Implementation Of Fuzzy Approach To Improve Time Estimation [Case Study Of A Thermal Power Plant Is Considered
Abstract
Fuzzy System has demonstrated their ability to solve different kinds of problem in various application domains. There is an increasing interest to apply fuzzy concept to improve tasks of any system. Here case study of a thermal power plant is considered. Existing time estimation represents time to complete tasks. Applying fuzzy linear approach it becomes clear that after each confidence level least time is taken to complete tasks. As time schedule is less than less amount of cost is needed. Objective of this paper is to show how one system becomes more efficient in applying Fuzzy Linear approach.
In this paper we want to optimize the time estimation to perform all tasks in appropriate time schedules. For the case study, optimistic time (to), pessimistic time (tp), most likely time(tm) is considered as data collected from thermal power plant. These time estimates help to calculate expected time(te) which represents time to complete particular task to considering all happenings. Using project evaluation and review technique (PERT) and critical path method (CPM) concept critical path duration (CPD) of this project is calculated. This tells that the probability of fifty percent of the total tasks can be completed in fifty days. Using critical path duration and standard deviation of the critical path, total completion of project can be completed easily after applying normal distribution. Using trapezoidal rule from four time estimates (to, tm, tp, te), we can calculate defuzzyfied value of time estimates. For range of fuzzy, we consider four confidence interval level say 0.4, 0.6, 0.8,1. From our study, it is seen that time estimates at confidence level between 0.4 and 0.8 gives the better result compared to other confidence levels.
 Publication:

International Conference on Modeling, Optimization, and Computing (ICMOC 2010)
 Pub Date:
 October 2010
 DOI:
 10.1063/1.3516344
 Bibcode:
 2010AIPC.1298..433P
 Keywords:

 fuzzy logic;
 power systems;
 functional analysis;
 matrix algebra;
 07.05.Mh;
 84.30.Jc;
 02.30.Sa;
 02.10.Yn;
 Neural networks fuzzy logic artificial intelligence;
 Power electronics;
 power supply circuits;
 Functional analysis;
 Matrix theory