IIVIFS-WASPAS: An integrated Multi-Criteria Decision-Making perspective for cloud service provider selection

https://doi.org/10.1016/j.future.2019.09.053Get rights and content

Highlights

  • IIVIFSs based cloud service selection approach is presented.

  • Integrated weight assessment approach to identify the weights of the QoS attributes.

  • Preference attitudinal score and accuracy functions for accurate service ranking.

  • Validations using case studies on a sample trust feedback dataset from Cloud Armor.

  • IIVIFSs validated in terms of sensitivity analysis and rank reversal phenomenon.

Abstract

Cloud service provider selection, an important Multi-Criteria Decision Making (MCDM) problem involves intrinsic relations among the multiple alternatives, attributes and decision experts for the selection of trustworthy Cloud Service Providers (CSPs). Due to uncertain and incomplete nature of the cloud service provider evaluation data, i.e., Quality of Service (QoS) and user feedbacks, the identification of suitable CSPs with accurate service ranking remains an open research challenge. To address the above-mentioned challenge, this work presents an Improved Interval-Valued Intuitionistic Fuzzy Sets-Weighted Aggregate Sum and Product Assessment (IIVIFS-WASPAS) based cloud service provider selection approach for the identification of Trustworthy CSPs (TCSPs). The proposed CSP selection approach employs integrated objective and subjective weight assessment method and IVIFS-WASPAS method to determine the importance of QoS attributes and to rank the TCSPs respectively. Further, a novel preference-attitudinal score and accuracy function of IIVIFS have been designed based on the decision maker’s attitude to rank the TCSPs. Case studies using Cloud Armor, a real-world trust feedback dataset demonstrates the accuracy, effectiveness, and feasibility of IIVIFS-WASPAS approach for CSP selection problem in terms of sensitivity analysis and Rank Reversal Phenomenon (RRP).

Introduction

Over the last few decades, rapid advancements in Information and Communication Technologies (ICT) have an enormous impact on the fifth utility computing - ‘Cloud Computing’ through the provision of data-intensive services and computational resources on a subscription basis [1] . The innate advantages of cloud computing have attracted various governmental, academic and business organizations to migrate their business solutions to cloud infrastructures to make their business processes agile with minimal cost and management effort [2]. However, the proliferation of a wide range of Cloud Service Providers (CSPs) offering functionally-equivalent services with different performance and cost makes the selection of an appropriate and Trustworthy Cloud Service Providers (TCSPs) an unsolved research challenge For example, Google Drive, Onedrive, Bitriz24, MyDrive, etc. provides Storage as a Service (StaaS) at different performance (QoS attributes: availability, response time, price, and so on), cost with varied importance on the QoS attributes (Criteria).

Problem Statement and Context: ‘Cloud service provider selection’, a significant and interesting research problem due to the existence of a wide range of CSPs (alternatives) and uncertain assessment data (objective and subjective). In general, the performance of the CSPs is well-reflected by the QoS attributes. A user-service QoS evaluation/performance matrix which contains QoS values or users’ feedback values over the QoS attributes forms the major data source for CSP evaluation. Moreover, trustworthiness is a quality metric that expresses the performance of the CSPs with respect to different QoS attributes. Several trust-based service selection approaches based on rough set theory [3], [4], [5], [6], [7], evidence theory [8], probability theory [9], [10], fuzzy set theory [11], etc., have been developed for the design of a robust cloud service selection model.

CSP selection problem can be formulated as a Multi-Criteria Decision-Making (MCDM) problem due to the intrinsic relationship among the multiple QoS attributes, alternatives and decision makers’ opinions [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22]. Recent research works based on Fuzzy based MCDM (FMCDM) approaches [23], [24], [25] like Intuitionistic Fuzzy Set (IFS) [26], [27], [28], [29], Neutrosophic Fuzzy Set (NFS) [30], Hesitant Fuzzy Set (HFS) [31], etc., demonstrated their effectiveness in dealing with uncertain data and ambiguous decisions for cloud service provider selection. However, determining the trustworthiness of the service providers with respect to a set of user-specific QoS attributes is complex due to personal biases in the users’ preferences, imprecise weights of the QoS parameters, multi-dimensional evaluation process, vagueness in the decision-making process, and Rank Reversal Phenomenon (RRP) [32].

Solution and Novelty: To address the afore-mentioned issues in CSP selection, this work presents an Improved Interval-Valued Intuitionistic Fuzzy Set-Weighted Aggregate Sum and Product Assessment (IIVIFS-WASPAS) approach for the identification of TCSPs. IIVIFS-WASPAS employs (i) IVIFS — analyse the fuzziness of the user preferences, (ii) Integrated weight assessment method — identify the importance of QoS attributes through IVIFS-Shannon Entropy (SE) for objective weight assessment and Rank-centroid for subjective weight assessment and (iii) IVIFS-WASPAS — rank the TCSPs. Further, a novel preference attitudinal score and accuracy function was developed to defuzzify the IVIFS value without loss of information.

The major motive behind the application of IVIFS is that it can handle vague and imprecise trust feedback data (crisp data) by expressing satisfaction, dissatisfaction and uncertainty of the CSPs through membership degree/truthfulness, non-membership degree/falsity, and hesitant degree in the form of intervals. Further, WASPAS aggregates Weighted Sum Model (WSM) and Weighted Product Model (WPM) to enable high ranking accuracy in dynamic environment scenarios. In general, the crisp data has uncertainty since it is hard for the users to express their level of satisfaction/dissatisfaction on the cloud service providers with respect to the QoS attributes. Therefore, the conversion of crisp data into IVIFS enables the users to express their satisfaction, dissatisfaction & uncertainty and thereby enabling accurate cloud service ranking.

Contributions: The major contributions of the IIVIFS-WASPAS are highlighted as follows:

  • 1.

    An Improved Interval Valued Intuitionistic Fuzzy Set (IIVIFS) based cloud service provider selection approach is presented to identify the trustworthy CSPs.

  • 2.

    Uncertainty in the assessment data is handled by converting the user feedbacks (crisp numbers) into IVIFS numbers since it reflects truthfulness, falsity and hesitancy of the CSPs with respect to the QoS attributes.

  • 3.

    A novel integrated objective and subjective weight assessment approach is designed to determine the importance of the QoS attributes for accurate service ranking.

  • 4.

    A novel preference attitudinal score and accuracy functions based WASPAS is designed for service ranking without loss of information.

  • 5.

    The effectiveness of the IIVIFS-WASPAS was demonstrated with the case study using the sample dataset from the Cloud Armor trust feedback dataset in terms of sensitivity analysis and rank reversal phenomenon.

The rest of the paper is organized as follows: Section 2 discusses the literature review about cloud service provider selection using different FMCDM approaches. Section 3 provides a detailed insight into the basics of IVIFS, score and accuracy functions of IVIFS. Section 4 presents a novel preference-attitudinal score and accuracy functions for ranking CSPs and working of the IIVIFS, the proposed cloud service provider selection approach. Section 5 presents the case study using Cloud Armor, a trust feedback dataset to analyse the efficiency and reliability of IIVIFS over the state-of-the-art IVIFS based cloud service provider selection approaches. Section 6 concludes the paper.

Section snippets

Literature review

‘Cloud service provider selection’ is modelled as a MCDM problem since the trustworthiness of the CSPs can be evaluated based on: (i) discovering cloud service providers (Alternatives); (ii) identify QoS attributes (Criteria); (iii) assess the ratings of the CSPs’ and weights of the QoS attributes; (iv) aggregate the ratings of the CSPs and weights of the QoS attributes to compute the trustworthiness of each CSP across the QoS attributes, and (v) select the trustworthy CSP. Recent research

Interval-Valued Intuitionistic Fuzzy Set (IVIFS)

Interval-Valued Intuitionistic Fuzzy Set (IVIFS) formulated by Atanassov and Gargov is defined in terms of interval numbers rather than crisp numbers to express the Membership Degree Function (MDF) and the Non-Membership Degree Function (NMDF) [47]. Further, the Hesitancy Degree Function (HDF) is defined using MDF and NMDF.

Definition 3.1

[48]

Let ϓ be the universe of discourse and ß is any element in ϓ, then IFS (ύ) in ϓ is defined as in Eq. (1). ύ=ß,fύß,ύß|ßϓwhere, fύß and ύß is the MDF and the NMDF of ύ

An improved score and accuracy function of IVIFS based on the preference attitude of assessor/decision maker/user

IIVIFS, the proposed cloud service provider selection approach evaluates the cloud service providers based on the aggregated IVIFS. Therefore, it is necessary to find out the ways to compare two IVIFS for accurate service ranking. There exists several score and accuracy functions to compare two IVIFS, however they fail to identify the exact difference between two IVIFS. To address the above-mentioned challenge, we present a novel preference attitudinal based accuracy and score function with an

Case study

The performance of IIVIFS-WASPAS, the proposed cloud service selection methodology has been evaluated using a sample dataset extracted from the Cloud Armor project, University of Adelaide [61], a real-world trust feedback dataset which focus on the development of a robust trust management framework for the cloud environment. It contains 10,080 user feedbacks for nine QoS attributes (Benefit attribute: Availability (AV), Response Time (RT), Security (Sec), Speed (S), Storage Space (SS), Features

Conclusions

‘Service Selection Problem’ remains a challenging research area in the field of cloud computing due to the emerging new cloud service providers with different functionalities and dynamic user requirements. Further, the uncertainty in the objective and subjective assessment data, a primary source for cloud service evaluation complicates the service selection problem. To address the same, this work presents Improved Interval-Valued Intuitionistic Fuzzy Sets-Weighted Aggregated Sum and Product

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by The Council of Scientific & Industrial Research, India and The Department of Science & Technology, India (Grant No: CSIR - SRF Fellowship/143345/2K17/1 and SR/FST/ETI-349/2013). The authors gratefully acknowledge Tata Realty-IT City-SASTRA Srinivasa Ramanujan Research Cell of SASTRA Deemed University for the financial support extended to us in carrying out this research work.

Obulaporam Gireesha is a Senior Research Fellow at Center for Information Super Highway (CISH) School of Computing, SASTRA Deemed University, Thanjavur, INDIA. She was awarded The Council for Scientific & Industrial Research (CSIR) - Senior Research Fellowship’ 2018. She obtained her Bachelor’s degree in Information Technology from Sri Padmavati Mahila VisvaVidyalayam (SPMVV) Women’s University, Tirupati, India in 2014. She received his Master’s degree in Software Engineering from Sree

References (64)

  • NivethithaS. et al.

    An improved robust heteroscedastic probabilistic neural network based trust prediction approach for cloud service selection

    Neural Netw.

    (2018)
  • NivethithaS. et al.

    A trust centric optimal service ranking approach for cloud service selection

    Future Gener. Comput. Syst.

    (2018)
  • Wen-juanF. et al.

    A multi-dimensional trust-aware cloud service selection mechanism based on evidential reasoning approach

    Int. J. Autom. Comput.

    (2015)
  • JunweiZ. et al.

    A customer-centric trust evaluation model for personalized service selection

    Sci. Program.

    (2018)
  • MehdiM. et al.

    Probabilistic approach for QoS-aware recommender system for trustworthy web service selection

    Appl. Intell.

    (2014)
  • KumarR.R. et al.

    Prioritizing the solution of cloud service selection using integrated MCDM methods under fuzzy environment

    J. Supercomput.

    (2017)
  • GireeshaO. et al.

    WNN-EDAS: A wavelet neural network based multi-criteria decision-making approach for cloud service selection

  • HamzehA. et al.

    Cloud service evaluation method-based multi-criteria decision-making: A systematic literature review

    J. Syst. Softw.

    (2018)
  • MichaelM. et al.

    (MC2)2: criteria, requirements and a software prototype for cloud infrastructure decisions

    Softw. Pract. Exp.

    (2013)
  • NeerajY. et al.

    Two-way ranking based service mapping in cloud environment

    Future Gener. Comput. Syst.

    (2018)
  • ChandrashekarJ. et al.

    SELCLOUD: a hybrid multi-criteria decision-making model for selection of cloud services

    Soft Comput.

    (2019)
  • Ranjan KumarR. et al.

    A novel framework for cloud service evaluation and selection using hybrid MCDM methods

    Arab. J. Sci. Eng.

    (2018)
  • GireeshaO. et al.

    GCRITICPA: A critic and grey relational analysis based service ranking approach for cloud service selection

  • SarbjeetS. et al.

    Compliance-based multi-dimensional trust evaluation system for determining trustworthiness of cloud service providers

    Future Gener. Comput. Syst.

    (2017)
  • JagpreetS. et al.

    Improved TOPSIS method based trust evaluation framework for determining trustworthiness of cloud service providers

    J. Grid Comput.

    (2017)
  • ChandrashekarJ. et al.

    Evaluating the efficiency of cloud services using modified data envelopment analysis and modified super-efficiency data envelopment analysis

    Soft Comput.

    (2017)
  • JagpreetS. et al.

    Design and comparative analysis of MCDM-based multi-dimensional trust evaluation schemes for determining trustworthiness of cloud service providers

    J. Grid Comput.

    (2017)
  • SantosoW. et al.

    Evaluation of cloud services: A fuzzy multi-criteria group decision making method

    Algorithms

    (2016)
  • T. Subramanian, N. Savarimuthu, Cloud service evaluation and selection using fuzzy hybrid MCDM approach in marketplace,...
  • BoopathyA.K. et al.

    E-fpromethee: An entropy based fuzzy multi criteria decision making service ranking approach for cloud service selection

  • Shui-xiaC. et al.

    Cloud-based ERP system selection based on extended probabilistic linguistic MULTIMOORA method and Choquet integral operator

    J. Comput. Appl. Math.

    (2019)
  • Ru-xinN. et al.

    A three-cycle decision-making selection mechanism with intuitionistic trapezoidal fuzzy preference relations

    J. Intell. Fuzzy Syst.

    (2019)
  • Cited by (55)

    View all citing articles on Scopus

    Obulaporam Gireesha is a Senior Research Fellow at Center for Information Super Highway (CISH) School of Computing, SASTRA Deemed University, Thanjavur, INDIA. She was awarded The Council for Scientific & Industrial Research (CSIR) - Senior Research Fellowship’ 2018. She obtained her Bachelor’s degree in Information Technology from Sri Padmavati Mahila VisvaVidyalayam (SPMVV) Women’s University, Tirupati, India in 2014. She received his Master’s degree in Software Engineering from Sree Vidyanikethan Engineering College (SVEC), Tirupati, India in 2016. She is a lifetime member of Ramanujan Mathematical Society. Her current research interests include Cloud service selection, Multi criteria decision making, and Elliptic curve cryptography.

    Nivethitha Somu is an Institute Post Doctoral Fellow at Smart Energy Informatics Laboratory (SEIL), Department of Computer Science and Engineering, Indian Institute of Technology (IIT-B), Bombay. She was a Senior Research Fellow at Center for Information Super Highway (CISH) School of Computing, SASTRA Deemed University, Thanjavur, INDIA (2013–2018). She was awarded The Department of Science and Technology - “Innovation in Science Pursuit for Inspired Research (INSPIRE)” Fellowship’2013. She received her Master’s degree in Science from Anna University, Chennai, INDIA in 2011. She also received her Master’s degree in Technology from SASTRA Deemed University, Thanjavur, INDIA in 2013. She is a lifetime member of Ramanujan Mathematical Society. Her current research interests include Cloud service selection, QoS Prediction, Machine learning, Energy informatics, and Intrusion detection systems.

    Kannan Krithivasan is a Professor in the Department of Mathematics, SASTRA Deemed University, Thanjavur, INDIA. He obtained his Bachelor’s and Master’s degrees from the University of Madras, India, in 1980 and 1982, respectively. He also received his Bachelor’s and Master’s degrees in Education from Madurai Kamaraj University, India, in 1984 and 1986 respectively. He obtained his M.Phil. degree in Mathematics from Regional Engineering College, Tiruchirapalli, India, in 1988. He was conferred Ph.D. in Mathematics in the area of Computational Fluid Dynamics by Alagappa University, Karaikudi, India, in 2000. He is a member of IEEE. He is also a lifetime member of Ramanujan Mathematical Society. He has been in Academia for the past 34 years. His specific areas of interest include Combinatorial optimization, Hypergraph based image processing, and Bayesian computing.

    Shankar Sriram V.S. is a Professor of Information Technology; and Chair Professor for TATA Communications — Cyber Security at School of Computing, SASTRA Deemed University, Thanjavur, Tamil Nadu, INDIA. He received his Bachelor’s degree in Science from Madurai Kamaraj University, Madurai, India. He obtained his Master’s degree in Computer Applications from Madurai Kamaraj University, Madurai, India. He also received his Master’s degree in Engineering from Thapar University, Punjab, India. He was conferred Ph.D. in Information and Network Security from Birla Institute of Technology, Mesra, India. He has been in the Academia for the past 19 years. He is a member of IEEE. He was awarded the IBM Shared University Research (SUR) Award’ 2017. His current area of research includes information and Network security, Cloud computing, Big data analytics, Machine learning, and Graph-based data mining.

    View full text