Software Requirements Negotiation: a Systematic Literature Review

doi:10.21203/rs.3.rs-2187638/v1

Download PDF

Research Article

Software Requirements Negotiation: a Systematic Literature Review

https://doi.org/10.21203/rs.3.rs-2187638/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Software requirements negotiation (SRN) is one of the most essential stages of software requirements engineering. SRN involves the stakeholder's interaction to reach a mutual understanding of the requirements for developing a software project. The increased research interest in requirements engineering has resulted in huge literature in the SRN domain. There is a need to investigate the broad of techniques, processes, and evaluation mechanisms used in the SRN research community. This study aims to examine and identify the existing methods, processes, evaluation mechanisms, quantity of publications, publication trends and demographics shaping SRN research domain. To accomplish our aim, we used an evidence-based systematic approach, and 67 relevant studies were ultimately chosen from the search process based on the formulated research questions. Our study result shows broad and promising SRN techniques that include agent-based negotiation, TAICOS, wikiwinwin and winbook. However, we found that the existing SRN techniques suffer from limitations that have stakeholders' communication gaps, interface issues to non-technical users, decision-making, and managing requirements changes.

Moreover, our study found that 63% of the selected studies stated their SRN processes, with 22% adopting the basic SRN processes in winwin model. Five evaluation mechanisms were discovered, with case study and experiment most adopted by the selected studies with 44% and 30%, respectively. In conclusion, although research on SRN is recently gaining some traction, the works in the domain are insufficient. Concrete proposals are needed to improve SRN in software requirements engineering research domain.

requirements

negotiation

stakeholders

techniques

processes

Software requirements engineering (SRE) is the early phase of software engineering that deals with identifying stakeholders and their needs, and documenting the needs for analysis, communication, and implementation [1]. The common challenges of this phase include the geographical distribution of the stakeholders, their varying goals, requirements conflict and the tasks to accomplish [2]. The basic SRE iterative activities defined in [3] are requirements elicitation, requirements negotiation, requirements documentation, requirements validation and requirements management. Software Requirements negotiation (SRN) is considered to be the output of the interaction between the 'stakeholders' which leads to mutual understanding [4]. SRN begins by sharing the 'stakeholders' goals and completed when all the stakeholders agree on the goals [5]. SRN results are set to satisfy the requirements of two or more parties in the presence of limited common knowledge and conflicting preferences [6, 7]. There are many advantages to negotiating stakeholders' requirements before architecture design or coding. This is because many software projects have failed. After all, their requirements were poorly negotiated among stakeholders[8]. Therefore, effective SRN is significant because a software product developed based on negotiated needs can be expected to be accepted by all the stakeholders [9]. To negotiate requirements, stakeholders will have to weigh the elicited requirements and compare them based on relative criteria of importance. There exist a variety of SRN techniques and processes used for the conduct of SRN in the software industry. This helps to resolve software industry challenges such as stakeholders' communication gap, scalability, decision-making, and managing requirements changes.

Over the past decade, many studies have been published to enable the multiple stakeholders to resolve conflicting requirements through proposals of novel SRN techniques and processes. The studies that published systematic reviews in SRN include [10-12]. However, no systematic literature review (SLR) comprehensively describes the SRN techniques and their limitations, SRN processes, demographic characteristics, and evaluation mechanisms. Therefore, our study aims to provide a comprehensive SLR that identifies, categorizes, and synthesizes published literature, presents an overview of the existing techniques and processes used in the conduct of SRN, and then highlights the evaluation mechanisms in the selected studies. To achieve the study aims, an evidence-based systematic methodology was utilized from ' 'kitchenham's [11] SLR to ensure that important studies on SRN published between (2010 - 2021) are identified and retrieved. The methodology employs a systematic selection and evaluation process with a rigorous and repeatable evidence-based studies. The result of our SLR was based on the general demographics and characteristics, the proposed SRN techniques and methods, the evaluation mechanisms used by the selected studies, and the limitations of the existing SRN techniques.

In summary, this SLR tends to clarify the available evidence regarding

The existing SRN techniques
their limitations
Processes
Evaluation mechanisms
General demography and characteristics

Conclusively, related work is presented in Section 2. Section 3 describes the methodology used in the study. Section 4 presents the results of our findings concerning the defined research questions. Section 5 outlines the general research findings of the SLR. In Section 6, the Threat to validity is presented. Finally, the study is concluded in Section 7.

This section highlights and discusses the existing survey and review papers in the SRN research domain. However, few systematic literature reviews have been published in the domain. The study of Lenz et al. presented a review on electronic requirements negotiation [10]. The study scope was limited to literature surveys on state-of-the-art in electronic requirements negotiation. In another study by Tito et al., the authors conducted a mapping study in SRN with a focus on SRN techniques, evaluation mechanisms, and experimental setting [13]. However, the study did not clearly state the SRN techniques, SRN processes, and limitations.

Additionally, Terpra adopted the guidelines in [11] to systematically review requirement conflicts and SRN methods from 2010 to 2015 [12]. The study describes the conflicts as limitations in SRN, but the limited searching strategy might lead to missing relevant papers. Kerstan et al. conducted a review on e-negotiation techniques [14]. The authors reviewed the negotiation system, negotiation support, and negotiation analysis. Similarly, Riaz et al. presents SLR in SRN domain that reviews the existing SRN techniques, the setting of conducting SRN studies, and a brief description of the publication channels [15]. The study did not review the demographic characteristics of the selected studies, SRN processes, SRN techniques descriptions and limitations. Lastly, Hutchinson and Kotonya reviewed SRN techniques in component-based engineering [16]. The authors highlighted the SRN approaches and identified the strengths and limitations of the selected studies.

However, based on the highlighted review studies, there is no comprehensive systematic literature review in SRN. Moreover, the study selection process adopted in reviewed studies was not rigorous without any quality assessment evaluation. Additionally, no study comprehensively describes the SRN techniques and their limitations, SRN processes, demographic characteristics, and evaluation mechanisms. Therefore, this study provides a comprehensive SLR in SRN to fill the identified research gaps. Table 1 presents the comparison between the related work reviewed and this study.

Table 1

Related Work comparison
Study	Reference	Techniques	Processes	Evaluation Mechanism	Limitations	Demography
Lenz et al.	[10]	✓	X	X	X	X
Tito et al.	[13]	✓	X	✓	X	X
Terpra	[12]	✓	X	X	✓	X
Kerstan et al.	[14]	✓	X	X	X	X
Riaz et al.	[15]	✓	X	X	X	✓
Hutchinson and Kotonya	[16]	✓	X	X	X	X
Our Study		✓	✓	✓	✓	✓

This study followed the guidelines for conducting systematic literature reviews (SLRs) in [11]. In an SLR, researchers have to interpret and evaluate available research related to the formulated research questions and topic [17]. The SLR was conducted to summarise the existing literature and identify gaps in current research to suggest areas for further investigation [18–20]. These sections highlight the SLR planning and details. The details of the SLR include the research questions, inclusion and exclusion criteria, data sources and search strategy. This SLR is conducted to achieve the following objectives to get an explicit view of the current trend in SRN. Firstly, we would like to understand the techniques used in SRN. Secondly, we would like to investigate and find out the processes in these SRN techniques (if any). Thirdly, we intend to find the limitations of the SRN techniques. Finally, we would like to know the selected primary studies' demographic characteristics and evaluation mechanisms

3.1 Research Questions

In this sub-section, the objectives of the study defined are translated into specific research questions as presented in Table 2.

Table 2

Research Questions
RQs	Aim
RQ1: What are the existing SRN techniques proposed by the selected studies?	To identify the existing SRN techniques proposed by the selected studies.
RQ2: What are the processes in SRN Techniques?	To examine the processes used in conducting SRN by the selected studies.
RQ3: What are the evaluation mechanisms in adopted by the selected studies?	To investigate the techniques used in evaluating SRN studies.
RQ4: What are the limitations of the existing SRN techniques?	To describes the limitations and research gap in the existing SRN techniques.
RQ5: What are the selected studies demographics and characteristics?	To highlight the relevant publication channels, relevant publication sources, the annual publication frequency, and presents the selected studies based on their Quality assessment scores.

3.2 Data Source

To carry out an exhaustive search for data to be used in this study, we queried five electronic databases (see Table 3) that were considered primary databases for relevant study retrieval. However, our study did not consider Google Scholar due to its low precision and overlapping results compared to other databases [20]. Additionally, there is no provision of advanced search options in the Google Scholar database which is needed for a more accurate search of relevant selected studies

Table 3

Electronic databases used in this study
ID	Database Name	Link
ED1	ACM	https://dl.acm.org/
ED2	Science Direct	https://www.sciencedirect.com/
ED3	Scopus	https://www.scopus.com/
ED4	Dimension	https://www.dimensions.ai/
ED5	IEEE Xplore	https://ieeexplore.ieee.org/

3.3 Search String

This study used the search guidelines defined in[21] to construct the search string.

a) Terms derivation from the formulated research questions.

b) Identification of synonyms and spelling of the derived terms

c) Identification of keywords from the related studies

d) Use of Boolean operations (AND, OR and *) in the defined words and terms.

The string formed and used in this study is as follows:

((Requirement*) AND (negotiation) OR (Requirement) AND (negotiation) AND (technique* OR “Process” OR “Limitation”))

The databases were searched by title, keyword and abstract. The search string was adjusted during the search process to suit the specific requirements of each database in (Table 3). The search was limited by the year of publication: 2010–2021.

3.4 Study Selection Procedure

The study selection process is intended to identify the relevant primary studies that provide direct evidence about the formulated research questions [18]. The study selection procedure was carried out in stages, as illustrated in Fig. 1, to eliminate bias and establish unanimity. The first stage was the exclusion of articles based on duplicates on the initial 3031 articles obtained from the databases, followed by the exclusion of primary studies based on the title and abstract, and then the exclusion of primary studies based on the full text. The duplicate articles were removed prior to the screening based on the titles. As a result, the second step resulted in 304 articles being screened. To ensure the accuracy of the publications' coverage, we used the snowballing approaches (backwards and forward) on 304 studies that were screened. The papers were examined using forward snowballing, which looked at how many times they cited each other (forward citation).

On the other hand, backward snowballing involves critically analyzing publications based on references (backward citation). Articles that do not fulfil our inclusion requirements are removed in both circumstances. In the end, 67 publications were chosen for the research.

3.5 Inclusion and Exclusion Criteria

To further filter the findings received from the search process, inclusion and exclusion criteria were established in this study. As illustrated in Fig. 1, the criteria were applied to the studies retrieved from the adopted electronic databases. To find relevant studies, a eleven-year search timeframe was established (from January 2010 to June 2021), ensuring that only relevant and up-to-date articles published were included in the study. The inclusion and exclusion criteria in this study are described as applied in the second and third stages of the defined protocol in Fig. 1.

The inclusion criteria are:

IC1: The paper published from 2010–2021
IC2: The paper that focuses on SRN
IC3: The paper has the potential to answer at least one of the study research questions
IC4: The paper is published in the English language
IC5: The paper is available for download

The exclusion criteria are:

EC1: Paper that is duplicates of other studies
EC1: Paper outside the scope of SRN
EC1: The paper is non-peer reviewed, has no full text, and cannot address at least one of the study research questions
EC1: The paper written in a language other than English

3.6 Quality Assessment

The quality assessment (QA) is a more precise inclusion/exclusion criterion for evaluating the quality of the primary studies that have been chosen[18]. It can be done by assessing studies to determine the ones relevant to the research questions set [22]. The quality assessment questions used by this study to examine and evaluate the relevance and truthfulness of the selected studies are listed in Table 5. Figure 1 shows that the screening stage resulted in 305 papers, and using quality evaluation criteria on this paper eliminated another 82 papers. In Q1 and Q2, a three-point likert scale was used: "Yes (+ 1)," "Partially (+ 0.5)," and "No (+ 0)." Whereas in Q3, Q4 and Q5, a two-point possible likert scale was used: """"Yes (+ 1"""" " " " )"," No (+ 0""")". Moreover, for the Q6, we used the Journal Citation Report (JCR) list and the Computer Science Conference Ranking (CORE) to address this quality question. Hence, the possible answers are presented in Table 4.

The evaluation of each quality assessment question against each of the primary selected studies in this study is presented in Table 5. The sum of all the quality assessment questions determined the total quality assessment score of the main chosen studies.

Table 4

Possible Answers to Q5
Journals papers	Conference, workshop, Symposium and Technical Report
Q1 (+ 2)	CORE A (+ 1.5)
Q2 (+ 1.5)	CORE B (+ 1)
Q3 or Q4 (+ 1)	CORE C (+ 0.5)
Without JCR and others (+ 0)	Without CORE ranking (+ 0)

Table 5

Quality Assessment Questions
Questions	Response options for Score
Q1: Is the Objective of the study clearly articulated?	Yes = 1 Partially = 0.5 No = 0
Q2: The study clearly described the proposed technique?	Yes = 1 Partially = 0.5 No = 0
Q3: The study clearly describes its evaluation mechanism?	Yes = 1 Partially = 0.5 No = 0
Q4: The SRN processes clearly described?	Yes = 1 No = 0
Q5: The contribution and limitations of the study are clearly highlighted?	Yes = 1 No = 0
Q6: The study has been published in a higher ranked publication source?	Refer to Table 3

3.7 Data Extraction

This phase aids the researchers in accurately extracting relevant data from their chosen papers. The information retrieved from the complete text of the primary studies used for this investigation was stored in a Microsoft excel file developed by the authors. The following items were extracted: the SRN technique, the SRN processes employed, the evaluation mechanism used, the study limitation (if applicable), the title, authorship information, and publishing information.

This section presents the findings and discussion of our SLR. Firstly, the SRN techniques were highlighted as identified, followed by the SRN processes presented in the identified SRN techniques. Thirdly, the limitations in SRN techniques were described. Furthermore, the demographic of the selected studies was presented. Table 6 highlights the primary selected studies.

Table 6

Highlights of the primary studies with their reference numbers
ID	Reference	Year	Citations	Publication Channel	Publication Source
S1	[23]	2013	1	Conference	Asia-Pacific Software Engineering Conference (APSEC)
S2	[24]	2013	7	Conference	IEEE International Requirements Engineering Conference (RE)
S3	[25]	2011	7	Conference	IEEE Ninth European Conference on Web Services
S4	[26]	2018	10	Conference	International Requirements Engineering Conference (RE)
S5	[27]	2011	4	Conference	International Requirements Engineering Conference
S6	[28]	2010	17	Conference	Hawaii International Conference on System Sciences
S7	[29]	2017	6	Conference	International Requirements Engineering Conference Workshops (REW)
S8	[30]	2013	49	Conference	IEEE International Requirements Engineering Conference (RE)
S9	[31]	2011	10	Workshop	International Workshop on Managing Requirements Knowledge
S10	[32]	2010	25	Conference	IEEE International Requirements Engineering Conference
S11	[33]	2018	2	Conference	International Requirements Engineering Conference (RE)
S12	[34]	2014	24	Conference	International Requirements Engineering Conference (RE)
S13	[35]	2010	0	Conference	Annual Computer Software and Applications Conference Workshops
S14	[36]	2012	9	Conference	Asia-Pacific Software Engineering Conference
S15	[37]	2014	11	Workshop	International Workshop on Empirical Requirements Engineering (EmpiRE)
S16	[37]	2012	3	Conference	IEEE International Requirements Engineering Conference (RE)
S17	[38]	2020	0	Workshop	CEUR Workshop Proceedings
S18	[39]	2019	0	Conference	International Conference on Power Electronics, Control and Automation,
S19	[40]	2019	0	Conference	Euromicro Conference on Software Engineering and Advanced Applications
S20	[41]	2019	1	Report	Lecture Notes in Computer Science
S21	[42]	2018	5	Workshop	CEUR Workshop Proceedings
S22	[43]	2017	4	Report	Lecture Notes in Business Information Processing
S23	[44]	2017	8	Report	Lecture Notes in Computer Science
S24	[45]	2017	10	Report	Lecture Notes in Computer Science
S25	[46]	2016	6	Journal	Journal of Telecommunication, Electronic and Computer Engineering
S26	[47]	2014	1	Conference	Australasian Conference on Information Systems
S27	[48]	2014	6	Conference	Frontiers in Artificial Intelligence and Applications
S28	[49]	2014	3	Journal	Interdisciplinary Journal of Information, Knowledge, and Management
S29	[50]	2013	13	Conference	ACM International Conference Proceeding Series
S30	[51]	2012	3	Journal	International Journal of Digital Content Technology and its Applications
S31	[52]	2012	3	Conference	IEEE International Workshop on the Twin Peaks of Requirements and Architecture
S32	[53]	2012	53	Journal	Empirical Software Engineering
S33	[54]	2012	2	Conference	Proceedings of the IADIS International Conference TPMC 2012
S34	[55]	2012	2	Conference	ACM International Conference Proceeding Series
S35	[56]	2012	19	Conference	International Conference on Software and System Process
S36	[57]	2011	0	Journal	International Arab Journal of Information Technology
S37	[58]	2010	7	Conference	Proceedings - International Computer Software and Applications Conference
S38	[59]	2010	5	Conference	ACM International Conference Proceeding Series
S39	[60]	2010	20	Journal	IET Software
S40	[61]	2010	4	Journal	Journal of Networks
S41	[62]	2013	4	Journal	Cybernetics and Information Technologies
S42	[63]	2016	3	Conference	International Conference on Knowledge Based and Intelligent Information and Engineering Systems
S43	[64]	2011	17	Conference	International Conference on Web Engineering
S44	[65]	2015	6	Conference	IEEE international conference on requirement engineering
S45	[66]	2013	7	Report	ACM SIGSOFT Software Engineering Notes
S46	[67]	2012	13	Symposium	IEEE Symposium on Computers & Informatics (ISCI)
S47	[68]	2010	5	Symposium	Proceedings of the Second Asia-Pacific Symposium on Internet ware
S48	[69]	2012	15	Conference	International Conference on software engineering
S49	[70]	2011	1	Conference	International Conference on software engineering
S50	[71]	2016	1	Symposium	ACM Symposium on Engineering Interactive Computing Systems
S51	[72]	2015	24	Journal	Springer journal of requirement engineering
S52	[73]	2011	14	Journal	Journal of computer science and technology
S53	[74]	2015	0	Journal	Journal of software engineering
S54	[75]	2010	10	Conference	International conference on cognitive informatics
S55	[76]	2013	29	Journal	Journal of expert system with application
S56	[77]	2011	5	Journal	Journal of software
S57	[78]	2010	24	Conference	IEEE international conference on advance computing
S58	[79]	2014	24	Conference	International conference on information system
S59	[80]	2015	22	Journal	International journal of decision support system
S60	[81]	2015	62	Journal	Journal of internet services and application
S61	[82]	2021	0	Conference	UK Academy for information Technology international conference
S62	[83]	2021	0	Conference	International conference on evaluation and novel approach to Software Engineering
S63	[84]	2021	1	Journal	IEEE Transactions on Network Science and Engineering
S64	[85]	2021		Conference	Conference of the North American Chapter of the Association for Computational Linguistics (NAACL)
S65	[86]	2021	0	Journal	IEEE Access
S66	[87]	2021	0	Journal	New Trends in Mathematical Sciences
S67	[88]	2020	1	Journal	Group Decision and Negotiation

4.1 RQ1: What are the existing SRN techniques proposed by the selected studies?

In this study, many SRN techniques were discovered as presented in Table 7. Finding shows that 50 techniques were identified from the primary selected studies. From the techniques, 8 of the primary studies used agent-based requirements negotiation technique (S50, S55, S56, S57, S59, S61, S65, S66 and S67) followed by wikiwinwin (S6, S14), winbook (S35, S48), and TAICOS (S17, S20) with 2 studies each. Finally, the remaining 46 techniques were used by 46 of the selected primary studies.

An agent is an automated negotiation tool that improves negotiation by facilitating stakeholders to organize behavior and achieve agreement [89]. The agent allows geographically separated stakeholders to resolve conflict via utility function [71]. Garcia and Sebastia proposed a multi-agent system in which an agent can serve as single stakeholder [76]. The technique implements a UserAgent that can be configured to exhibit the behaviour desired by the corresponding stakeholder, UserAgents negotiate agent which builds a group profile to satisfy 'stakeholders' minimum requirements and Negotiator agent that facilitate the negotiation process. Similarly, study of Cao developed an agent-based decision making model to automate SRN process [77]. This model resolves the conflicting behaviour between multi-agents. The study also proposed an algorithm for running the decision model. In another study, Guar et al. proposed an agent oriented approach that adopted a spiral model to handle stakeholders' multi-dimensional goals on the software to be developed [78]. The approach extends the agent using a User category card to elicit and display requirements.

Furthermore, the approach uses an agent card to help the stakeholders identify the software requirements in terms of agent. The negotiation process obtains a comprehensive view of the participating stakeholders over conflicting requirements, leading to ascertain, correct, prioritize, and provide a comprehensive list of requirements. The study of Schmid et al. used design science research methodology to develop the Negotiation Bot for Electronic Negotiation [82]. The study scope was limited to developing software agents called Bot that imitate human behaviour and makes them interesting for interactive processes and improved the stakeholders communication. Gokay and Erdam present a new framework that incorporates a brokered pricing protocol for agent-based negotiations and a pricing model study for data driven web services hosted on a cloud environment [87]. The study of salah et al. propose Maut technique that allows the incorporation of a bilateral multi-attribute negotiation process with stakeholder selection technique to improve the negotiation concession and unique competent stakeholder proposal [87]. The agents, the broker and the unique stakeholder exchange proposals and counter-proposals during SRN to reach a mutual agreement.

Arrais-Castro et al. developed a Collaborative Negotiation Platform for the distributed stakeholders in the software ecosystem to select businesses [80]. The approach works based on the dynamic multi-criteria decision model (DMCDM). The study of Vahidov et al. investigated the performance of human negotiation using stakeholders in comparison with SRN using software agents [71]. This investigation was experimented to determine the possibility of adopting software negotiation agents in real ' 'stakeholder's transactions. Although the experiment was conducted online instead of in the lab, the result shows a positive sign in adopting the agent automated negotiation in software ecosystems.

Furthermore, the study by Wu et al. developed a wiki-win-win technique which allowss stakeholders to collaborate in a distributed environment and establishes a win-win approach to form a shared understanding of the requirements between the stakeholders [28]. The Wiki-win-win contain the webpages linked together for the requirements elicitation, issues, and options. Meanwhile, the prioritization details, glossary terms, and negotiated agreements are captured in separate web pages. The arrangements are recorded by updating each of the requirements, issues, and options web pages. The wikiwin is very effective in SRN, but not friendly to non-technical users which can fail a software project. This drawback led to the development of winbook by [56]. Winbook was developed using social network sites similar to Gmail and Facebook. This technique used artifact of winwin model to provide collaborative means for the stakeholders to achieve agreement. Winbook has a feature to capture the rationale of the stakeholders and the technique is friendly to both technical and non-technical stakeholders. Haindl et al. present the Task-Interest-Constraint-Satisfying (TAICOS) technique that hierarchically divides the stakeholders requirements into tasks to give a comprehensive view of the requirements conflict to achieve agreements [38]. In another study, Syeff et al. extend the easy-winwin model by combining it with sustainability concept [26]. The last 2 Steps of easy win-win are modified to support sustainability. In this technique, it is assumed that the stakeholders still want to keep requirements. Still, they consider the outcomes of the discussion and modify the original requirements to minimize the negative impacts of sustainability dimensions. Fricker and Glinz proposed a negotiation process with handshaking. This technique implement proposals that measure requirements and design volatility for the stakeholder to understand requirements [32]. Similarly, Alharti et al. presented an ontological win-win approach to negotiate software requirements. The process adds ontological characteristics to the easywinwin SRN model [47]. The ontological winwin enables testing SRN documents to assess the efficiency and effectiveness of the 'stakeholders' communication. In Calefato et al. winwin, wiki, and JSPwiki are adopted to develop JSPWikiwinwin [53]. The technique improves the JSPwiki with a compatible database, higher security, and negotiation efficiency. In a study similar to winbook technique, Syeff et al. used an artifact in easywinwin to propose a social network site approach that allows users to use social sites like Facebook to participate in SRN [81].

Similarly, [30] use social sites to develop web-based techniques called requirements bazaar that engage the stakeholders in the SRN process. Ahmad et al. extend winwin to develop SRN conceptual model that will assist conflict detection and resolution [46]. The technique used rule-based reasoning to provide the stakeholders with an opportunity to add rationale of the requirements to facilitate the negotiation process. In another study by [70], a high-level negotiation framework used to deploy an empirical investigation to assess the effectiveness of negotiation in requirements engineering was proposed.

Moreover, the approach in [37] assists in viewing the requirements of the stakeholders from other disciplines that are difficult to understand. Additionally, Farshidi et al. proposed DSST that adopts MoSCoW technique to describe and detect the conflicts in the assigned priorities to the domain feature requirements by decision-makers [90]. The DSST provides a discussion and negotiation platform to enable Software Producing Organizations to make group decisions. Also Lenz and Schoop proposed a decision matrix to enable decision support for the stakeholders through problem identification [43]. On one hand, Mairiza et al. integrate multi-criteria decision analysis and TOPSIS to value the stakeholders requirements based on weighing scale [79]. The multi-criteria decision analysis is based on the vector space model of computation. In another study by Peng et al., The authors proposed a new and effective approach called requirement maturity measurement approach (RMMA) based on artifacts in wiki. The approach can measure requirements stability during SRN [58]. RMMA also supports requirements selection when there is a large number of stakeholders to participate in the SRN process.

Elrakaiby et al. proposed calculus requirement engineering (CaRE) technique to solve SRN problem [33]. CaRE is well suited for capturing the arguments between stakeholders and requirements engineers during the requirements engineering process. In another study, Carvallo and France presented a Quality model to support processes of SRN [27]. The approach supports both emerging and initial requirements elicited. Hence, the approach detects and resolves the conflict between the requirements.

Similarly, Yang and Liang proposed the extended i* framework that used power relationships between the stakeholders to reason about stakeholder groups for SRN [23]. The reasoning results will help SRN activity. A Valen software ecosystem negotiation technique was developed based on dimensions in the software ecosystem (Valen 2013). The technique helps in resolving conflict in the software ecosystem negotiation activities. In a study by Urbieta et al., the authors proposed a Web requirement meta-model techniques to help in conflict identification and detection in stakeholders requirements [64].

To manage requirement changes in SRN, Zhang et al. propose a compromise-based negotiation framework to manage requirements change when there is deadlock in SRN results [62]. The compromise is a negotiation strategy that helps avoid the deadlock, contributing to the negotiation success. Additionally, Kushiro et al. proposed an extended goal graph to manage requirements and identify conflicting requirements, solutions, criteria, and stakeholders [63]. The extended goal graph constructed an environment for analyzing quality of requirements meeting. Similarly, Ghosh et al. proposed a collaborative and context aware framework for requirement management (C-FaRM) that helps manage the requirement knowledge from the negotiation artifacts to ease and improve SRN [31]. Bulajic et al. improve SRN effectiveness by proposing a generalization requirements approach that validates requirements [49]. Certain rules guide the SRN process for specifying the details of the requirements to produce a code that automatically validates the requirements. Menghi et al. proposed COVER technique to support the goal from the requirements analyst to keep it updated and alive [45]. The technique allows the goal model to be compared with the required interest to resolve conflict. In a study by Mu et al., ' 'Booth's negotiation framework was developed to provide effective and flexible approaches to manage requirements changes by the stakeholders [73]. Alharti and Salim proposed a Zest-based visualization technique to help visualise SRN and resolve conflicts [57]. The technique utilized the Zest algorithm used in visualization of standard email discussion for the conduction of SRN visualization. In another study by Xu et al., an automated service negotiation framework was proposed for SRN [35]. The framework includes the generic service negotiation protocol that improves the negotiation effectiveness. Ahmad et al. proposed a fuzzy CASB strategy that used triangular fuzzy numbers and weights for ranking requirements [39].

On the other hand, Minhas et al. proposed 'Stakeholders' Weight Vote Priority (SWVP) SRN technique that allows stakeholders to conduct SRN based on 'stakeholders' weight, vote, and priority [40]. De Lange et al. proposed a novel approach called NRT collaborative modeling used for real-time web-based collaboration and web application generation [44]. The approach provides an effective service interface that improves the communication between stakeholders. In another study by Lupeikiene and Caplinskas, a view-based approach was presented to derive balanced quality service requirements from the set of goals elicited by the stakeholders [48]. Kukreja and Boehm proposed a two step prioritization technique to prioritize requirements of the stakeholder and system to resolve conflict [50]. The system requirements were divided into minimal marketable features (MMF), and each subdivide is further divided into the requirements. However, the decomposition of the system and user requirements reduces the communication gap between the SRN participants.

Similarly, Petrova-Antonova developed Dynamic QoS-aware web service composition (DYSCO) that automates web service composition processes [55]. In another study by Petrova-Antonova proposed an architectural view that automates the web service composition process [55]. It starts with the definition of requirements and finishes with the production of executable business process. Oster et al. developed a Goal-Oriented Requirements Engineering (GORE) framework that used conditional importance networks (CI-net) for reasoning preference in goal models [65]. The GORE improves the usability and scalability of CI-nets to analyze and specify requirements preferences in SRN. Bajaj and Arora proposed a Fuzzy Analytic Hierarchy Process (FAHP) & Alpha cut approach that extend the capability of FAHP to capture the stakeholder requirement. The approach used Eigenvalue and alpha cut to derive the stakeholders weight [66]. In a study by Yi et al., a Web-based Collaborative Feature Modeling (CoFM) technique was developed to perform SRN [68]. In CoFM, the stakeholders negotiate the brainstormed requirements collaboratively using the framework's distributed feature. Knauss et al. provided an Automatic identification technique to signal top management about the risk associated with elicited requirements [72]. This is to avoid issues in the software project developed that can result in negative consequences on the Software Company and lead to re-negotiation. The approach works based on automatic analysis of 'stakeholders' online requirement communication. Carod and Cechich proposed a Cognitive driven technique that used cognitive psychology, AHP, and map to come up with a framework that determined the stakeholders knowledge level before negotiation commencement [75]. The study of Ishaya and Kuldar used clustering techniques and Expert-based techniques to propose a novel framework for conflict identification and resolution. The framework resolves the conflict triggered by the multiple stakeholders elicited requirements in a particular problem [83].

Finally, Dubois et al. proposed a Broker-based technique where the SRN are handled by a third party (external negotiation brokers) [25]. In conclusion, we found that 11 studies (S7, S18, S22, S23, S24, S30, S37, S44, S45, S61, and S66) did not clearly state the SRN technique that their studies adopted.

Table 7

Existing Techniques of SRN
Technique	ID
Agent	S50, S55, S56, S57, S59, S61,S65, S66
Wikiwinwin	S6, S14
Winbook	S35, S48
TAICOS	S17, S20
Extended i*	S1
SECO-Based RN Model	S2
Broker-based BS	S3
Winwin with Sustainability	S4
Quality Models	S5
RE & CrowdRE	S7
Requirements Bazaar	S8
C-FaRM	S9
Winwin with handshaking	S10
CaRE	S11
Automated RN framework	S13
Pattern-based approach	S16
Fuzzy CASB	S18
SWVP	S19
DSST	S21
Decision matrix	S22
NRT collaborative modeling	S23
COVER	S24
Winwin with rule-based reasoning	S25
Ontological winwin	S26
View-based approach	S27
Generalized requirement approach	S28
Two-step prioritization approach	S29
DYSCO	S33
Architectural view process	S34
Zest algorithm	S36
RMMA	S37
JSPWikiWinWin	S38
Compromise-based Framework	S41
Extended goal graph	S42
Web requirement meta-model	S43
GORE	S44
FAHP & Alpha cut.	S45
CoFM	S47
High-level RN model	S49
Automatic clarification identification	S51
Booth’s RN framework	S52
Negotiation meta-strategy	S53
Cognitive driven technique	S54
MCDA and TOPSIS	S58
Social network sites	S60
Conflict Identification and Resolution Framework (CIRF)	S62
Novel Blockchain Framework	S63
CaSiNo	S64
MAUT technique	S67

4.2 RQ2: What are the processes involved in software requirements negotiation?

The basic processes used in the conduct of SRN as defined in [91] are: win conditions, issues, options, and agreements. Requirements were elicited as win conditions, followed by identifying conflicting requirements as issues, then revealing the alternatives options, and the stakeholders achieved mutual agreements based on options. Some primary studies adopted the basic negotiation artefact to propose their SRN processes (S4, S6, S14, S17, S22, S25, S26, S35, S38, S42, S48, S49, and S50). Additionally, the remaining studies define different processes that their studies used for their SRN. The major processes in SRN as identified from the selected studies are highlighted in Table 8.

Table 8

SRN Processes
ID	Processes	ID	Process
S1	a) Classes and Object Properties of Ontology for Reasoning Stakeholder group b) Rules of Reasoning about Stakeholder c) Using Reasoning Results to Assist Requirements Negotiation	S2	a) Negotiation phase b) SECO’s social dimension c) Software Platform Management
S3	a) Stakeholder requirement b) User handler c) Optimization engine d) Requirements Negotiator e) Modified goal	S4	a) Review and Expand Negotiation Topics b) Brainstorm Stakeholder Interests c) Converge on Win Conditions d) Capture a Glossary of Terms e) Prioritize Win Conditions f) Reveal Constraints and Assumptions g) Identify Issues and Options h) Negotiate Agreements
S5	a) Identification of platform b) Construction of quality model c) Evaluation of alternatives using the quality model d) Analysis and issue of recommendations	S6,S14	a) Agreement b) Brainstorm idea c) Prioritization to point of agreement d) Voting from agreement e) Sequence number f) Reduce number of taxonomy categories g) Added site map h) Tagging multiple contributors i) Tagging multiple win conditions j) Added related topic
S8	a) Requirements specification b) Workflow for co-creation c) Workspace integration d) Personalizable requirements prioritization	S9	a) Artifact creation and alerting b) Ontology mapping c) Tagging recommendation d) Requirements tagging e) XML based dossier f) Requirement analysis, review and negotiation
S10	a) Early-phase requirements with goal models b) Late-phase requirements with a solution oriented c) System specification d) traceability between requirements and implementation proposals	S11	a) Argument construction b) Rebuttal attack c) Argument construction d) Refinement
S13	a) Preparation of negotiation process b) Negotiation process c) Post negotiation	S17	a) Capturing user task b) Collecting stakeholder interest c) Tailoring stakeholder interest d) Specifying constraints
S18	a) Identify participants b) functional and non-functional requirements c) Obtaining of DM’s by utilizing L-1 R-1 operation and GMI illustration for discovering the goals of an institute d) Prioritize these goals according to the participants	S19	a) The weight assignment process b) The communication parameter setting process c) The value assignment process d) The prioritization process e) The negotiation process
S21	a) Requirement and priorities b) Inference engine processing c) Knowledge base processing d) Ranked and feasible solution.	S22	a) Objective definition b) Negotiation issue c) Negotiation alternative d) Attribute negotiation alternative
S24	a) Goal model design b) Requirement specification c) Binding task to system event	S25	a) Register glossary b) Post win conditions c) Register preference d) Identify issue e) Prioritize issue f) Address issue g) Register options
S26	a) Elaborate taxonomy b) Capture domain language c) Map negotiation result to taxonomy d) Taxonomy element	S28	a) Requirements elicitation b) Source code generation based on requirement specification c) Working software in RN process
S29	a) Goal articulation and prioritization b) MMF decomposition c) Win conditions capture and prioritizations	S33	a) Manage negotiation between QoS conditions and provides b) Creation of SLA c) QoS metrics monitoring
S35, S50	a) Win condition b) Issue c) Option d) Equilibrium e) Agreement	S36	a) Produce zest view visualization b) Add stakeholder year of experience c) Classify stakeholder role d) Mixed visualization e) Grid view visualization f) Statistical visualization
S37	a) Browsing b) Modifying c) Marking d) Voting on e) Commenting	S38	a) Win condition b) Issue c) Option d) Agreement
S41	a) Requirement Presentation b) Requirements division based on priority c) Define solution d) Compromise algorithm	S42	a) Define requirement b) Manage functional and non-functional requirement c) Identify solutions d) Detect implicit solution and criteria
S43	a) Requirement gathering b) Requirement modelling c) Structural analysis d) Semantic analysis e) Conciliation process f) Refinement	S45	a) Computation of eigenvalue from the stakeholder b) Comparison of conflicting requirement c) Integrating stakeholders need using FAHP d) Computing alpha cut and total ordering e) Complexity value of conflicting requirement
S47	a) Communicating b) Resolving conflicts c) Submitting modeling operations	S48	a) Win condition b) Issue c) Option d) Equilibrium e) Agreement
S49	a) Define and share the glossary b) Identify conflicts c) Share perspectives, views, and expectation on the requirements d) Assess the system feasibility e) Justify the requirements need f) Prioritize the requirements	S54	a) Determine elicitation preference b) Determine cognitive profile c) Assign importance values d) Analyze results
S55	a) Negotiation object b) Negotiation protocol c) Negotiation strategy	S56	a) Preparation b) Negotiation process c) Negotiation agent
S57	a) Elicitation b) Definition c) Negotiation d) Validation e) Specification	S58	a) Construct the normalized decision matrix b) Construct the weighted normalized decision matrix c) determine the positive and negative ideals d) calculate the separation measures e) rank the preference
S59	a) Underlying MCDM model b) Criteria fusion & alternatives rating	S61	a) Problem Identification b) Objectives of solution c) Design and Development d) Implementation e) Evaluation f) Conclusion
S62	a) Definition b) Incomplete blocks c) Correction for ties d) Significance test	S63	a) Prepare b) Exchange c) Bargain d) Conclude e) Execute

4.3 RQ3: What are the evaluation mechanisms in the selected SRN studies?

To answer this RQ, we found that 55 studies out of the 67 have used evaluation mechanisms. The selected primary studies are categorized according to the evaluation mechanism used to assess their findings. The identified evaluation mechanisms are case studies, experiment, empirical, exploratory, and expert review. According to the distribution presented in Table 9, 30 (44%) of the entire papers used case study evaluation, followed by an experiment with 20 (30%). Hence, this shows the case study evaluation as a more prepared mechanism in SRN research. One of the reasons is probably because it involves stakeholders from the software industry in the evaluation. Some studies adopt empirical, exploratory, and expert evaluation with 4%, 3%, and 1%, respectively. Also interesting to observe is the lack of an adequate description of the evaluation mechanism in 12 (18%) of the reviewed studies.

Table 9

Evaluation mechanisms in primary studies
Evaluation Mechanism	Studies
Case Study	S2, S3,S4, S5, S10, S12, S13, S15, S16, S17, S19, S20, S24,S31, S34, S37,S38, S41, S42, S45, S46, S47, S51, S52, S55, S56, S57, S62, S63,S67.
Experiment	S6,S7,S18,S23,S28,S29, S35,S39,S40, S41, S46,S48, S49, S50, S54, S58, S59,S64,S65,S66
Emprical Evaluation	S30,S32, S61
Exploratory evaluation	S8,S60
Expert Review	S21
Not Stated	S1,S9,S14,S22,S25,S26,S27,S33,S36,S43,S44,S53

4.4 RQ4: What are descriptions and limitations of the existing SRN techniques?

This RQ describes the limitations of the existing SRN techniques reviewed in this SLR. The identified limitations would serve as a direction for future studies in the research domain. The finding shows the existence of SRN techniques is effective, but several several limitations still several limitations exist which need to be addressed by veteran researchers. These limitations are: 'stakeholders' communication gap, scalability, managing requirements changes, conflict resolution and decision-making.

The first limitation is the communication gap between the stakeholders during the SRN. The SRN process is dependent on the stakeholder's communication [72]. This SLR observed that SRN techniques such as Wikiwinwin and TAICOS are not friendly to the non-technical stakeholders, lack efficiency, and work based on a synchronous mode of communication (in the presence of stakeholders) [28]. Prominent techniques such as winbook [56] and JSPwikiwinwin [59]. were developed to address this limitation. Although these techniques improve the efficiency, the interface of the wikiwinwin proved unsatisfactory for the SRN stakeholders. Another limitation is the 'stakeholders' communication in the software ecosystem (SECO). SECO is defined as """"a set of businesses functioning as a unit and interacting with a shared market for software and services, together with the relationships among them"""" [92]. The organization's collaborations in the software ecosystem improve the competitive advantage of the organization and the ecosystem. An effort has been made in [50] where requirements are divided into minimal marketable features (MMF), and each subdivide is further divided into the requirements to reduce the 'stakeholders' communication gap in SECO. The study of de Lange et al. improves the collaborative stakeholders in his approach for real-time [44].

Similarly, [47] developed ontological winwin to assess the efficiency and effectiveness of the stakeholders communication. Nevertheless, more techniques are needed to improve 'stakeholders' communication in SRN and support organizations' collaboration in the software ecosystem. Another Limitation identified by this study is the issue of scalability in some existing SRN techniques. In the study of Haindl et al. it is reported that the scalability of TAICOS is limited to small-scale software engineering [38]. However, the study proceeded to recommend that the future work to focus mainly on ensuring TAICOS scalability on large-scale software engineering projects and handle various needs of the SRN participating stakeholders. The techniques like winwin with sustainability, winwin with handshaking, ontological winwin shows how these SRN techniques were scalable to be enlarged and integrated with other SRN processes. This study recommends future studies to extend SRN techniques into many real-time processes.

Further, there exist requirements changes in management and validation limitations. The software requirements are inherently immune to changes; almost 50% are altered throughout the process of software development [93]. Modifying requirements effectively minimizes the software project's time, resources, and cost [94]. The effective SRN technique enables requirements changes when there is doubt in stakeholders decisions [95]. This study found that most SRN techniques have limited features to allow requirements changes. A compromise-based negotiation framework proposed in [62] does not fully handle requirements changes in negotiation deadlock. Moreover, the authors in [31, 63] proposed C-FaRM and extended goal graphs to support requirements changes and management. This study further found that the limitations still exist and need attention from novel researchers in the SRN research domain.

Equally, conflict identification and detection is another limitations discovered by this study. Conflicts are negative undesired interactions due to the failure of one or more stakeholders to effectively detect the kind of knowledge to seek in SRN [47]. A large proportion of the studies reviewed addressed the issue of conflict detection and resolution. The authorThe authors have proposed many techniques have proposed many techniques such as web requirement meta-model, RMMA, COVER, TAICOS, C-FaRM, and quality model yet these techniques fail to resolve requirements conflict efficiently. The causes of requirements conflict include an inadequate interface, policy conflict, goal conflict, violation of assumptions, and concurrency [47, 96, 97]. The SRN was used when conflicts were detected and cannot be ignored. Moreover, it will help to make an effective decision that is critical to the success of the stakeholders [96, 98]. The identified limitations include varying goals of multiple stakeholders [46]. This might be due to a lack of defined negotiation team problems that addressed the stakeholders' misunderstanding [99]. The clustering techniques and Expert-based technique was incorporated in (study of Ishaya and Kuldar) to proposed a novel framework for conflict identification and resolution. It is believed that the framework would resolve conflicts and stakeholders' requirements. The last limitations of the existing SRN techniques identified by this study are decision-making. These limitations are due to some issues that need extra attention by the ' 'stakeholder's before the SRN commencement. First, problem identification and preference by the stakeholders [43]. ]. Secondly, communication by the set of collaborative organizations in SECO. In SECO, requirements are elicited jointly by the organization representative instead of being brainstormed by individual stakeholders [92]. Hence, agile collaboration networks are needed for business organizations' implementation and for software industries to grow beneath their geographical location [80]. Similarly, the issue of team selection and formation process in the software project. The project manager often neglect members' opinions in the team formation process. An apparent negotiation team problem that will resolve the misunderstanding and provide mutual decision between the team manager and the members is needed. The study of Wang et al. proposed an agent-based SRN technique that iterates negotiation processes until the requirements of the manager and the members are satisfied [99]. In the study of Arrais-Castro et al. a dynamic multi-criteria decision model was used to improve decision-making in the software ecosystem [80]. Finally, although there is a lot of decision-making in SRN, there is little understanding about how stakeholders make decisions in requirement engineering. The study of [82] is evaluated by comparing the Bot with an existing agent. The result shows that although the Bot sometimes provides unsuitable arguments, the Bot imitates human behaviour well and ensures coherent communication processes..

4.5 RQ 5: What are the characteristics and the demographics of the selected studies?

This sub-section presents the characteristic and demographic of the 67 selected papers used in this SLR. Table 4 gives the overview of the primary selected studies.

4.5.1 Publication Frequency

Figure 3 highlights the evolution of publication in SRN classified by their publication year. The trend in publication from the year 2010–2012 is positive. Although there is a slight decrease in 2011, 28 publications were found. However, from 2013 the publications are declining consistently up to 2016. The justification is that top-ranking requirements engineering conferences from CORE (IEEE international conference in requirements engineering) do not publish sufficient papers in SRN within these years (see Table 6). Only 3 papers were published (S2, S8, S28) by the conference. Moreover, in 2017 we observed little increase, with stable linear publication in 2018 and 2019 respectively. Finally, although our SLR found were published in 2020 (S17, S67), there isonly two were published in 2020 (S17, S67), there is a significant increase in paper publication in the SRN research domain in 2021 with six papers as of june.

4.5.2 Quality Score and Publication Channels

The quality assessment questions were formulated to ascertain how the primary studies answered our research questions. Hence, the assessment result would help the researchers in the SRN research domain to identify the relevant studies in terms of the quality score, which clearly stated their SRN techniques, processes, limitations and evaluation mechanism (see Table 7). With regard to the quality assessment in our SLR, Table 7 presents the quality assessment conducted in the study. We found that 9 (14%) studies (S2, S3, S4, S5, S6, S8, S10, S29, and S38) possessed a quality score of greater or equal to 6. Furthermore, 15 (23%) studies (S1, S20, S21, S35, S41, S44, S48, S49, S52, S54, S55, S59, S63, S65 and S67) have a quality score of 5 and above. 36% of the total studies possessed a quality score of 5 and above.

On the publication channels (As shown in Table 5), a significant number of studies were published in Conference proceedings (36), followed by Journals publications with (19), and technical reports with (5). A limited number of Workshop proceedings (4) were published and finally, Symposium proceedings (3). Also, It is worth noting that 28% of primary selected studies have journal articles, while the conference, workshops, symposium and technical report proceedings have the remaining 72% of the total articles used in the study

4.5.3 Publication Source

Table 10

Quality Assessment Result of the Primary Studies
ID	Q1	Q2	Q3	Q4	Q5	Q6	Total Quality Score
S1	1	1	0	1	1	1	5
S2	1	1	1	1	1	1.5	6.5
S3	1	1	1	1	1	1	6
S4	1	1	1	1	1	1.5	6.5
S5	1	1	1	1	1	1.5	6.5
S6	1	1	1	1	1	1	6
S7	1	0.5	1	0	0	0	2.5
S8	1	1	1	1	1	1.5	6.5
S9	1	1	1	1	1	0	5
S10	1	1	1	1	1	1.5	6.5
S11	1	1	0	1	0	1.5	4.5
S12	1	0	1	0	0	1.5	3.5
S13	1	1	1	1	0	0	4
S14	1	1	0	0	1	1	4
S15	1	0	1	0	0	0	2
S16	1	1	1	0	0	1.5	4.5
S17	1	1	1	1	1	0	5
S18	1	0.5	1	1	0	0	3.5
S19	1	1	1	1	0	0.5	4.5
S20	1	1	1	1	1	0	5
S21	1	1	1	1	1	0	5
S22	1	0.5	0	1	0	0	2.5
S23	1	0.5	1	0	1	0	3.5
S24	1	0.5	1	1	0	0	3.5
S25	1	1	0	1	0	0	3
S26	1	1	0	1	0	1	4
S27	1	1	0	0	0	0	2
S28	1	1	1	1	0	0	4
S29	1	1	1	1	1	1.5	6.5
S30	1	0.5	1	0	0	0.5	3
S31	1	0	1	0	0	0	2
S32	1	0	1	0	0	0	2
S33	1	1	0	1	0	0	3
S34	1	1	1	0	0	1.5	4.5
S35	1	1	1	1	0	1.5	5.5
S36	1	1	0	1	0	1	4
S37	1	0.5	1	1	1	0	4.5
S38	1	1	1	1	1	1.5	6.5
S39	1	0	1	0	0	1	3
S40	1	0	1	0	0	2	4
S41	1	1	1	1	1	0	5
S42	1	1	1	1	0	0	4
S43	1	1	0	1	0	0.5	3.5
S44	1	0.5	0	1	1	1.5	5
S45	1	1	1	1	0	0	4
S46	1	0	1	0	0	0	2
S47	1	1	1	1	0	0	4
S48	1	1	0	1	1	1	5
S49	1	1	1	1	0	1	5
S50	1	0	1	0	0	1.5	3.5
S51	1	1	0	0	0	0	2
S52	1	1	1	0	1	1	5
S53	1	1	0	0	0	0	2
S54	1	1	1	1	1	0	5
S55	1	1	1	1	1	0	5
S56	1	1	1	1	1	1	6
S57	1	1	0	1	1	0	4
S58	1	0.5	1	1	1	0	4.5
S59	1	1	1	1	1	0	5
S60	1	1	1	0	1	0	4
S61	1	1	1	0	0	1	3
S62	1	1	0	1	0	1	4
S63	1	1	0	0.5	1	1.5	5
S64	1	0.5	1	0	0	0.5	3
S65	1	0	1	1	1	1.5	5.5
S66	1	0	1	0.5	1	1	4.5
S67	1	1	0	0.5	1	1.5	5

The impact of SRN in software project development cannot be overemphasized. Based on our SLR, the following findings were identified.

Our study found that research publications in SRN have been declining, with an average of 4 publications from reputable journals and proceedings per year. Although the publication from 2010–2012 is positive, we strictly observed a decrease in publication since 2013. Our studies also revealed that SRN research activities started declining in 2013 and continued to decrease until 2016. However, publications increase in 2017, with a stable and linear number of publications in 2018 and 2019. Finally, we found that only 2 papers were published in 2020, but the publication in SRN domain increased significantly in 2021.

A large proportion of the primary studies are published in China, Germany and the USA. The SRN conference proceedings comprised 42% of the total studies, were mostly published in IEEE and ACM, and were affiliated to universities in China, Germany and the USA. Additionally, from the 67 primary selected studies, 66 are affiliated to learning institutions, with only 1 affiliated to the software industry. We strongly believe that our study would improve the attention of the researchers in software development industries to be focusing on SRN domain for their future research.

Our study has found that SRN research and adoption is only growing substantially within China, Germany and USA with few other countries such as Switzerland, Brazil, Spain, Tunisia and the United Kingdom, and has been non-existent in underdeveloped regions such as Africa. Therefore, future research should investigate the barriers to SRN adoption in other parts of the globe such as Africa. On the quality assessment, we found that 14% of the selected studies have a quality score of 6 and above, whereas 23% have a quality score of greater than or equal to 5. 36% of the total studies possessed a quality score of 5 and above. This shows the quality of the primary selected studies.

On SRN techniques, we found that agent-based technique is the most commonly used technique as it has been utilized by 8 of the 67 selected studies. However, TAICOS, winbook, and wiki techniques have 2 studies each from the primary studies. We observe the rise of SRN techniques proposals that leverage agent-based negotiation [72, 77, 78]. Using agents reflects the current trend of machine learning in software engineering. Observing the improved performance especially between distributed stakeholders in the software ecosystem, we think that there would be more publications on agent-based SRN techniques in the future. An additional observation was that 82% of the selected studies adopt different techniques for conducting SRN activities. However, we discovered that most of the techniques are hybrid-based and composed of the basic SRN models [25, 32, 46]. The process models used are winwin, easywinwin, spiral, broker-based, and so on. This study could predict that there would be more work on hybrid-based techniques in years to come as it generates the SRN techniques with improved performance and efficiency.

On the processes used in SRN activities, we found that 63% of primary studies clearly stated the processes in their SRN techniques, with 22% adopting the basic SRN processes in [91] to propose and formulate their SRN processes. This could be attributed to the fact that the basic processes adopted eases the SRN process and quickly allows the stakeholders to achieve agreements. Additionally, the remaining studies define different processes that they used to resolve the SRN conflict. It helps the participating stakeholders save most of the time taken and eliminates the barriers between them.

This study discovered that the most utilized evaluation mechanism in the selected SRN studies is case study with 44%, followed by experimental evaluation, which has 30% of the publications. The result revealed that few studies exhibit more than 1 evaluation mechanism that we classified in both mechanisms. Additionally, empirical evaluation, exploratory evaluation and expert review utilized 4%, 3% and 1% each, respectively. Furthermore, SRN researchers mostly use case study evaluation mechanisms. One possible explanation of this finding is from the assertion that case study evaluation is more preferable compared to other evaluation mechanisms because it investigates phenomena in its real-life context [100].

We have identified the limitation in the existing RN techniques in response to RQ4. The identified techniques have several challenges such as ' 'stakeholder's communication gaps, decision-making, managing requirements changes, conflict resolution and scalability. Various approaches were introduced to resolve these limitations to improve the efficiency of the SRN and reduce the communication gaps between the SRN stakeholders. These include winbook [56] and JSP wikiwinwin [59] and ontological winwin [47]. Additionally, agent techniques were developed to address the software organizations collaboration in SECO [71]. However, more approaches are needed from the veteran researchers to help stakeholders minimize the communication gap, and resolve conflict. The SRN techniques such as web requirement meta-model, RMMA, COVER, TAICOS, C-FaRM, and quality model fail to aid stakeholders detect and resolve conflict effectively. This study recommends the future studies to focus on investigating the root cause of the requirements conflict. This study found that most of the SRN techniques have limited features to allow requirements changes and half of the stakeholders requirements are changed during the software development process. This is a visible limitation that needs further research by providing novel SRN techniques that will handle changes in requirements. Likewise, another limitation in SRN techniques is decision-making [99]. In SECO, decisions are often taken by the representative of an organization; each stakeholder in the organization has varying needs on the software project to be developed. Future SRN techniques should have a self-organizing team to handle decisions of all the stakeholders.

Overall, it is obvious that SRN techniques addressed various SRN challenges. Converging and reviewing the elicited requirements are some features that make SRN techniques feasible. The prioritization of requirements helps the stakeholders in decision-making and resolving the conflict. Exploring objective criteria processes also helps to reduce the communication gap between the stakeholders. The adoption of agent-based negotiation techniques improve the performance, support collaborative negotiation and provides the mutual understanding between the stakeholders.

The limitations of this review have to be considered to have an overall analysis of the results gained from this review. Therefore, the main threats to the validity of this SLR are twofold: the incompleteness of study search and the biases on the study selection. Hence, in this section, all these threats are discussed.

Firstly, concerning the incompleteness of study search, key studies can be missed in retrieving the studies. This can affect the general completeness of the study search. Therefore, to alleviate this Threat and ensure that all significant and prospective studies have been covered, a general search was done on five databases (see Table 3). These data sources contain many Journals, Conference, Workshop, and Symposium proceedings in SRN domain that are indexed. Furthermore, the selected studies were searched backwards and forward to ensure that significant studies are included. Even though we took measures to enhance the completeness of the study search, the study can still suffer from selection bias. This is because other libraries such as EI Compendex, Springer, and Citeceerx were not considered.

Secondly, concerning the study selection process, to decrease bias by researchers, we formulated a very clear and precise IC/EC criteria. Each researcher can have a different understanding of the IC/EC criteria, hence, the study selection results of each researcher are possibly going to differ. To mitigate this bias, we conducted a pilot selection to ensure that an agreement between the researchers is attained on the general understanding of the criteria. The possible mismanagement of duplicate studies is another threat as well. This Threat may have marginally changed our results. 32 possible duplications were identified and thoroughly assessed to determine if they are the same study. In addition, to select a study, the two researchers who conducted the search process take the final decision. Therefore, any disagreement between the two researchers will be fixed between them through discussion until a tangible agreement is established. Furthermore the remaining researchers will review the final selected studies. For this study, we only include peer-reviewed studies. However, there is a likelihood that we might miss some vital non-peer-reviewed study in the SRN domain.

This study aims to utilize the SLR and identify the state of art in SRN. The SLR methodology adopted facilitates formulating the research questions based on the study objectives. The research questions were formulated to identify and ascertain existing SRN techniques and their limitations, SRN processes, evaluation mechanisms and general demography in the selected studies. Prisma protocol was used in the study search and selection process to ensure relevant studies were used. The research initiation, research question formulation, study selection, quality assessment, and data extraction processes were executed in line with the review protocol. Quality assessment was conducted on identified studies to improve thestudy inclusion criteria. This paper is set to equip the research community in requirements engineering with detailed knowledge of the SRN techniques, processes, limitations, and evaluation mechanisms. Many SRN techniques have been proposed, yet improvements are still needed. The future research enhancements are needed on stakeholder's communication gap, conflict resolution, managing requirements changes, scalability, and decision-making.

Furthermore, 5 evaluation mechanisms identified are case study, experiment, exploratory evaluation, empirical evaluation and expert review. The last three mechanisms were the least utilised by the selected primary studies. Finally, this study highlights the future research direction based on the gap identified in the literature.

Ethical Approval

Hereby, I Shamsu Abdullahi consciously assure that for the manuscript SOFTWARE REQUIREMENTS NEGOTIATION: A SYSTEMATIC LITERATURE REVIEW the following is fulfilled:

1) This material is the authors' own original work, which has not been previously published elsewhere.

2) The paper is not currently being considered for publication elsewhere.

3) The paper reflects the authors' own research and analysis in a truthful and complete manner.

4) The paper properly credits the meaningful contributions of co-authors

5) The results are appropriately placed in the context of prior and existing research.

6) All authors have been personally and actively involved in substantial work leading to the paper, and will take public responsibility for its content.

Competing interests

The authors declare that there no any form of competing interest in this paper.

Authors' contributions

The research objectives was formulated by the first Author, Data extraction was conducted by the first and third author, the fourth Author guide the first Author in conducting the Quality Assessment. Similarly, the second, third, fourth and six author validate and improve the quality assessment. First author draft the first draft of the manuscript through the help of the fifth author.

Funding

The Postgraduate study of Shamsu Abdullahi at University of Malaya, Malaysia is funded by Tertiary Education Trust Fund Nigeria through department of Computer Science Hassan Usman Katsina Polytechnic, Katsina State Nigeria. The funders had no role in study decision to publish, or preparation of this manuscript.

Availability of data and materials

The authors declare that this paper did not used dataset

Bozyiğit, F., et al., Linking software requirements and conceptual models: A systematic literature review. 2021. 24(1): p. 71–82.
Pohl, K., Requirements engineering fundamentals: a study guide for the certified professional for requirements engineering exam-foundation level-IREB compliant. 2016: Rocky Nook, Inc.
Davis, K.J.I.J.o.P.M., A method to measure success dimensions relating to individual stakeholder groups. 2016. 34(3): p. 480–493.
Kukreja, N., Using Social Networking Technology to Improve Collaborative Requirements Elicitation, Negotiation, Prioritization and Evolution. 2015, University of Southern California.
Urbinati, A., et al., Stakeholder management in open innovation projects: a multiple case study analysis. 2020.
Braun, P., et al., E-negotiation systems and software agents: Methods, models, and applications, in Intelligent decision-making support systems. 2006, Springer. p. 271–300.
Sofian, H.B., et al., A requirements negotiation process model that integrates EasyWinWin with quality assurance and multi-criteria preference techniques. 2014. 39(6): p. 4667–4681.
In, H.P. and D.J.J.U. Olson, Requirements Negotiation Using Multi-Criteria Preference Analysis. 2004. 10(4): p. 306–325.
Pitula, K. and T.J.R.E. Radhakrishnan, On eliciting requirements from end-users in the ICT4D domain. 2011. 16(4): p. 323.
Lenz, A., M. Schoop, and G. Herzwurm. Electronic requirements negotiation–a literature survey on the state-of-the-art. in UK Academy for Information Systems Conference, Proceedings of UKAIS. 2016.
Kitchenham, B., et al., Systematic literature reviews in software engineering–a systematic literature review. 2009. 51(1): p. 7–15.
Terpstra, E.J.S.U.R.E.-j., A Systematic Literature Review of Requirements Negotiation Methods from 2010 till 2015. 2015. 1.
Tito, L., et al. A Systematic Mapping of Software Requirements Negotiation Techniques. in ICEIS (2). 2017.
Kersten, G.E.J.I.R.I., Montréal, Canada, The science and engineering of e-negotiation: Review of the emerging field. 2002.
Riaz, S., et al., A Systematic Literature Review of Software Requirements Negotiation Practices. 2010.
Hutchinson, J. and G. Kotonya. A review of negotiation techniques in component based software engineering. in 32nd EUROMICRO Conference on Software Engineering and Advanced Applications (EUROMICRO'06). 2006. IEEE.
Pacheco, C., I.J.J.o.S. Garcia, and Software, A systematic literature review of stakeholder identification methods in requirements elicitation. 2012. 85(9): p. 2171–2181.
Keele, S., Guidelines for performing systematic literature reviews in software engineering. 2007, Citeseer.
Zakari, A., et al., Multiple fault localization of software programs: A systematic literature review. 2020: p. 106312.
Zakari, A., et al., Spectrum-based Fault Localization Techniques Application on Multiple-Fault Programs: A Review. 2020.
Beecham, S., et al., Protocol for a systematic literature review of motivation in software engineering. 2006.
Achimugu, P., et al., A systematic literature review of software requirements prioritization research. 2014. 56(6): p. 568–585.
Yang, H. and P. Liang. Reasoning about Stakeholder Groups for Requirements Negotiation based on Power Relationships. in 2013 20th Asia-Pacific Software Engineering Conference (APSEC). 2013. IEEE.
Valen, G. Requirements negotiation model: A social oriented approach for software ecosystems evolution. in 2013 21st IEEE International Requirements Engineering Conference (RE). 2013. IEEE.
Dubois, E., K. Kritikos, and S. Kubicki. An automatic requirements negotiation approach for business services. in 2011 IEEE Ninth European Conference on Web Services. 2011. IEEE.
Seyff, N., et al. Tailoring requirements negotiation to sustainability. in 2018 IEEE 26th International Requirements Engineering Conference (RE). 2018. IEEE.
Carvallo, J.P. and X. Franch. Requirements negotiation for multilayer system components. in 2011 IEEE 19th International Requirements Engineering Conference. 2011. IEEE.
Wu, D., D. Yang, and B. Boehm. Finding success in rapid collaborative requirements negotiation using wiki and shaper. in 2010 43rd Hawaii International Conference on System Sciences. 2010. IEEE.
Schneider, K., et al. Reframing societal discourse as requirements negotiation: vision statement. in 2017 IEEE 25th International Requirements Engineering Conference Workshops (REW). 2017. IEEE.
Renzel, D., et al. Requirements bazaar: Social requirements engineering for community-driven innovation. in 2013 21st IEEE International Requirements Engineering Conference (RE). 2013. IEEE.
Ghosh, S., A. Dubey, and S. Ramaswamy. C-FaRM: A collaborative and context aware framework for requirements management. in 2011 4th International Workshop on Managing Requirements Knowledge. 2011. IEEE.
Fricker, S. and M. Glinz. Comparison of requirements hand-off, analysis, and negotiation: case study. in 2010 18th IEEE International Requirements Engineering Conference. 2010. IEEE.
Elrakaiby, Y., A. Ferrari, and J. Mylopoulos. Care: A refinement calculus for requirements engineering based on argumentation semantics. in 2018 IEEE 26th International Requirements Engineering Conference (RE). 2018. IEEE.
Valença, G., et al. Competition and collaboration in requirements engineering: A case study of an emerging software ecosystem. in 2014 IEEE 22nd International Requirements Engineering Conference (RE). 2014. IEEE.
Xu, T., et al. Negotiating Service Requirements among Strategic Actors. in 2010 IEEE 34th Annual Computer Software and Applications Conference Workshops. 2010. IEEE.
Peng, R. and H. Lai. DRE-specific wikis for distributed requirements engineering: a review. in 2012 19th Asia-Pacific Software Engineering Conference. 2012. IEEE.
Maalej, W., Z. Kurtanović, and A. Felfernig. What stakeholders need to know about requirements. in 2014 IEEE 4th International Workshop on Empirical Requirements Engineering (EmpiRE). 2014. IEEE.
Haindl, P., R. Plösch, and C. Körner. Tailoring and Evaluating Non-Functional Interests Towards Task-Oriented Functional Requirements. in Software Engineering (Workshops). 2020.
Ahmad, S., S. Mehfuz, and J. Beg. Fuzzy Cloud Access Security Broker for Requirements Negotiation and Prioritization. in 2019 International Conference on Power Electronics, Control and Automation (ICPECA). 2019. IEEE.
Minhas, N.M., et al. SWVP-A Requirements Prioritization Technique for Global Software Development. in 2019 45th Euromicro Conference on Software Engineering and Advanced Applications (SEAA). 2019. IEEE.
Haindl, P., R. Plösch, and C. Körner. A Research Preview on TAICOS–Tailoring Stakeholder Interests to Task-Oriented Functional Requirements. in International Working Conference on Requirements Engineering: Foundation for Software Quality. 2019. Springer.
Farshidi, S., et al. Multiple Criteria Decision Support in Requirements Negotiation. in REFSQ Workshops. 2018.
Lenz, A. and M. Schoop. Decision Problems in Requirements Negotiations–Identifying the Underlying Structures. in International Conference on Group Decision and Negotiation. 2017. Springer.
de Lange, P., et al. Engineering web applications using real-time collaborative modeling. in CYTED-RITOS International Workshop on Groupware. 2017. Springer.
Menghi, C., P. Spoletini, and C. Ghezzi. Integrating goal model analysis with iterative design. in International Working Conference on Requirements Engineering: Foundation for Software Quality. 2017. Springer.
Ahmad, S., et al., An Enhancement of Software Requirements Negotiation with Rule-based Reasoning: A Conceptual Model. 2016. 8(10): p. 193–198.
Alharthi, M., J. Campbell, and C. McDonald. An Ontological Win-Win Model for Requirements Negotiation: Visual Decision-Making Aid for Software Development Teams. 2014. ACIS.
Lupeikiene, A. and A. Caplinskas. Requirements Engineering for Service-Oriented Enterprise Systems: Quality Requirements Negotiation. in DB&IS. 2014.
Bulajic, A., et al., The Generalized Requirement Approach for Requirement Validation with Automatically Generated Program Code. 2014. 9.
Kukreja, N. and B. Boehm. Integrating Collaborative Requirements negotiation and prioritization processes: a match made in heaven. in Proceedings of the 2013 International Conference on Software and System Process. 2013.
Ahmad, S., et al., Requirements Negotiation: Does Consensus Reduce Software Development Cost? 2012. 6(22): p. 46.
Wu, Y., et al. Towards understanding requirement evolution in a software product line an industrial case study. in 2012 First IEEE International Workshop on the Twin Peaks of Requirements and Architecture (TwinPeaks). 2012. IEEE.
Calefato, F., D. Damian, and F.J.E.S.E. Lanubile, Computer-mediated communication to support distributed requirements elicitations and negotiations tasks. 2012. 17(6): p. 640–674.
Petrova-Antonova, D. and S. Ilieva. DYSCO: A Platform for Dynamic QoS-Aware Web Service Composition. in IADIS International Conference on Theory and Practice in Modern Computing. 2012.
Petrova-Antonova, D. and S. Ilieva. An architectural approach for dynamic web service composition. in Proceedings of the 13th International Conference on Computer Systems and Technologies. 2012.
Kukreja, N. and B. Boehm. Process implications of social networking-based requirements negotiation tools. in 2012 International Conference on Software and System Process (ICSSP). 2012. IEEE.
Alharthi, M. and S.S.J.I.A.J.o.I.T. Salim, Zest Based Visualization for Requirements Negotiation. 2011. 8(3): p. 318–325.
Peng, R., Q. Ye, and M. Ye. A requirements maturity measurement approach based on sklsewiki. in 2010 IEEE 34th Annual Computer Software and Applications Conference Workshops. 2010. IEEE.
Wan, P., et al. Balancing load of shaper in WikiWinWin requirements negotiation environment: an empirical evaluation. in Proceedings of the 11th International Conference on Product Focused Software. 2010.
Erra, U. and G.J.I.s. Scanniello, Assessing communication media richness in requirements negotiation. 2010. 4(2): p. 134–148.
Dong, W.J.J.o.N., The research on web service based network management. 2010. 5(7): p. 849.
Zhang, Y., et al., A negotiation framework for managing the requirements changes. 2013. 13(Special-Issue): p. 75–87.
Kushiro, N., T. Shimizu, and T.J.P.C.S. Ehira, Requirements elicitation with extended goal graph. 2016. 96: p. 1691–1700.
Urbieta, M., et al. Detecting conflicts and inconsistencies in web application requirements. in International Conference on Web Engineering. 2011. Springer.
Oster, Z.J., G.R. Santhanam, and S. Basu. Scalable modeling and analysis of requirements preferences: A qualitative approach using CI-Nets. in 2015 IEEE 23rd International Requirements Engineering Conference (RE). 2015. IEEE.
Bajaj, P. and V.J.A.S.S.E.N. Arora, Multi-person decision-making for requirements prioritization using fuzzy AHP. 2013. 38(5): p. 1–6.
Rodina, A., T. Amjed, and M.K. Zarinah. An empirical assessment of the use of different communication modes for requirement elicitation and negotiation using students as a subject. in 2012 IEEE Symposium on Computers & Informatics (ISCI). 2012. IEEE.
Yi, L., et al. CoFM: a web-based collaborative feature modeling system for internetware requirements' gathering and continual evolution. in Proceedings of the Second Asia-Pacific Symposium on Internetware. 2010.
Kukreja, N. Winbook: a social networking based framework for collaborative requirements elicitation and WinWin negotiations. in 2012 34th International Conference on Software Engineering (ICSE). 2012. IEEE.
Ahmad, S. and N.A. Muda. An experimental design to exercise negotiation in requirements engineering. in International Conference on Software Engineering and Computer Systems. 2011. Springer.
Vahidov, R., R. Saade, and B. Yu. Effects of negotiation tactics and task complexity in software agent: human negotiations. in Proceedings of the 18th Annual International Conference on Electronic Commerce: e-Commerce in Smart connected World. 2016.
Knauss, E., et al., Patterns of continuous requirements clarification. 2015. 20(4): p. 383–403.
Mu, K.-D., et al., Managing software requirements changes based on negotiation-style revision. 2011. 26(5): p. 890.
Jamoussi, Y.J.J.o.S.E., Enhancing satisfaction of actor's requirements in web service composition: A guided negotiation based approach. 2015. 9(3): p. 429–450.
Carod, N.M. and A. Cechich. Cognitive-driven requirements prioritization: a case study. in 9th IEEE International Conference on Cognitive Informatics (ICCI'10). 2010. IEEE.
Garcia, I. and L.J.E.S.w.A. Sebastia, A negotiation framework for heterogeneous group recommendation. 2014. 41(4): p. 1245–1261.
Cao, M.J.J., Decision Making Model for Intelligent Agent in Automated Negotiation. 2011. 6(8): p. 1537–1544.
Gaur, V., A. Soni, and P. Bedi. An agent-oriented approach to requirements engineering. in 2010 IEEE 2nd International Advance Computing Conference (IACC). 2010. IEEE.
Mairiza, D., D. Zowghi, and V. Gervasi, Utilizing TOPSIS: A multi criteria decision analysis technique for non-functional requirements conflicts, in Requirements Engineering. 2014, Springer. p. 31–44.
Arrais-Castro, A., et al., Collaborative negotiation platform using a dynamic multi-criteria decision model. 2015. 7(1): p. 1–14.
Seyff, N., et al., Using popular social network sites to support requirements elicitation, prioritization and negotiation. 2015. 6(1): p. 7.
Schmid, A., et al., Are you for real? A Negotiation Bot for Electronic Negotiations. 2021.
Gambo, I. and K. Taveter. Identifying and Resolving Conflicts in Requirements by Stakeholders: A Clustering Approach. in ENASE. 2021.
Cong, X., et al., DTNB: A blockchain transaction framework with discrete token negotiation for the delay tolerant network. 2021.
Chawla, K., et al., CaSiNo: A Corpus of Campsite Negotiation Dialogues for Automatic Negotiation Systems. 2021.
Najjar, A., et al., One-to-Many Negotiation QoE Management Mechanism for End-User Satisfaction. 2021. 9: p. 59231–59243.
Akkus, G.B. and E.J.N.T.i.M.S. Ucar, Brokered pricing model for agent-based negotiation of data-driven services in the cloud. 2021. 9(2): p. 38–47.
Salah, S.B., et al., A Bilateral Multi-Attribute Negotiation-Based Approach for a VE Configuration. 2020. 29(5): p. 923–947.
Aguiar, L.G.F., et al. Conceptual mapping as a communication tool of legal requirements. in 2016 11th Iberian Conference on Information Systems and Technologies (CISTI). 2016. IEEE.
Farshidi, S., et al., A decision model for programming language ecosystem selection: Seven industry case studies. 2021: p. 106640.
Boehm, B., et al. Software requirements negotiation and renegotiation aids: A theory-W based spiral approach. in 1995 17th International Conference on Software Engineering. 1995. IEEE.
Jansen, S., A. Finkelstein, and S. Brinkkemper. A sense of community: A research agenda for software ecosystems. in 2009 31st International Conference on Software Engineering-Companion Volume. 2009. IEEE.
Ahmad, Z., et al. Impact minimization of requirements change in software project through requirements classification. in Proceedings of the 9th International Conference on Ubiquitous Information Management and Communication. 2015.
Bhatti, M.W., et al. A methodology to manage the changing requirements of a software project. in 2010 International Conference on Computer Information Systems and Industrial Management Applications (CISIM). 2010. IEEE.
Fricker, S., et al., Handshaking with implementation proposals: Negotiating requirements understanding. 2010. 27(2): p. 72–80.
Jamoussi, Y.J.J.o.S.E., Enhancing satisfaction of Actors' requirements in Web service composition: A Guided negotiation based Approach. 2015. 9(3): p. 1819–4311.
Weiss, M., B. Esfandiari, and Y.J.C.n. Luo, Towards a classification of web service feature interactions. 2007. 51(2): p. 359–381.
Yan, J., et al., Autonomous service level agreement negotiation for service composition provision. 2007. 23(6): p. 748–759.
Wang, J. and J.J.E.A.o.A.I. Zhang, A win–win team formation problem based on the negotiation. 2015. 44: p. 137–152.
Inayat, I., et al., A systematic literature review on agile requirements engineering practices and challenges. 2015. 51: p. 915–929.

No competing interests reported.

Appendix.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Software Requirements Negotiation: a Systematic Literature Review

Status:

Version 1

Abstract

Figures

1. Introduction

2. Related Work

3. Research Methodology

3.1 Research Questions

3.2 Data Source

3.3 Search String

3.4 Study Selection Procedure

3.5 Inclusion and Exclusion Criteria

3.6 Quality Assessment

3.7 Data Extraction

4. Result And Discussion

4.1 RQ1: What are the existing SRN techniques proposed by the selected studies?

4.2 RQ2: What are the processes involved in software requirements negotiation?

4.3 RQ3: What are the evaluation mechanisms in the selected SRN studies?

4.4 RQ4: What are descriptions and limitations of the existing SRN techniques?

4.5 RQ 5: What are the characteristics and the demographics of the selected studies?

4.5.1 Publication Frequency

4.5.2 Quality Score and Publication Channels

4.5.3 Publication Source

5. Research Findings

6. Threat To Validity

7. Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1