MULTI-CRITERIA DECISION MAKING FOR SELECTING SEMANTIC WEB SERVICE CONSIDERING VARIABILITY AND COMPLEXITY TRADE-OFF

• Published in: Journal of Theoretical and Applied Information Technology Vol. 86; no. 2; p. 316
• Main Authors: Setiawan, Nanang Yudi, Sarno, Riyanarto
• Format: Journal Article
• Language: English
• Published: Islamabad Journal of Theoretical and Applied Information 01.04.2016
• Subjects: Complexity; Decomposition; Mathematical models; Quality; Recall; Semantics; Tradeoffs; Web services
• ISSN: 1817-3195

Business process decomposition helps to focus on the base business process rather than the whole business processes with a higher complexity. Variability of business process decomposition taken into a different complexity level within common business process. By considering business process variability and complexity trade-off, we selected the best solution for semantic web service alternatives with different service quality. The selected web service should be fulfilled most of service quality capability defined in user's preferences. For our case study we had chosen the Penerimaan Peserta Didik Baru, an online student submission system consists the registration procedure for multi-degree of school, multi-selection (regular, inclusion, pre-welfare), and multi-phase for every educational level. The base process is selected from four basic models, and we identified the variant models that representing the behavior of all possible requirements. In this paper, semantic service selection is used by discovering semantic web services through their service registry. Web services was annotated with its business process formalization and quality attributes. However, the variability and complexity in business process decomposition are contradictory, thus, constitute a trade-off. We interpreted it as the multi-criteria decision problem and proposed Fuzzy AHP+TOPSIS method to bring the best optimum solution that reflect the needs and preferences of the decision maker. This approach proved to solve the multi-criteria decision problem for selecting the best options of semantic web services considering the trade-off among variability and complexity of business process. In our study, we had tested 153 service requests and gained a precision of 91.3%, and a recall or sensitivity of 89.4% that result the harmonic mean of precision and recall of 0.903. Our approach is success to deliver the most preferred number of business process variant with minimum complexity level in accordance with the acceptable service quality (service cost, capacity, and latency) delivered by service providers.