18-08-2012, 03:02 PM
DEVELOPMENT OF INTELLIGENT SYSTEMS FOR PERFORMANCE ANALYSIS AND DECISION MAKING USING FUZZY LOGIC
DEVELOPMENT OF INTELLIGENT SYSTEMS .doc (Size: 605.5 KB / Downloads: 28)
ABSTRACT
With incorrect decision, the efficiency and effectiveness of an organization may degrade therefore decision making for an employee continuance or removal from an organization is a critical issue. There is a need for certain algorithm, tools, techniques etc which can provide precise and accurate information for the performance analysis and decision making. This paper presents a method for performance analysis of faculty members which has given us more precised and accurate results. This analysis is divided into three parts i.e. input fuzzification, rule evaluation and defuzzification. All the steps have been tested on a number of faculty members and the experimental results have demonstrated a fast, robust, and reliable analysis simulation. The faculty performance analysis inputs are taken as “Feedback from students”, “Self-development effort“, “proficiency in teaching” and “management feedback”. The output is obtained in term of “overall performance” of the faculty which is utilized as standard for decision making. The proposed performance analysis technique is simulated using fuzzyTECH 5.7 developed by Inform Software Corporation.
INTRODUCTION
Fuzzy Logic control systems Fig. 1 have been reported in wide range of applications that include industrial processes, transportation system, robotics and consumer products. The concepts of fuzzy logic (FL) was conceived by Lotfi Zadeh, a professor at University of California at Berkley, and presented as a way of processing data by allowing partial set membership rather than crisp set membership or non membership [1]. “Figure 1 to be included here". Professor Zadeh reasoned that people do not require precise, numerical information input, and yet they are capable of highly adoptive control. Fuzzy logic is a problem solving control system methodology that lends itself to implementation in system ranging from simple, small, embedded microcontrollers to large network, multi channel PC or work station based data acquisition and control systems. Fuzzy Systems are used in various fields such as game design [2], scoring method [3], clinical practice guidelines [4], autonomous systems [5] and planning systems [6] and etc.
SIMULATION RESULTS & DISCUSSION
This analysis simulation was programmed using Inform Software Corporation’s fuzzyTECH 5.7. Every input was divided into four sets. The simulation results use 256 rules which are placed at Appendix. The summary of results is shown in Table - 1. “Table-1 is to be placed here”. Total 13 input patterns are taken as inputs for 13 faculty members and their corresponding output is evaluated using fuzzy logic system. For example if a faculty member is having a feedback from student is of 4, proficiency in teaching is 7 and self development effort is 7 and management feedback is 5, then the overall performance of faculty member using fuzzy logic is derived out to be 2.332.
CONCLUSION
Fuzzy logic is a problem solving control system methodology that lends itself to implementation in system ranging from simple, small, embedded microcontrollers to large network, multi channel PC or work station based data acquisition and control systems. In this paper, overall performance analysis of 13 faculty members of an Engineering Institution is taken as sample. The overall performances of the faculty members were analyzed and the result is utilized for decision making for continuance or removal of an employee from an organization. This can further be extended as per the requirement of the educational institute / organization. This method provides an easy and precise method for accurately defining and analyzing the overall performance of a faculty member. These methods can also be extended for the performance analysis of students in teaching institutions and also in industries for performance evaluation of employee at end of every year.