Goal Programming Approach to Solve Multi- Objective Intuitionistic Fuzzy Non- Linear Programming Models

Similar documents
Irene Hepzibah.R 1 and Vidhya.R 2

FUZZY GOAL PROGRAMMING VS ORDINARY FUZZY PROGRAMMING APPROACH FOR MULTI OBJECTIVE PROGRAMMING PROBLEM

CONTINUOUS PARAMETER FREE FILLED FUNCTION METHOD

Interactive Bi-Level Multi-Objective Integer. Non-linear Programming Problem

International Journal of Pure and Applied Sciences and Technology

36.1 Why is it important to be able to find roots to systems of equations? Up to this point, we have discussed how to find the solution to

Chapter 3 Differentiation and Integration

Fuzzy Approaches for Multiobjective Fuzzy Random Linear Programming Problems Through a Probability Maximization Model

LEVEL SET OF INTUITIONTISTIC FUZZY SUBHEMIRINGS OF A HEMIRING

Assortment Optimization under MNL

Fuzzy approach to solve multi-objective capacitated transportation problem

Neutrosophic Goal Programming

International Journal of Mathematical Archive-4(12), 2013, Available online through ISSN

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Local Approximation of Pareto Surface

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017

Kernel Methods and SVMs Extension

Complement of Type-2 Fuzzy Shortest Path Using Possibility Measure

Which Separator? Spring 1

Trajectory Optimization of Flexible Mobile Manipulators Using Open-Loop Optimal Control method

: Numerical Analysis Topic 2: Solution of Nonlinear Equations Lectures 5-11:

Intuitionistic Fuzzy G δ -e-locally Continuous and Irresolute Functions

Absorbing Markov Chain Models to Determine Optimum Process Target Levels in Production Systems with Rework and Scrapping

An Interactive Optimisation Tool for Allocation Problems

Amiri s Supply Chain Model. System Engineering b Department of Mathematics and Statistics c Odette School of Business

Application of B-Spline to Numerical Solution of a System of Singularly Perturbed Problems

Double Layered Fuzzy Planar Graph

Lecture 2 Solution of Nonlinear Equations ( Root Finding Problems )

Speeding up Computation of Scalar Multiplication in Elliptic Curve Cryptosystem

CHAPTER 4 MAX-MIN AVERAGE COMPOSITION METHOD FOR DECISION MAKING USING INTUITIONISTIC FUZZY SETS

EE 330 Lecture 24. Small Signal Analysis Small Signal Analysis of BJT Amplifier

A Hybrid Variational Iteration Method for Blasius Equation

Single Variable Optimization

The Second Anti-Mathima on Game Theory

An Interactive Approach for Solving Multi-Objective Nonlinear Programming and its Application to Cooperative Continuous Static Games

Some Concepts on Constant Interval Valued Intuitionistic Fuzzy Graphs

Problem Set 9 Solutions

CHAPTER 4d. ROOTS OF EQUATIONS

Bayesian decision theory. Nuno Vasconcelos ECE Department, UCSD

Taylor Series Approximation to Solve Neutrosophic Multi-objective Programming Problem

A Simple Research of Divisor Graphs

MMA and GCMMA two methods for nonlinear optimization

DETERMINATION OF TEMPERATURE DISTRIBUTION FOR ANNULAR FINS WITH TEMPERATURE DEPENDENT THERMAL CONDUCTIVITY BY HPM

OPTIMISATION. Introduction Single Variable Unconstrained Optimisation Multivariable Unconstrained Optimisation Linear Programming

A new Approach for Solving Linear Ordinary Differential Equations

A Satisfaction Degree of Optimal Value for. Grey Linear Programming*

A New Algorithm for Finding a Fuzzy Optimal. Solution for Fuzzy Transportation Problems

Reliability-Based Design Optimization Under Stationary Stochastic Process Loads

Some modelling aspects for the Matlab implementation of MMA

On the Multicriteria Integer Network Flow Problem

Yong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 )

Support Vector Machines. Vibhav Gogate The University of Texas at dallas

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis

CISE301: Numerical Methods Topic 2: Solution of Nonlinear Equations

SOLVING CAPACITATED VEHICLE ROUTING PROBLEMS WITH TIME WINDOWS BY GOAL PROGRAMMING APPROACH

Numerical Heat and Mass Transfer

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS

Lecture 13 APPROXIMATION OF SECOMD ORDER DERIVATIVES

Designing of Combined Continuous Lot By Lot Acceptance Sampling Plan

Determine the Optimal Order Quantity in Multi-items&s EOQ Model with Backorder

Lecture 16 Statistical Analysis in Biomaterials Research (Part II)

IIM Shillong PGP ( ) Admissions Process

/04/$ IEEE. 2

The Minimum Universal Cost Flow in an Infeasible Flow Network

CS 331 DESIGN AND ANALYSIS OF ALGORITHMS DYNAMIC PROGRAMMING. Dr. Daisy Tang

University of Washington Department of Chemistry Chemistry 452/456 Summer Quarter 2014

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg

Advanced Topics in Optimization. Piecewise Linear Approximation of a Nonlinear Function

VARIATION OF CONSTANT SUM CONSTRAINT FOR INTEGER MODEL WITH NON UNIFORM VARIABLES

COS 521: Advanced Algorithms Game Theory and Linear Programming

Chapter 2 A Class of Robust Solution for Linear Bilevel Programming

Solutions HW #2. minimize. Ax = b. Give the dual problem, and make the implicit equality constraints explicit. Solution.

Transfer Functions. Convenient representation of a linear, dynamic model. A transfer function (TF) relates one input and one output: ( ) system

Chapter 5 rd Law of Thermodynamics

A Bayes Algorithm for the Multitask Pattern Recognition Problem Direct Approach

Using T.O.M to Estimate Parameter of distributions that have not Single Exponential Family

Bayesian decision theory. Nuno Vasconcelos ECE Department, UCSD

Polynomial Regression Models

NETWORK IDENTIFICATION AND TIME SERIES ANALYIS USING S-SYSTEMS

Test Problem Construction for Single-Objective Bilevel Optimization

Chapter 6. Supplemental Text Material

DESIGN OPTIMIZATION OF CFRP RECTANGULAR BOX SUBJECTED TO ARBITRARY LOADINGS

The optimal delay of the second test is therefore approximately 210 hours earlier than =2.

An Application of Fuzzy Hypotheses Testing in Radar Detection

System in Weibull Distribution

Bayesian predictive Configural Frequency Analysis

An Admission Control Algorithm in Cloud Computing Systems

OPTIMALITY CONDITIONS FOR NONLINEAR PROGRAMMING WITH MIXED CONSTRAINTS AND ρ -LOCALLY ARCWISE CONNECTED FUNCTIONS

Endogenous timing in a mixed oligopoly consisting of a single public firm and foreign competitors. Abstract

Heuristic Algorithm for Finding Sensitivity Analysis in Interval Solid Transportation Problems

Linear programming with Triangular Intuitionistic Fuzzy Number

Estimating the Fundamental Matrix by Transforming Image Points in Projective Space 1

Lagrange Multipliers Kernel Trick

Adaptive multi-point sequential sampling methodology for highly nonlinear automotive crashworthiness design problems

The Quadratic Trigonometric Bézier Curve with Single Shape Parameter

Difference Equations

Formulas for the Determinant

Mixed Taxation and Production Efficiency

Transcription:

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 Goal Prorammn Approach to Solve Mult- Objectve Intutonstc Fuzzy Non- Lnear Prorammn Models S.Rukman #, R.Sopha Porchelv * # Assstant Proessor, Department o Mathematcs, S..E.. Women s Collee, Mannarud, amlnadu, Inda * AssocateProessor, Department o Mathematcs, A.D.M. Collee or Women, Naappattnam, amlnadu, Inda Abstract In ths Paper Mult-objectve Intutonstc Fuzzy Non-lnear Prorammn Problem has been solved throuh oal prorammn approach. Solvn non-lnear prorammn problem strahtaway s qute dcult. So, the Mult-objectve Intutonstc Fuzzy Non-lnear Prorammn Problem s transormed nto Mult-objectve Intutonstc Fuzzy Lnear Prorammn Problem by the use o aylor Polynomal Seres. Intutonstc uzzy oal prorammn approach s dscussed by ormn sutable membershp and nonmembershp uncton n whch the aspraton level and the tolerance levels are ed by the decson maker or by ndn the best and worst soluton o each objectve uncton. Usn the membershp and non-membershp uncton o each objectve, the ven problem was ormulated and the solutons are ound by varous approaches and the obtaned solutons are dentcal. Numercal eample wll llustrate the ecency o the proposed approach. Keywords Non-Lnear Prorammn, Goal Prorammn, Intutonstc Fuzzy Set. I. Introducton Lke Lnear Prorammn, Non- Lnear Prorammn s a mathematcal technque or determnn the optmal solutons to many busness problems. Lnear Prorammn problem s characterzed by the presence o lnear relatonshp n decson varables n the objectve uncton as well as lnear constrants. But n the real le systems, nether the objectve uncton nor the constrants are lnear unctons n decson varables,.e., the relatonshps are non-lnear. Samr Dey et al., proposes an ntutonstc uzzy optmzaton approach to solve a non-lnear prorammn problem n the contet o structural applcaton. Mahapatra.G.S et al., solved a ntutonstc mult-objectve nonlnear prorammn problem n the contet o relablty applcaton. Based on these, Multobjectve Intutonstc Fuzzy Non-lnear Prorammn Problem has been solved throuh oal prorammn approach n ths paper. In the rst secton, Mult-objectve Intutonstc Fuzzy Non-lnear Prorammn Problem s dscussed and aylor Polynomal seres s ntroduced n the subsequent secton. he membershp and non-membershp unctons are ormed ordn to the eptance and rejecton level n the thrd secton. Varous ormulatons o the problem are dscussed n the ourth secton. Step-wse soluton procedure s eplaned n the net secton. Fnally, a numercal eample s ven and t s solved by varous ormulatons n two cases and the obtaned solutons are dscussed at the end. II. Prelmnares A. Formulaton o Mult-Objectve Intutonstc Fuzzy Non-Lnear Prorammn Problem Consder the Mult-objectve Intutonstc FuzzyNon-lnear Prorammn Problem, Ma or MnZ X =,,. k the constrants j = b, =,,, m; j =,,, n ISSN: -7 http://www.jmttjournal.or Pae

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 ISSN: -7 http://www.jmttjournal.or Pae 6 and j, j =,,.., n In the above, s are real valued unctons o n varables,,.., n and each,,. k are ntutonstc uzzy non-lnear unctons. o overcome the computatonal dcultes o solvn non-lnear prorammn problems, the ollown aylor olynomal seres s used to transorm the non-lnear uncton nto lnear uncton. B. aylor Polynomal Seres For each objectve uncton, =,,, k the ndvdual optmzed value =,,.., n s to be determned and transorm the non-lnear objectves, =,,, k nto lnear objectves by the use o aylor Polynomal Seres, P = + + + + n n n By replacn the non-lnear uncton by P o all objectve unctons, =,,.., k become the lnear uncton. Usn aylor Polynomal seres, the ven non-lnear unctons are transormed nto lnear unctons. Accordn to these lnear unctons, the ollown membershp and non-membershp unctons ormed. C. Formulaton o Membershp and non-membershp unctons Case : he aspraton level and tolerance levels are ed by the decson maker Form the membershp and non-membershp unctons or and or, Here, and are eptance and rejecton tolerance or the th oal. Case : he eptance and rejecton levels are ed by ndn the best and worst solutons o each objectve uncton. he membershp and non-membershp uncton o mamzaton problem s dened as ollows:

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 L U L U U re re L re L L re L U U L U re U re he membershp and non-membershp uncton o mnmzaton problem s dened as: U U L L L U re U re L re L re re L U re re U L re U Here, U = ma {Z } and L = mn {Z } or membershp uncton. Moreover, U re = U ε, L re = L and U re = U, L re = L + ε are non-membershp uncton o mamzaton type and mnmzaton type respectvely, ε, =,., k based on the decson maker s choce. III. Problem Formulaton Rajesh Danwal et al., proposed a method to solve Mult-Objectve Lnear Fractonal Prorammn Problem by vaue set. Here, Mult-objectve Intutonstc Fuzzy Non-lnear Prorammn Problem has been solved n whch the membershp unctons can be chosen as per hs method. Furthermore, the non-membershp should be mnmzed and ν are any one o the our possbltes dened above and each o them are reater than or equal to zero. Moreover, n ntutonstc uzzy sets μ + ν, the membershp and non-membershp are ormed separately but the sum o both values are less than or equal to one another. hereore, the ntutonstc uzzy oal prorammn can be ormulated as: Formulaton I For membershp k Ma λ = = μ μ = / μ = or μ = L U L / μ = U U L μ j = b ISSN: -7 http://www.jmttjournal.or Pae 7

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 and j, μ, =,,, m; j =,,., n For non-membershp k Mn λ = = ν ν = / ν = or ν = U U re re / ν = L re L U re re L ν j = b and j, ν, =,,, m; j =,,., n o et the best achevement, the membershp should be mamzed and non-membershp should be mnmzed and n ntutonstc uzzy sets, μ + ν. hereore, the ven Mult-objectve Intutonstc Fuzzy Non-lnear Prorammn Problem can be reormulated as: Formulaton II k Ma λ = = {μ ν } μ ν μ + ν < μ + ν j = b and j, μ, ν, =,,, m; j =,,., n Snce, the hhest deree o membershp s and the lowest deree o non-membershp s. hereore, n oal prorammn approach, Acheve: { / μ + d d + = & ν Z + d d + = }, j In whch, d, d and d +, d + are the under attanment and over attanment, respectvely o the th oal, the devatonal varables d + and d can be removed rom the oal prorammn because the overachevement o membershp uncton and underachevement o non-membershp uncton are eptable. hereore, the ven problem s ormulated as: Formulaton III Mn λ = d + + d μ + d ν d + j = b and j, d, d +, =,,, k; j =,,., n he ollown soluton procedure s used to solve the problem wthout dculty by the above ormulatons. IV. Soluton Procedure ISSN: -7 http://www.jmttjournal.or Pae 8

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 Step : Determne the optmal soluton or each objectve uncton, =,,, k n subject to the ven set o constrants. Step : ransorm the non-lnear objectve uncton nto equvalent lnear objectve uncton at the optmal ponts =,,.. n by aylor Polynomal Seres Step : I the decson maker desres to the oal, then the aspraton level, eptance tolerance and rejecton tolerance or each o the obtaned lnear objectve uncton based on the decson maker s choce. I the decson maker may have the decson deadlock to decde the oals, then the eptance and rejecton levels by ndn the best and worst values o each lnear objectve uncton. Step 4 : Construct the membershp and non-membershp uncton or each lnear objectve uncton wth the help o tolerance and rejecton levels n step. Step : Construct the ntutonstc uzzy oal prorammn models I, II and III as presented n secton III. Step 6 : Solve the ntutonstc uzzy oal prorammn models by the use o optmzaton sotware ORA. he ollown numercal eample wll demonstrate the ecency o the above soluton procedure. It has been solved n two proposed cases and the results are dscussed at the end. V. Numercal Eample Ma = + Ma = 4 + 6 Ma = + subject to + 4 4, +,, By ndn the optmzed solutons or each objectve uncton subject to the ven constrants, ma, =, ma, = 4, ma, = By usn aylor s seres.., the lnear objectve unctons are, = +, +, = 4 +, +, = +, +, he obtaned lnear unctons are, = +, = + +, = + + 4. he ntutonstc uzzy oal prorammn s as ollows: Case : I the oals and tolerance values are decded by the decson maker Accordn to the decson maker s choce,, 6 and he eptance and rejecton tolerance or oal G, oal G and oal G are as ollows: G : = and = G : = and = 4 G : = and = 4 he membershp and non-membershp unctons ormed wth respect to the eptance and rejecton levels dened above are: ISSN: -7 http://www.jmttjournal.or Pae 9

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 ISSN: -7 http://www.jmttjournal.or Pae 6 6 4 Usn these membershp and non-membershp uncton dened above, the problem wll be ormulated by I, II and III as ollows: Formulaton I For membershp Ma λ =. +..7 μ + + = ; μ + + = ; μ + = ; μ ; μ ; μ ; + 4 4; + ; And,, μ, μ, μ he optmal soluton s =., =.67, μ =.8, μ =, μ =.44 For non-membershp, Mn λ =.9 +.7 +.8 ν + + = ; 4ν + + = 4; 4ν + = ; ν ;ν ;ν ; + 4 4; + ; and,, ν, ν, ν he optmal soluton s =., =.67, ν =., ν =, ν =.4 Formulaton II Ma λ =. + 4. 4 + ; + 4; ; + 9; 7 + 7 8; 8; 7 7 ; + ; ; + 4 4; + ;

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 and, he optmal solutons s =., =. 67 By substtutn these values n the correspondn membershp and non-membershp unctons, the ollown values obtaned: μ =. 8, μ =, μ =. 446, ν =., ν =, ν =. 4 Formulaton III Mn λ = d + d + d + d + + d + + + d + + d ; + + d + + + d > 4; + + 4d + 4; + + d ; ; + + 4d + ; + 4 4; + ; and,, d, d, d, d +, d +, d + he optmal solutons are =., =.67, d =.7, d =, d =.6, d., d + =, d + =.4 From the membershp and non-membershp unctons dened above, the values are, μ =.8, μ =, μ =.44, ν =., ν =, ν =.4 Case : he oals and tolerance values are ed by ndn the best and worst values o the lnear unctons. he best and worst values o each o the objectve unctons are, Ma = 4.67; Mn = ; Ma = 6; Mn = ; Ma = ; Mn = ; Accordn to these values, the membershp and non-membershp unctons are, 4.67 4.67 4.67 4 4.67 4.67 6 4.67 4.67 6 6 6 6 6 ISSN: -7 http://www.jmttjournal.or Pae + =

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 Usn these membershp and non-membershp uncton dened above, the problem wll be ormulated by I, II and III or the values ε =., ε =. and ε = as ollows: Formulaton I For membershp Ma λ =.98 +.47 +.66 4.67μ = ; 4μ = ; μ + = ; μ ; μ ; μ ; + 4 4; + ; And,, μ, μ, μ he optmal soluton s =., =.67, μ =, μ =, μ =.44 For non-membershp, Mn λ =.44.64 4.6ν + + = 4.6;.97ν + + =.97; ν + = ; ν ;ν ;ν ; + 4 4; + ; and,, ν, ν, ν he optmal soluton s =., =.67, ν =, ν =, ν =. Formulaton II Ma λ =.8 +.7 +. 8.64 + 7.96.7;.4 +.6 ;.4 +.6.7;.94 +.94.88;.6 +.6 ;.6 +.6.88; 4; ; + 4; + 4 4; + ; and, he optmal solutons s =., =.67 By substtutn these values n the correspondn membershp and non-membershp unctons, the ollown values obtaned: μ =, μ =, μ =.4, ν =, ν =, ν =. Formulaton III Mn λ = d + d + d + d + + d + + + d + + 4.67d 4.67; + + 4.6d + 4.6; + + 4d 4; + +.97d +.97; + + d ; d + ; + 4 4; + ; and,, d, d, d, d +, d +, d + ISSN: -7 http://www.jmttjournal.or Pae

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 he optmal solutons are =., =.67, d =, d =, d =., d + =, d + =, d + =. From the membershp and non-membershp unctons dened above, the values are, μ =, μ =, μ =.4, ν =, ν =, ν =. Comparson o Optmal Solutons or Case able : Approach Soluton Ponts, Achevement Level μ, μ, μ Non-achevement level ν, ν, ν Formulaton I.,.67.8,,.44.,,.4 Formulaton II.,.67.8,,.44.,,.4 Formulaton III.,.67.8,,.44.,,.4 Comparson o Optmal o Solutons or Case able : Approach Soluton Ponts, Achevement Level μ, μ, μ Non-achevement level ν, ν, ν.,.67,,.4,,. Formulaton I Formulaton II.,.67,,.4,,. Formulaton III.,.67,,.4,,. In comparson o the optmal solutons n the three ormulatons under two varous cases, he values are dentcal n all the ormulatons and the possblty o the achevements under case, the oal s 8%, oal s % and oal s 4%. he non-achevement level o oal s %, oal s % and oal s 4%. he values are dentcal n all the ormulatons and the possblty o the achevements under case, the oal s %, oal s % and oal s 4%. he possbltes o non-achevement o the oal s %, oal s % and oal s %..e. the rst two oals are ully acheved and the thrd oal s acheved by 4%. Hence, the solutons are dentcal n all the ormulatons and n addton the second case ves more ecent soluton than the rst one. Moreover, n the second case the epslon ε value can be chaned to et the better soluton. VI. CONCLUSION Mult-objectve Intutonstc Fuzzy Non-lnear prorammn problem has been solved n two varous cases under three ormulatons and the obtaned solutons are compared at the end. aylor s Polynomal Seres s used to convert the non-lnear nto lnear to solve the problem rather easly. he decson maker s achevement level was satsed wth ood percentae n all the objectves. he stepwse soluton procedure wll helpul to the reader to solve the mult-objectve non-lnear prorammn problem wth better soluton. Moreover, the numercal eample clearly eplaned or better understandn o the soluton procedure. REFERENCES [] H.J. Zmmermann 978, Fuzzy prorammn and lnear prorammn wth several objectve unctons, Fuzzy Sets and Systems, vol., 4- ISSN: -7 http://www.jmttjournal.or Pae

Internatonal Journal o Mathematcs rends and echnoloy IJM Volume Number 7 - January 8 [] P.P.Anelov 997, Optmzaton n ntutonstc uzzy envronment, Fuzzy Sets and Systems, vol.86, 99-6 [] R. Sopha Porchelv, An Alorthmc approach to Mult-objectve uzzy lnear prorammn problem, Internatonal Journal o Alorthms, Computn & Mathematcs, Eashwar Publcatons, 4,6-66 [4] Rajesh Danwal, aylor seres soluton o mult-objectve lnear ractonal prorammn problem by vaue set, Internatonal Journal o Fuzzy Mathematcs and Systems, Vol., No., 4- [] Mahapatra.G.S, Intutonstc uzzy mult-objectve mathematcal prorammn on relablty optmzaton model, Internatonal Journal o Fuzzy Systems, Vol-, No-, pp.9-66 [6] Samr Dey and apan Kumar Roy, Intutonstc Fuzzy Goal Prorammn echnque or Solvn Non-Lnear Mult-Objectve Structural Problem, Journal o Fuzzy Set Valued Analyss, No., pp.79-9 [7] Pramank.S and Roy..K, An ntutonstc uzzy oal prorammn approach to vector oprmzaton problem, Notes on Intutonstc Fuzzy sets, Vol-, No-, pp.-4. [8] Attanassov.K.. 986, Intutonstc uzzy sets, Fuzzy Sets and Systems, Vol-, pp.87-96 ISSN: -7 http://www.jmttjournal.or Pae 4