experience. His research inter-
[107] M. Alenezi, A. Agrawal, R. Kumar, and R. A. Khan, ‘‘Evaluating perfor- ests include IT adoption, social media, Internet
mance of Web application security through a fuzzy based hybrid multi- addiction, mobile commerce, computer mediated
criteria decision-making approach: Design tactics perspective,’’ IEEE communication, IT outsourcing, big data, and IT security.
Access, vol. 8, pp. 25543–25556, 2020.
[108] R. Kumar, A. Baz, H. Alhakami, W. Alhakami, M. Baz, A. Agrawal,
and R. A. Khan, ‘‘A hybrid model of hesitant fuzzy decision-making
analysis for estimating usable-security of software,’’ IEEE Access, vol. 8,
pp. 72694–72712, 2020.
[109] A. K. Pandey, A. I. Khan, Y. B. Abushark, M. M. Alam, A. Agrawal,
R. Kumar, and R. A. Khan, ‘‘Key issues in healthcare data integrity: MUHAMMAD ILYAS received the Ph.D. degree
Analysis and recommendations,’’ IEEE Access, vol. 8, pp. 40612–40628,
in computer science from the University of
2020.
[110] M. Zarour, M. T. J. Ansari, M. Alenezi, A. K. Sarkar, M. Faizan, Malakand, Pakistan, where he is currently an
A. Agrawal, R. Kumar, and R. A. Khan, ‘‘Evaluating the impact of Assistant Professor with the Computer Sci-
blockchain models for secure and trustworthy electronic healthcare ence and IT Department. His research interests
records,’’ IEEE Access, vol. 8, pp. 157959–157973, 2020. include software outsourcing, empirical software
[111] A. Agrawal, A. K. Pandey, A. Baz, H. Alhakami, W. Alhakami, R. Kumar, engineering, systematic literature review, cloud
and R. A. Khan, ‘‘Evaluating the security impact of healthcare Web appli- computing, requirements engineering, and green
cations through fuzzy based hybrid approach of multi-criteria decision- computing/IT.
making analysis,’’ IEEE Access, vol. 8, pp. 135770–135783, 2020.

19160 VOLUME 9, 2021