mostafa al-emran khaled shaalan aboul ella hassanien

Studies in Systems, Decision and Control 295

Mostafa Al-EmranKhaled ShaalanAboul Ella Hassanien Editors

Recent Advances in Intelligent Systems and Smart Applications

Studies in Systems, Decision and Control

Volume 295

Series Editor

Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,Warsaw, Poland

The series “Studies in Systems, Decision and Control” (SSDC) covers both newdevelopments and advances, as well as the state of the art, in the various areas ofbroadly perceived systems, decision making and control–quickly, up to date andwith a high quality. The intent is to cover the theory, applications, and perspectiveson the state of the art and future developments relevant to systems, decisionmaking, control, complex processes and related areas, as embedded in the fields ofengineering, computer science, physics, economics, social and life sciences, as wellas the paradigms and methodologies behind them. The series contains monographs,textbooks, lecture notes and edited volumes in systems, decision making andcontrol spanning the areas of Cyber-Physical Systems, Autonomous Systems,Sensor Networks, Control Systems, Energy Systems, Automotive Systems,Biological Systems, Vehicular Networking and Connected Vehicles, AerospaceSystems, Automation, Manufacturing, Smart Grids, Nonlinear Systems, PowerSystems, Robotics, Social Systems, Economic Systems and other. Of particularvalue to both the contributors and the readership are the short publication timeframeand the world-wide distribution and exposure which enable both a wide and rapiddissemination of research output.

** Indexing: The books of this series are submitted to ISI, SCOPUS, DBLP,Ulrichs, MathSciNet, Current Mathematical Publications, Mathematical Reviews,Zentralblatt Math: MetaPress and Springerlink.

More information about this series at http://www.springer.com/series/13304

http://www.springer.com/series/13304

Mostafa Al-Emran • Khaled Shaalan •

Aboul Ella HassanienEditors

Recent Advancesin Intelligent Systemsand Smart Applications

123

EditorsMostafa Al-EmranDepartment of Information TechnologyAl Buraimi University CollegeAl Buraimi, Oman

Khaled ShaalanFaculty of Engineering and ITThe British University in DubaiDubai, United Arab Emirates

Aboul Ella HassanienFaculty of Computers and ArtificialIntelligenceCairo UniversityGiza, Egypt

ISSN 2198-4182 ISSN 2198-4190 (electronic)Studies in Systems, Decision and ControlISBN 978-3-030-47410-2 ISBN 978-3-030-47411-9 (eBook)https://doi.org/10.1007/978-3-030-47411-9

© Springer Nature Switzerland AG 2021This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or partof the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmissionor information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilarmethodology now known or hereafter developed.The use of general descriptive names, registered names, trademarks, service marks, etc. in thispublication does not imply, even in the absence of a specific statement, that such names are exempt fromthe relevant protective laws and regulations and therefore free for general use.The publisher, the authors and the editors are safe to assume that the advice and information in thisbook are believed to be true and accurate at the date of publication. Neither the publisher nor theauthors or the editors give a warranty, expressed or implied, with respect to the material containedherein or for any errors or omissions that may have been made. The publisher remains neutral with regardto jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AGThe registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

https://doi.org/10.1007/978-3-030-47411-9

Preface

The field of intelligent systems and smart applications has extremely evolved withnew trends during the last decade. Several practical and theoretical findings aregrowing enormously due to the increasing number of successful applications andnew theories derived from numerous diverse issues. The applications of these trendshave been applied to various domains, including education, travel and tourism,health care, among others. This book is dedicated to the intelligent systems andsmart applications area in several ways. First, it aims to provide and highlight thecurrent research trends in intelligent systems and smart applications. Second, itattempts to concentrate on the recent design, developments, and modifications ofintelligent systems and smart applications. Third, it aims to provide a holistic viewof the factors affecting the adoption, acceptance, or continued use of intelligentsystems and smart applications. It is important to understand these issues in order todetermine future needs and research directions. Fourth, this edited book aims tobring scientists, researchers, practitioners, and students from academia and industryto present the recent and ongoing research activities about the recent advances,techniques, and smart applications and to allow the exchange of new ideas andapplication experiences.

This book is intended to present the state of the art in research on intelligentsystems, smart applications, and other related areas. The edited book was able toattract 60 submissions from different countries across the world. From the 60submissions, we accepted 35 submissions, which represents an acceptance rate of58.3%. The accepted papers were categorized under five different themes, includinginformation systems and software engineering, knowledge management, technol-ogy in education, emerging technologies, and social networks. Each submission isreviewed by at least two reviewers, who are considered specialized in the relatedsubmitted paper. The evaluation criteria include several issues, such as correctness,originality, technical strength, significance, quality of presentation, and interest andrelevance to the book scope. The chapters of this book provide a collection ofhigh-quality research works that address broad challenges in both theoretical andapplication aspects of intelligent systems and smart applications. The chapters of

v

this book are published in Studies in Systems, Decision and Control Series bySpringer, which has a high SJR impact.

We acknowledge all those who contributed to the staging of this edited book.We would also like to express our gratitude to the reviewers for their valuablefeedback and suggestions. Without them, it would not be possible for us to maintainthe high quality and the success of the Recent Advances in Intelligent Systems andSmart Applications edited book. Therefore, on the next page, we list the reviewersalong with their affiliations as a recognition of their efforts.

Al Buraimi, Oman Mostafa Al-EmranDubai, United Arab Emirates Khaled ShaalanGiza, EgyptApril 2020

Aboul Ella Hassanien

vi Preface

List of Reviewers

Andrina Granić, University of Split, CroatiaGarry Wei-Han Tan, Taylor’s University, MalaysiaGonçalo Marques, Universidade da Beira Interior, PortugalHussam S. Alhadawi, Ton Duc Thang University, VietnamIbrahim Arpaci, Tokat Gaziosmanpasa University, TurkeyLee Voon Hsien, Universiti Tunku Abdul Rahman, MalaysiaLeong Lai Ying, Universiti Tunku Abdul Rahman, MalaysiaLuigi Benedicenti, University of New Brunswick, CanadaMohammed Al-Sharafi, Universiti Malaysia Pahang, MalaysiaMohammed N. Al-Kabi, Al Buraimi University College, OmanMohanaad Shakir, Al Buraimi University College, OmanMona Mohamed Zaki, The University of Manchester, UKNoor Al-Qaysi, Universiti Pendidikan Sultan Idris, MalaysiaReham Marzouk, Alexandria University, EgyptTariq Rahim Soomro, Institute of Business Management, PakistanVitaliy Mezhuyev, FH JOANNEUM University of Applied Sciences, Austria

vii

Contents

Information Systems and Software Engineering

A Systematic Review of Metamodelling in Software Engineering . . . . . . 3Murni Fatehah, Vitaliy Mezhuyev, and Mostafa Al-Emran

A Systematic Review of the Technological Factors Affectingthe Adoption of Advanced IT with Specific Emphasis on BuildingInformation Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29Mohamed Ghayth Elghdban, Nurhidayah Binti Azmy,Adnan Bin Zulkiple, and Mohammed A. Al-Sharafi

A Study on Software Testing Standard Using ISO/IEC/IEEE 29119-2:2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43Cristiano Patrício, Rui Pinto, and Gonçalo Marques

Towards the Development of a Comprehensive Theoretical Modelfor Examining the Cloud Computing Adoption at the OrganizationalLevel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63Yousef A. M. Qasem, Rusli Abdullah, Yusmadi Yah, Rodziah Atan,Mohammed A. Al-Sharafi, and Mostafa Al-Emran

Factors Affecting Online Shopping Intention Through VerifiedWebpages: A Case Study from the Gulf Region . . . . . . . . . . . . . . . . . . . 75Mohammed Alnaseri, Müge Örs, Mustefa Sheker, Mohanaad Shakir,and Ahmed KH. Muttar

Knowledge Management

Critical Review of Knowledge Management in Healthcare . . . . . . . . . . . 99Afrah Almansoori, Mohammed AlShamsi, Said A. Salloum,and Khaled Shaalan

ix

Knowledge Sharing Challenges and Solutions Within SoftwareDevelopment Team: A Systematic Review . . . . . . . . . . . . . . . . . . . . . . . 121Orabi Habeh, Firas Thekrallah, Said A. Salloum, and Khaled Shaalan

The Role of Knowledge Management Processes for Enhancingand Supporting Innovative Organizations: A Systematic Review . . . . . . 143Sufyan Areed, Said A. Salloum, and Khaled Shaalan

The Impact of Artificial Intelligence and Information Technologieson the Efficiency of Knowledge Management at ModernOrganizations: A Systematic Review . . . . . . . . . . . . . . . . . . . . . . . . . . . 163Saeed Al Mansoori, Said A. Salloum, and Khaled Shaalan

Technology in Education

A Novel Approach for Predicting the Adoption of SmartwatchesUsing Machine Learning Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . 185Ibrahim Arpaci, Mostafa Al-Emran, Mohammed A. Al-Sharafi,and Khaled Shaalan

Vocabulary Improvement by Using Smart MobileApplication—A Pilot Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197Petra Poláková, Blanka Klímová, and Pavel Pražák

Examining the Acceptance of WhatsApp Stickers Through MachineLearning Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209Rana A. Al-Maroof, Ibrahim Arpaci, Mostafa Al-Emran, Said A. Salloum,and Khaled Shaalan

Exploring the Effects of Flipped Classroom Model Implementationon EFL Learners’ Self-confidence in English Speaking Performance . . . 223Mohamad Yahya Abdullah, Supyan Hussin, Zahraa Mukhlif Hammad,and Kemboja Ismail

Developing an Educational Framework for Using WhatsApp Basedon Social Constructivism Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243Noor Al-Qaysi, Norhisham Mohamad-Nordin, and Mostafa Al-Emran

Research Trends in Flipped Classroom: A Systematic Review . . . . . . . . 253Rana A. Al-Maroof and Mostafa Al-Emran

Perceptions and Barriers to the Adoption of Blended Learningat a Research-Based University in the United Arab Emirates . . . . . . . . 277Rawy Thabet, Christopher Hill, and Eman Gaad

Applying a Flipped Approach Via Moodle: New Perspectivesfor the Teaching of English Literature . . . . . . . . . . . . . . . . . . . . . . . . . . 295Emira Derbel

x Contents

An Integrated Model of Continuous Intention to Use of GoogleClassroom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311Rana Saeed Al-Maroof and Said A. Salloum

A Game-Based Learning for Teaching Arabic Letters to Dyslexicand Deaf Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337Sahar A. EL_Rahman

Information Communication Technology Infrastructure in SudaneseGovernmental Universities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363Abdalla Eldow, Rawan A. Alsharida, Maytham Hammood,Mohanaad Shakir, Sohail Iqbal Malik, Ahmed Kh. Muttar,and Kais A. Kadhim

Emerging Technologies

Internet of Things and Cyber Physical Systems: An Insight . . . . . . . . . 379Charu Virmani and Anuradha Pillai

Web Services Security Using Semantic Technology . . . . . . . . . . . . . . . . 403Firoz Khan and Lakshmana Kumar Ramasamy

DistSNNMF: Solving Large-Scale Semantic Topic Model Problemson HPC for Streaming Texts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429Fatma S. Gadelrab, Rowayda A. Sadek, and Mohamed H. Haggag

Predicting MIRA Patients’ Performance Using Virtual RehabilitationProgramme by Decision Tree Modelling . . . . . . . . . . . . . . . . . . . . . . . . 451Nurezayana Zainal, Ismail Ahmed Al-Qasem Al-Hadi, Safwan M. Ghaleb,Hafiz Hussain, Waidah Ismail, and Ali Y. Aldailamy

Segmentation of Images Using Watershed and MSER:A State-of-the-Art Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463M. Leena Silvoster and R. Mathusoothana S. Kumar

Block Chain Technology: The Future of Tourism . . . . . . . . . . . . . . . . . 481Aashiek Cheriyan and S. Tamilarasi

Biomedical Corpora and Natural Language Processing on ClinicalText in Languages Other Than English: A Systematic Review . . . . . . . 491Mohamed AlShuweihi, Said A. Salloum, and Khaled Shaalan

A Proposed Context-Awareness Taxonomy for Multi-data Fusionin Smart Environments: Types, Properties, and Challenges . . . . . . . . . . 511Doaa Mohey El-Din, Aboul Ella Hassanein, and Ehab E. Hassanien

Systematic Review on Fully Homomorphic Encryption Schemeand Its Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537Hana Yousuf, Michael Lahzi, Said A. Salloum, and Khaled Shaalan

Contents xi

Ridology: An Ontology Model for Exploring Human BehaviorTrajectories in Ridesharing Applications . . . . . . . . . . . . . . . . . . . . . . . . 553Heba M. Wagih and Hoda M. O. Mokhtar

Social Networks

Factors Affecting the Adoption of Social Media in Higher Education:A Systematic Review of the Technology Acceptance Model . . . . . . . . . . 571Noor Al-Qaysi, Norhisham Mohamad-Nordin, and Mostafa Al-Emran

Online Social Network Analysis for Cybersecurity Awareness . . . . . . . . 585Mazen Juma and Khaled Shaalan

Mining Dubai Government Tweets to Analyze Citizens’Engagement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 615Zainab Alkashri, Omar Alqaryouti, Nur Siyam, and Khaled Shaalan

The Impact of WhatsApp on Employees in HigherEducation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 639Jasiya Jabbar, Sohail Iqbal Malik, Ghaliya AlFarsi, and Ragad M. Tawafak

Effects of Facebook Personal News Sharing on Building Social Capitalin Jordanian Universities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 653Mohammed Habes, Mahmoud Alghizzawi, Said A. Salloum,and Chaker Mhamdi

xii Contents

Information Systems and SoftwareEngineering

A Systematic Review of Metamodellingin Software Engineering

Murni Fatehah, Vitaliy Mezhuyev, and Mostafa Al-Emran

Abstract Metamodelling has become a crucial technique to handle the complex-ity issues in the software development industry. This paper critically reviews andsystematically classifies the recent metamodelling approaches to show their currentstatus, limitations, and future trends. This systematic review retrieved and analyzed atotal of 1157 research studies published on the topic of metamodelling. The retrievedstudies were then critically examined to meet the inclusion and exclusion criteria,in which 69 studies were finally nominated for further critical analysis. The resultsshowed that the main application domains of metamodelling are the cyber-physicaland safety-critical systems development. Moreover, the majority of used approachesinclude metamodels formalization, adding spatial and time semantics, and consider-ing nonfunctional properties. Further, themain trends ofmetamodelling developmentinclude the support of complex systems, behavior modeling, and multilevel model-ing. The results of this systematic review would provide insights for scholars andsoftware engineering practitioners looking into the state-of-the-art of metamodellingand assist them in improving their approaches.

Keywords Metamodelling · Software engineering · Systematic review

M. FatehahFaculty of Computing, Universiti Malaysia Pahang, Gambang, Malaysiae-mail: [email protected]

V. MezhuyevInstitute of Industrial Management, FH JOANNEUM University of Applied Sciences,Werk-VI-Straße 46, 8605 Kapfenberg, Austriae-mail: [email protected]

M. Al-Emran (B)Department of Information Technology, Al Buraimi University College, Al Buraimi, Omane-mail: [email protected]

© Springer Nature Switzerland AG 2021M. Al-Emran et al. (eds.), Recent Advances in Intelligent Systemsand Smart Applications, Studies in Systems, Decision and Control 295,https://doi.org/10.1007/978-3-030-47411-9_1

3

http://crossmark.crossref.org/dialog/?doi=10.1007/978-3-030-47411-9_1&domain=pdf

mailto:[email protected]



https://doi.org/10.1007/978-3-030-47411-9_1

4 M. Fatehah et al.

1 Introduction

Extensive applications of software in daily life help users to save time, money, andenergy. However, the development of modern tools requires more and more effortsfrom the side of software specialists and companies. Safety, security, interoperabilityare just a few issues that arise during the software development process. To addressthe complexity of modern software, the resources needed for its development shouldbe multiplied.

Model-Driven Engineering (MDE) is an assuring practice in addressing soft-ware complexity with the application of Domain-Specific Modelling Languages(DSMLs), transformation engines, and code generators [1]. The MDE approachis used to increase productivity and simplify the software design by incorporatingdomain knowledge into the software development process. The domain model is amodel that integrates both the properties and behavior of a specific domain. MDEallows making a domain knowledge explicit and formal.

In MDE, a model has become a crucial representation of an idea and rules of asoftware system.However, amodel development itself needs amodel design at higherlevels of abstraction, which are calledmetamodels [2].Metamodels help to define theconstraints, rules, and the relationship between concepts of a domain model. Thereare several definitions of the metamodelling in scientific literature. One of the firstdefinitions is that a metamodelling can be interpreted as a modeling process, whichhas a higher level of conceptualization and logic than a standard modeling process[3]. Metamodelling can be also defined as a modeling framework that consists ofthe metamodel, model and an instance of the model, where a metamodel definesthe syntax of the modeling language [4]. Some articles define the metamodel as astandardized DSML for software modeling [5, 6].

The application of metamodelling approach to domain-specific software engi-neering has become a common practice nowadays. Metamodeling has become acrucial technique in the process of software development [7]. Metamodelling is usedto define DSMLs for software systems modeling, where the result of the metamod-elling (a DSML) is used to carry out the domain-specific properties and behavior [8].Metamodelling defines rules and methods for creating metamodels as a part of orga-nized and systematic metamodels development process. Many studies are devotedto the development of new and application of existing metamodelling approaches.However, to the best of our knowledge, there is no recent study devoted to the crit-ical analyses of the state-of-the-art of the metamodelling. To solve this problem, asystematic literature review (SLR) was conducted to summarize knowledge on thecurrent applications and limitations of the metamodelling in software engineering.By deriving evidence from the analyzed articles, this research is believed to assistsoftware engineers in the development of new and existingmetamodeling techniques.

A Systematic Review of Metamodelling in Software Engineering 5

2 Methodology

Prior research suggested that software engineering researchers should adopt theevidence-based practice in their study, which is known as Evidence-Based SoftwareEngineering (EBSE) [9]. A systematic literature review (SLR) is the recommendedapproach to perform an EBSE practice. This study follows the SLR guidelines pro-posed by Kitchenham et al. [9], and other systematic reviews conducted in the past[10–13].

2.1 Research Questions and Motivation

Over the years, different metamodels, metamodelling approaches, and techniqueshave been proposed tofind the bestwayof software design.Unlike the existing studiesof metamodelling in software engineering (SE), the proposed research questions ofthe present study are:

RQ1. What are the purposes and application domains of current metamodellingresearch?

RQ2. What are the trends of the metamodelling development in software engineer-ing?

RQ3. What is the quality of publications that reflect the metamodelling research inSE domain?

RQ4. What are the limitations of the current studies and the prospects for futureresearch?

The motivation for formulating RQ1 is to analyze the research purpose of theselected articles along with the application domain of the recent metamodellingapproaches.RQ2discusses the current trends of themetamodelling to define a generaltendency for its development. To identify the quality of the current metamodellingresearch (RQ3), a quality assessment questionnaire was used. Finally, RQ4 discussesthe limitations of the currentmetamodelling approaches in the SE domain, alongwithfuture improvements.

2.2 Search Process

An extensive search was conducted in the digital libraries of scientific literatureto answer the formulated research questions. Journal articles, workshop articles,conference papers, and books were included in the search process. Firstly, five digitallibraries were selected as platforms for the search process. These libraries are theScienceDirect, IEEE Xplore, Clarivate Analytics (formerly known as ISI Web ofKnowledge), ACMDigital Library, and SCOPUS. Next, the keywords for the search

6 M. Fatehah et al.

Table 1 Keywords definition A B C

Metamodelling Meta-modelling Softwareengineering

Metaprocessmodelling

Meta-processmodelling

Metadatamodelling

Meta-datamodelling

process were defined. As the use of general keywords results in a large number ofresearch papers, specific keywords were formulated with the aim of finding the mostrelevant studies. The search string was defined as “(A OR B) AND C”, where thekeywords are shown in Table 1.

The search process also takes into account a different spelling of the word “mod-elling” in British and American versions of English (i.e., the search query wasrepeated for “modelling” and “modeling”). The motivation to use the more generalkeyword “Software engineering” comparatively to the “Model Driven Engineering”is to cover other possible applications of the metamodelling (e.g., validation, stan-dards compliance, and risk management). At the same time, MDE is considered asa part of SE, so the search string does not need to include both terms.

2.3 Inclusion and Exclusion Criteria

A total of 1157 papers were found as a result of the search process since specifickeywords were applied. Hence, the inclusion criteria are crucial to find the papersrelevant to this research. The importance of the keyword “software engineering”can be illustrated in the example that metamodelling now has intensive applicationsin different domains such as spatial correlation modeling (also known as Krigingmetamodel) [14]. The collected papers have undergone eight phases of selectionthrough the use of inclusion and exclusion criteria, as outlined in Table 2. The reviewprocess and the number of studies identified at each stage were undertaken accordingto the preferred reporting items for systematic reviews andmeta-analysis (PRISMA),as depicted in Fig. 1.

2.4 Data Collection

The relevant informationwas extracted fromselectedpapers to facilitate the data anal-ysis process. The data were categorized into three different types. The explanationof extracted data and their category is shown in Table 3.


Table 2 Inclusion and exclusion criteria

Phases Inclusion/exclusion criteria

1 Search engine results after executing a search query

2 Date of publication since 2012

3 Full-text access

4 Published in the English language

5 Published in software engineering venues

6 Not duplicated

7 Not summaries of workshop, panel, or tutorials

8 Title, abstract, or full text contains a search string with the predefined keywords inTable 1

2.5 Search Strategy

Although the keywords used were specific, a number of irrelevant papers to theresearch topic were found. Hence, the exclusion criteria were applied to identify therelevant papers. Figure 2 and Table 4 show the distribution of papers retrieved bythe application of search keywords and their distribution among the digital librariesafter applying the exclusion criteria. It can be seen that the IEEE dominates theinitial papers retrieved from the digital libraries with a percentage of 46% of the totalcollected papers. On the contrary, ACM Digital Library is the highest contributorto the final amount of papers, representing 31% of the total remaining papers. Thisresult shows the effectiveness of applying the inclusion and exclusion criteria. Avalidity threat of misinterpretation of the primary studies was mitigated by the firsttwo authors of this study. The candidate studies also went through multiple reviews.The inclusion of each article into the final list of relevant primary studies was checkedduring the concluding agreement meeting among the authors.

3 Results and Discussion

The answers to the formulated research questions are discussed in this section.

RQ1. What are the purposes and application domains of current metamodellingresearch?

Although there were 1157 papers found published on the metamodelling since 2012,only 69 papers were found relevant to this study after applying the exclusion criteria.Tables 5, 6, 7, 8, 9 and 10 describe the research purpose of the selected papers.The analysis of the studies shows that the scholars have focused on several commonissues, including formalization, safety and security aspects of the systems, multilevelmodeling, behavior modeling, and processes improvement.

8 M. Fatehah et al.

Fig. 1 PRISMA flowchart

Table 3 Extractedinformation

Category Data

Basic information • Title• Author(s)

Publication information • Source• Year

Research information • Application domain• Research purpose• Used metamodels and tools• Industrial application• Limitation of research• Future work


46%

18% 7%

15% 24% 26%

13% 10% 10%

31%

0%

20%

40%

60%

80%

100%

Distribu on of papers ini ally retrieved from thedigital libraries

Distribu on of papers a er applica on ofexclusion criteria

IEEE ScienceDirect SCOPUS Clarivate Analy cs ACM

Fig. 2 Distribution of the papers retrieved from digital libraries

Table 4 Distribution of initially retrieved and selected papers across digital libraries

Digital library Number of papers initiallyretrieved

Number of papers after applyingthe exclusion criteria

IEEE Xplore 529 12

ScienceDirect 78 10

SCOPUS 281 18

Clarivate analytics 152 8

ACM digital library 117 21

Total 1157 69

The advantages of adopting metamodelling in the software development lifecy-cle have been taken notice in software engineering. This claim is proved by themultiple metamodelling studies in various SE domains. Figure 3 illustrates the mainapplication domains of metamodelling in several specific categories. This figure pro-vides evidence regarding the industrywillingness to adaptmetamodelling in softwaredesign, analyses, and validation. Cyber-physical systems development and the Inter-net ofThings (IoT), togetherwithmodeling safety and security aspects of the systems,are recognized as themost important application domains ofmetamodelling. Figure 4demonstrates the types of systems involved in metamodelling applications.

RQ2. What are the trends of the metamodelling development in softwareengineering?

The current trends of metamodelling development can be divided into three parts,including the analysis of metamodeling improvement directions, studying the useof metamodels, and tools. Based on Tables 5, 6, 7, 8, 9 and 10, the developmentof metamodelling approaches has various ways. However, the analysis shows thatthe most frequent approaches are the formalization of metamodelling approaches,conceptual modeling, and multilevel modeling. Researchers also concentrate on the

10 M. Fatehah et al.

Table 5 Research purpose description for the studies published since 2017

References Research purpose Approach/domain

[15] To develop a framework that supportsthe behavior and structural complexityof combat system simulation

Simulation and software validation

[16] To enhance the semantics of UMLtemplates in OCL through thedevelopment of an aspectual template

Formalization, constraints, andsoftware development

[17] To simplify the development ofdomain-specific modeling tools

Code generation and softwaredevelopment

[18] To develop the UML profile formodeling cognitive behavior

Development, cognitive behavior, andUML

[19] To develop a flexible tool-supportedmodeling approach that augments asketching environment withlightweight metamodelling capabilities

Development and sketchingenvironment

[20] To transform the Open PlatformCommunications Unified Architecture(OPC UA) to the UML model

Transformation and development ofUML profile

[4] To facilitate the simulation applicationin software development

Simulation and software development

[5] To transform the machine learningelement to domain modeling

Transformation and Cyber-PhysicalSystems (CPS)

[6] To present the constraints validationmodel benchmark for a large graphmodel

Constraints, validation, andsafety-critical systems

[21] To create a metamodel for modeling asmart environment through functionaland data perspective

New metamodel, CPS, Internet ofThings, and smart environments

[22] To increase the modeler productivityby task automation

Automation, workflows, and softwaredevelopment

[23] To adopt a dual deep modelingapproach as conceptual data modeling

Formalization, multilevel modeling,and conceptual modeling

[24] To provide a uniform formalization onsub-model and sub metamodel throughconstraint properties

Formalization, constraints, andsoftware development

[25] To propose a theory for multi-levelconceptual modeling

Multi-level modeling and conceptualmodeling

methods formetamodels reuse, adaption, and integration. Since the present study ana-lyzed both the basic research and applied research, Fig. 5 also presents the improve-ment of the syntax and semantics and the metamodel development inside an existingmeta-metamodel (as a development of the new UML profiles).

Figure 6 summarizes the most used metamodelling tools since 2012. This figureshows that Eclipse Modelling Framework (EMF) is the most used tool. Table 11


Table 6 Research purpose description for the studies of 2016


[26] To investigate the need for strictmetamodelling practice

Formalization and safety-criticalsystems

[27] To develop and maintain modelinglanguage and generator withtest-driven development

Code generation and softwarevalidation

[28] To adopt Software Product Lines(SPL) at the metamodel level

Metamodel reuse and softwaredevelopment

[29] To define graphical syntax bydesigning metamodel to supportgraphical modeling language

New metamodel and softwaredevelopment

[30] To specify the safety compliance needsfor a critical system

Standards compliance andsafety-critical systems

[31] To discover basic elements in themodeling of software development

Multilevel modeling and softwaredevelopment

[32] To define an abstract framework formultilevel modeling

Multilevel modeling and conceptualmodeling

[33] To support the security patternspecification and validation

Formalization, validation, andsafety-critical systems

[34] To develop the metamodel that mergesevidence, goal, and risk into softwaredevelopment

Risk management and safety-criticalsystems

[35] To create a modeling language thatsupports abstract syntax and dynamicsemantic for the cyber-physical system

Formalization, CPS, and intelligentsystems

[36] To create an interface for safetyevaluation and fault simulation in anautomotive system

Evaluation, validation, andsafety-critical systems

[37] To represent metadata in the buildingsusing the proposed schema

Metadata, information systems, CPS,and intelligent houses

[38] To create a metamodel developmentmethod with different mathematicalsemantics

Formalization (spatial and timingsemantics) and CPS

[39] To introduce behavior modeling forautomatic discrete event systemspecification model

Formalization and behaviour modeling

[40] To investigate the capabilities ofsimulation metamodelling in handlingdecision support tool

Simulation and decision support

[41] To develop a modeling language fordata description

New metamodel, metadata, andinformation systems

[42] To provide information on thesimulation model characteristic and itslimitations

Simulation and intelligent systems




[43] To understand the requirements andneeds of modeling stakeholders

Requirement engineering andconceptual modeling

[44] To prove the efficiency of simulationmetamodelling in healthcare analysis

Simulation and CPS (Healthcare)

[45] To support runtime synchronizationbetween the model and program state

Synchronization and softwaredevelopment

[46] To evaluate the software modelinglanguage elements

Metamodel evaluation and conceptualmodeling

[47] To define the execution semantic in theconcurrent model

Execution semantics and softwaredevelopment

[48] To discuss the situation whenmultilevel modeling is beneficial

Multilevel metamodelling and softwaredevelopment



[49] To develop a framework for modelingsecurity patterns specification andvalidation

Validation and security domain

[50] To use the ontological approach toaddress a semantic gap in securitypatterns

Semantics and security domain

[51] To define the archetypes-basedframework for evolutionary,dependable, and interoperablehealthcare information systems

Metadata and CPS (Healthcare)

[52] To develop SysML-based automatingbuilding energy system for modelingand analysis

Automation and CPS (energy system)

[53] To improve the change impact analysisin software requirements

Analyses and requirement engineering

[54] To formalize coordination amongmetamodels

Formalization and CPS (healthcare)

[55] To define a seamless chain for softwarestructural, functional, and executionmodeling

Transformation and CPS (embeddedsystems)

[56] To address the conflict in informationsystem design using a conceptualmodeling approach

Multilevel metamodelling and softwaredesign




[57] To support the theory of relation andreasoning in different approaches torequirement modeling

Formalization and requirementengineering

[58] To define the rules to create a riskmanagement system model

Risk management and safety-criticalsystem

[59] To improve the component-basedapproach for specifying DSML’sconcrete syntax

Syntax, semantics, and softwaredevelopment

[60] To integrate safety analysis in thesystem engineering process

Safety analyses, system engineering,and safety-critical systems

[61] To evaluate the existing softwarearchitecture viewpoint language

Evaluation and software design

[62] To propose a tool that supports thesoftware process reuse and automatestheir execution

Process reuse and automation andsoftware processes

[63] To introduce a model-drivenprocess-centered software engineeringenvironment

Software processes and softwaredevelopment

[64] To identify the right time and scenarioto use multilevel modeling

Multilevel modeling and conceptualmodeling

[65] To develop a metamodel that supportsconstraint in a database managementsystem

Constraints and information systems

[66] To define a template for the rationale ofarchitecture design decision

Decision support and softwarerequirements and design

[67] To develop a technology that supportsrapid domain-specific languagedevelopment

Automation and software development

[68] To support the metamodel developmentbased on the graphical representation

Metamodel generation and sketchingenvironment

[69] To define a programming and modelinglanguage that treats requirement andarchitecture concepts explicitly

Formalization and requirementengineering

[70] To generate test cases andmetamodelling correctnessautomatically

Code generation, formalization, andsoftware validation

[71] To develop a metamodel thatgeneralizes DSML and MDE

Models execution and softwaredevelopment




[72] To provide a behavior model reasoningat different levels of abstraction

Formalization, behavioral modeling,and CPS (Healthcare)

[73] To provide a framework for metadataediting and development

Metadata and geographic informationsystems

[74] To provide an approach that supportsmodel and constraint evaluation

Automation, model adaptation, andsoftware design

[75] To propose a metamodelling in theadaptation of Service-OrientedArchitecture (SOA) roles in thetraditional organization structure

Model adaptation and webtechnologies (SOA)

[76] To combine the agent-orientedsoftware development and MDEparadigms

Integration and artificial intelligence

[77] To develop a metamodel that supportsconcurrent task tree modeling andexecution

Code execution and concurrentprocesses

[78] To improve the search engine abilitywith metamodelling

Improvement and web-technologies

[79] To investigate the need for a multilevelbusiness process

Multilevel modeling and businessprocesses

[80] To address the challenge requirementsspecification for a graphical modelinglanguage

Syntax and requirement engineering

02468

10121416

Req

uire

men

t eng

inee

ring

Softw

are

desi

gn, a

naly

ses

and

valid

atio

n

Mod

ellin

g so

ftwar

epr

oces

ses a

nd w

orkf

low

s

Cyb

er-p

hysi

cal s

yste

ms

and

IoT

Info

rmat

ion

syst

ems

Inte

llige

nt sy

stem

s

Safe

ty a

nd se

curit

y as

pect

sof

the

syst

ems

Stan

dard

s com

plia

nce

and

risk

man

agem

ent

Fig. 3 The main application domains of metamodelling


Safety critical systems

Aircraft systems

Automotive systems

Combat systems

Healthcare systems

Intellegent systems

Smart devices

Building energy systems

Smart environment systems

Multi-agent systems

Management system

Disaster management

systems

Emergency management

systems

Workflow modeling

Risk analysis systems

Information Systems

Enterprise databases

Organization structure databases

Spatial and temporal databases

Web-based systems (search engines)

Fig. 4 Types of systems used in metamodeling applications

02468

1012

Met

amod

els

form

aliz

atio

n

Con

cept

ual

mod

ellin

g (in

cl.

beha

viou

rm

odel

ling)

Mul

tilev

elm

odel

ling

Met

amod

els r

euse

,ne

w m

etho

ds fo

r the

mod

els a

dapt

ion

and

inte

grat

ion

Impr

ovem

ent o

f the

synt

ax a

ndse

man

tics

Dev

elop

men

t of t

hem

etho

ds fo

r the

met

amod

els

valid

atio

n

New

met

amod

els

deve

lopm

ent

Dev

elop

men

t of n

ewU

ML

prof

iles

Fig. 5 The main trends of metamodeling development

shows the advantages and disadvantages of EMF. The researchers noticed that themain advantage is the various software environments and hardware platforms sup-port. The EMF includes a constraint definition language, which can be used to definethe model metrics and automatically detect deviations from the model design heuris-tics. Moreover, the Eclipse framework offers a variety of tools, which are portablefor different platforms.

Other than modeling tools, this review study also studies the most used meta-metamodels and metamodels. Based on Fig. 7, ECORE is the most used meta-metamodel. ECORE offers many advantages, such as compatibility to support the


Fig. 6 Metamodelling tools

(N = 1)

DPF Diagram Predicate

Framework

(N = 11)

Eclipse Modelling Framework

(N = 2)

Generic Modelling

Environment

(N = 1)MetaDepth

(N = 2)MetaEdit+

(N = 1)SeMF

Table 11 Advantages anddisadvantages of eclipsemodelling framework

Advantages Disadvantages

• Provides various platformsfor a language definition

• Ease of use in any softwareenvironment

• Easy to define a graphicalmodel

• Easy to develop a graphicaleditor

• Does not allow to model asystem behavior

• Does not support a uniquenumeric identification

• Fixed number ofmeta-levels

Fig. 7 Most referencedmetamodels (N = 1)ADOxx

(N = 1)AUTOSAR

(N = 8)ECORE

(N = 1)Kermeta

(N = 1)SMP2

(N = 1)SPEM

(N = 1)Universal

Metamodel

various platforms. The advantages and the motivation of ECORE are described inTable 12, along with the other metamodel descriptions and researchers’ motivations.While all considered metamodels are used to define concrete and abstract syntax ofthe models, Table 12 is dedicated to show their specific features.


Table 12 Existing metamodels descriptions and motivations

Core metamodel Descriptions Motivations

ADOxx • Used to specify a platformwith multiple deploymentoptions

• Support for an independentdeclarative language

AUTOSAR • Specify architectures forautomotive systems

• Standardize architecture andmethodology of anautomotive system

ECORE • Describes three layers ofmodel abstraction

• Can describe the safety andsecurity concepts of aparticular system

• Needs expanding byconstraint language

• Defines metamodel and therules for the automation oftool generation

• Can support softwaremethodology Prometheuslanguage

• The definition of concretetextual syntax is supported

• To support modeldevelopment forcomponent-based systems

• To follow the EMFcompatibility metamodel

• To support metamodelcollaboration and varioustools

• To describe the modelingelement of Prometheusmethodology

• To place the proposedmethodology in EMF

• To map structural features ofa model to abstract syntax

Kermeta • Metamodel that supportsspecific design patterninterpretation

• Can support modelinterpreter, compiler, andchecker

Simulation ModelPortability standards 2(SMP2)

• Support automatic codegeneration from theplatform-independent model

• This metamodel supportmodel composability

Software and SystemsProcess EngineeringMetamodel (SPEM)

• Initiate a model-driven process • Support for a specific aspectof the model-drivendevelopment concept

Universal Metamodel • Use the diagram element for aspecialized class

• Metamodel that supportsmodel-management systems

RQ3. What is the quality of publications that reflect the metamodelling researchin SE domain?

Each selected studywas evaluated according to journal ranking, which divides papersinto four quartiles (i.e., Q1, Q2, Q3, and Q4). Q1/Q2 covers the journals of thehighest impact factor, while Q4 covers the lowest ones. Conference papers were alsoconsidered. Figure 8 shows an increasing number of the impact of journal articlespublished between 2012 and 2017. These results acknowledge the increasing amountofmetamodelling approaches in the scientific literature. The selectedpaperswere alsoundergone six quality evaluation questions, as presented in Table 13. Three possibleanswers (i.e., “Yes”, “Partially”, and “No”) were used to evaluate the selected articlesagainst the six quality evaluation questions. Figure 9 shows the quality assessmentresults of the selected papers.


1

2

4

0

2

1

2

1

0

1

4

7

0

1

0 0 0 0 0 0 0 0 0 0

7

10

4

8

11

2

2 0 1 2 2 0 1 3 2 0 1 4 2 0 1 5 2 0 1 6 2 0 1 7

Q1 Q2 Q3 Q4 Conference

Fig. 8 Articles ranking per year

Out of the total number of selected articles (N= 69), all the articles mentioned thestudy purpose and application domain of the metamodelling approach either witha wide explanation or by a simple statement. 13% of the selected papers can beclassified as basic research that extends the state-of-the-art in metamodeling, while62% of them apply metamodeling to some domain. More than 60% of the selectedarticles explained in detail the methodology of research, while 30% of them did notelaborate on it in a structural way. Additionally, 27 articles mentioned and explainedthe use of a proven meta-metamodel, while only 21 articles mentioned the meta-metamodel. The remaining 21 articles just mentioned the use of an existing meta-metamodelwithout a clear explanation. Further, 37%of the considered studies clearlyshowed the applicability of themetamodelling approach in the software developmentindustry, while 19% of them just mentioned the possible industrial applications.Moreover, about 25%of the selected studies indicated the futurework and limitationsof their studies, which could help to determine the trends of the metamodellingapproach. 28% of the selected studies just mentioned either the possible future workor limitations of their studies, while more than 40% did not mention the future workand limitations.

RQ4. What are the limitations of the current studies and the prospects for futureresearch?

The answers to this question provide an overviewof the limitations and futurework ofthe reviewed studies. The overview is divided into two parts, including the analysisof limitations and the proposed future improvements. First, the limitations of thereviewed studies were analyzed in Table 14. Second, based on these limitations,future works were also analyzed. A brief overview of future proposals is describedin Table 15. Tables 14 and 15 present the limitations and future proposals for thescholars who only reported the limitations and future works in their studies.

There are several common issues highlighted in the selected research papers.Table 14 allows to deduct the limitations of metamodelling. While its application isrecognized as an effective, many authors noticed that manual coding is still required.


Table 13 Quality assessmentquestionnaire

QA1 Did the study consider the purpose and applicationdomain of the considered metamodelling approach?• Yes: it explains the purpose of metamodelling approach andits application domain

• Partially: it explains the purpose of metamodelling approach,but not the application domain

• No: it does not mention the purpose and application domain

QA2 Did the study extend the state-of-the-art inmetamodeling?• Yes: it extends the state-of-the-art in metamodeling• Partially: it shows some new results in the metamodellingdomain

• No: it applies an existing metamodeling approach to somedomain

QA3 Did the study consider and give details on themethodology used?• Yes: it clearly explains the methodology used• Partially: it lists a few methods but not in a structural way• No: it did not discuss the methodology used

QA4 Did the study use a proven meta-metamodel?• Yes: it names the meta-metamodel and how it is used in theresearch

• Partially: it names the meta-metamodel but did not describehow the meta-metamodel is used

• No: it mentions the meta-metamodel name withoutexplanation

QA5 Did the study consider an industrial use of themetamodelling approach?• Yes: it discusses the use of metamodelling approach in theindustry

• Partially: it mentions the possible use of metamodelling inthe industry

• No: it does not mention the industrial use of metamodelling

QA6 Did the study consider the future work and discusslimitation?• Yes: it explains the possible future elaboration ofmetamodelling approach and its limitations

• Partially: it mentions either future works or limitations• No: it did not mention the limitations nor the possibility offuture work

To support the design of cyber-physical systems (e.g., intelligent houses), furthermetamodel formalization and expansion by spatial and time semantics are required.

Based on Table 15, the researchers highlighted several common directions forthe improvement of metamodelling. The important directions include metamodelformalization, support of constraint language, and allowing formal specification andautomatic reasoning. To solve these arising issues, the authors of this study pro-posed to improve the conceptual foundation of metamodelling approach. Metamodel

mostafa al-emran khaled shaalan aboul ella hassanien

Documents