BENCHMARKING INTERNET PERFORMANCE USING FUZZY LOGIC APPROACH

Abstract

Abstract

Performance Dimensions (PDs) are variables used in measuring Internet performance. The PDs are usually evaluated using fuzzy logic with the logic serving as a co-intensive model of reality of human reasoning. Though many frameworks for benchmarking the Internet exist, studies addressing users’ satisfaction or employing multiple PDs using fuzzy logic are few. This work was designed to develop an intelligent benchmarking tool that measures University Internet performance from the users’ satisfaction perspective through multiple PDs. The PHP+MySQL was used in developing an Internet performance type of an intelligent benchmarking tool, named University Internet Network Benchmarking (UINB) as an internet-based information system. The UINB tool was used to facilitate data collection, storage and benchmarking of nodes. Fuzzy logic was used to measure performance. The University of Ibadan network (UINET) data on the PDs [Service Quality (SQ), users’ logs, complaints, maintenance, equipment and monitoring] were used for analysing and benchmarking. The reliability of the UINB tool was tested using UINET. Benchmarked Internet performance of nodes clearly classified performance as satisfactory, dissatisfactory and very dissatisfactory. This tool was applied on the internet nodes of the Department of Economics, Faculty of the Social Sciences and the Department of Computer Science. It classified both Faculty of Social Science and the Department of Computer Science as unsatisfactory (-31.5, and -78.9 respectively). The intelligent benchmarking tool solved the problem of benchmarking internet performance. The multiple performance dimensions using fuzzy logic appropriately measured Internet performance through users’ satisfaction. This could be used as Internet benchmarking tool for Universities.