A Systematic Review on Fuzzy Linear & Non-Linear Model and Its Problems in Real World

Regression examination is capable strategy for modeling and anticipating in conventional regression investigation. In this examination, information ought to be fresh and ought to take after Gaussian presumption. In the event that informational index is too little, vulnerability and unclearness happens in this circumstance fuzzy regression model is proper and gives better outcomes. On the off chance that the informational index is fresh and fuzzy then we need to fit the Fuzzy regression model. In this model, the creators were concentrates on fuzzy regression model utilized as a part of linear and nonlinear programming issue. The fuzzy regression model is connected to value component of pre-assembled houses, assessing utilitarian connection amongst reliant and autonomous factors in fuzzy condition, connected to different issues, for example, and gauging building, idea of expanding basic leadership problems, a fuzzy way to deal with multi-objective linear programming problems, fuzzy linear programming issue with fuzzy coefficients. It proposed a detailing of fuzzy linear programming issue with fuzzy limitations and given strategies for its answer on imbalance connection between fuzzy numbers.

Fuzzy regression examination utilized as a part of linear programming and nonlinear programming issue to limit spread of triangular fuzzy numbers and gives preferred expectation over the conventional regression investigation. Review of Fuzzy regression model created by various authors is given as takes after. Ping-Teng Chang et. al.[1996][1] proposed fuzzy regression model in view of weighted slightest square methods. They had created technique in three phases viz, the integrity of attack of the model to the information was tried by utilizing the formal factual methodology, for example, the forward, in reverse, or Step astute systems. Besides, the fuzzy regression investigation was stretched out to fuzzy-random regression examination where the randomness of the information was considered. They watched that weighted slightest square strategy has decreased the affectability to anomalies colossally. K. K. Yen et. al. [1999][2]created fuzzy linear model utilized symmetric triangular fuzzy numbers as coefficients with the end goal that the inflexibility of the prior fuzzy linear models. They broadened the fuzzy linear model as utilized for non-symmetric the two sides of the non-symmetric triangle. They built up the connections among skew factor, spread and mistake for the fuzzy linear model. Hsiao-Fan Wang et al.[2000][3]proposed fuzzy linear regression model with fresh info and fuzzy yield portrayed by Tanaka was viewed as .They altered fuzzy minimum square strategy for the arrangement. They demonstrated that altered slightest square methods have preferable execution over both Tanaka's methods and the conventional fuzzy minimum square technique. Jing-Rung Yu et. al[2001] [4],proposed fuzzy piecewise regression investigation with programmed change-point discovery. They introduced plausibility and need regression models were used and furthermore the capacity acted diversely in various parts of the scope of crisp input factors. They watched that fuzzy piecewise regression model with distinguishing change-focuses was inferred. Miin-Shen Yang et. al. [2002][5]proposed fuzzy regression model was utilized as a part of assessing the useful connection between the needy and autonomous factors in a fuzzy situation. Most fuzzy regression models were thought to be fuzzy outputs and parameters catch fuzzy (crisp) inputs. They detailed multi target programming methods for the model estimation alongside a linear-programming issue. They created two estimation methods alongside a fuzzy slightest squares and interval remove fuzzy minimum square .They recommended that interval separate fuzzy slightest square was proficient alternative strategy to gauge parameters for fuzzy linear regression models with fuzzy outputs, fuzzy parameters and furthermore fuzzy inputs. Liem Tran et. al. [2002] [6], built up A multi-objective fuzzy regression model (MOFR).They joined focal tendency and plausibility properties of factual and fuzzy regressions additionally defeats a few inadequacies of these two methods. They introduced separate measure for two intervals that consider every one of the focuses in lower and upper limits intervals was introduced. They reasoned that both the focal tendency and possibility properties of measurable and fuzzy regressions had equipped for showing the great qualities. Additionally they successful for measured of the separation amongst lower and upper limits intervals. Y. Chen et. al. [2004] [7], proposed fuzzy linear regression with symmetric triangular fuzzy coefficients was considered to model the social capacities between designing qualities and client prerequisite. They stretched out symmetric triangular fuzzy coefficient to non-symmetric triangular fuzzy coefficients. They introduced configuration spending plan and client desires among the contenders was machine. They looked at symmetric and non-symmetric triangular fuzzy number with the assistance of reproduction examination at long last they presumed that the non-symmetric fuzzy number was preferable execution over the symmetric fuzzy number. M. Modarres et. al. [2005][8] created fuzzy linear regression model was connected for scientific programming. They expected that the needy variables were crisp while the autonomous variable was a symmetric fuzzy number. They presumed that fuzzy linear regression mode was connected scientific programming would do well to execution than Tanaka's strategy, Diamond slightest square technique and Kim-Bishu methods. A. Azadeh, et. al. [2011][9], proposed fuzzy regression–analysis of difference (ANOVA) calculation to assessed and anticipated power utilization in questionable condition. They thought about the fuzzy regression–analysis of change calculation with sixteen fuzzy regression models. They watched that there was no obvious as to which of the current fuzzy regression model was appropriate for given arrangement of real information as for power utilization .and furthermore hard to model indeterminate conduct of power utilization with conventional time arrangement and legitimate fuzzy regression could be a perfect substitute for such cases. They recommended that the fuzzy regression–analysis of fluctuation calculation used to recognized the ideal model with most minimal blunder. M. R. Taghizadeh, et. al. [2011] [10], investigated fuzzy linear regression model was connected for transport vitality demand determining utilized financial and transport related pointers. They analyzed fuzzy linear regression model depended on Gross Domestic Product (GDP), populace and the quantity of vehicles as three inputs in the model. They had taken Energy information from 1993 to 2005 were utilized to gauge each of the three inputs. They had outlined fuzzy linear regression (FLR) models, a multi-level fuzzy linear regression model and furthermore anticipated vitality demand. Lazim Abdullah et. al. [2012] [11], developed the fuzzy regression model utilized for street accident in Malaysia over the time of 1974 to 2007 utilized three predictors. They introduced the edge level h=0.5, 0.9 were accounted and furthermore model organized was the street accident utilizes enrolled vehicles, populace and street length as variables in view of one reaction variable was street accident. They demonstrated that the variables of enrolled vehicles and populace give higher effect to the quantity of street accident.

Fuzzy regression model utilized as minimum square methods, input fuzzy output, fuzzy input fresh output, fresh input fresh output and fuzzy input fuzzy output and so forth fuzzy slightest square was developed by various authors review summarized underneath. Ruoning Xu, et. al. [2001] [12], developed fuzzy multivariable linear regression by slightest squares and furthermore the methods was like the customary minimum squares technique. They had presented a fuzzy simple by a separation characterized on a fuzzy number space, and fuzzy multivariable slightest square linear regression model. They demonstrated that the model had a one of a kind solution and the solution was given in a systematic articulation. Yun-His, et. al. [2001] [13], Developed half and half fuzzy minimum squares regression. They were utilized weighted fuzzy-arithmetic mean and minimum squares fitting criterion. They had contrasted half breed regression and the common regression and other fuzzy regression methods. They recommended that mixture fuzzy regression model fulfilled a restricting conduct that the fluffiness diminishes; the equations were like the consequences of the normal regression. Volker Krätschmer [2006] [14] grew new fuzzy linear regression models. He had summed up the kind of single conventional equation in linear regression models by fused the physical vagueness of the included things as fuzzy information for the factors. He had recommended that standard slightest squares strategy was more prominent adaptability for modeling and estimation. Seung Hoe Choi, et. al. [2008] [15], proposed minimum supreme deviation estimators to build fuzzy regression models. They had analyzed a fuzzy regression model and utilized the slightest total deviation estimators and another fuzzy regression model utilized the minimum square technique. They were demonstrated that the fuzzy slightest outright deviation model was preferred execution over the fuzzy regression in light of minimum squares technique in the event that the information contains fuzzy exceptions. FurkanBaseretet et. al. [2010] [16], investigated half breed fuzzy minimum squares regression examination was connected to anticipate future claim costs by utilized the idea of London Chain Ladder (LCL) strategy. They had recommended that the half and half fuzzy minimum squares regression model was taken both randomness and fluffiness sort of vulnerability into a regression model. Fatemeh Torfi, et. al. [2011] [17], investigated a fuzzy minimum squares linear regression (FLLRs) to steering issue (VRP).They had created surmised remove fuzzy slightest squares (ADFL) estimator and furthermore connected to give information. They recommended that rough separation fuzzy minimum squares strategy was practical and productive estimator to confront the stochastic request in vehicle steering issue. B. Pushpa, et. al. [2013] [18] grew slightest supreme different fuzzy regression model. They had presented useful reliance of fresh inputs-fuzzy outputs factors utilized the summed up Hausdorff-metric between fuzzy numbers and also linear programming issue. They recommended that slightest outright fuzzy regression model utilized shape protecting operation was more compelling than Hung and Yang technique.

Fuzzy regression model utilized as left right fuzzy numbers. Utilized as fuzzy and crisp information created model preferred accuracy over conventional regression model. The distinctive looks into review as abridge underneath. Ralf Korner, et. al. [1998] [19], discussed Zadeh's expansion rule to established crisp evaluated to the slightest squares technique. They were created linear estimation hypothesis. Sadly, since fuzzy sets neglect to constitute a linear space w.r.t, addition and scalar multiplication and furthermore adjusted linear gauges regularly prompt preferable gauges over by utilization of the expansion rule. At last they utilized slightest squares estimation standard for fuzzy information was explored preferred execution over the expansion primary. H. hassanpour, et. al.[2009] [20],proposed fuzzy linear regression model and furthermore changed Kim and Bish's fuzzy linear regression examination model. They watched that Kim and Bish's technique regularly does not discover satisfactory fuzzy linear regression coefficients. At last they recommended that altered fuzzy linear regression model was preferred execution over the Kim and Bish's strategy.

Fuzzy regression model utilized for fuzzy information and crisp information to evaluated parameters and Minimizing lingering by the diverse authors in the distinctive field proposed methods were preferred accuracy over conventional regression investigation. Review summarized underneath. an element of a bio climatic list speaking to the water stretch. They demonstrated that fuzzy linear regression was given more exact expected development was gotten for high water worry than common minimum squares and orthogonal slightest squares methods. Chiang Kao et. al.[2002],proposed two-organize fuzzy linear regression model. In first stage, the fuzzy observations was de-fuzzified with the goal that the customary slightest squares technique was connected to locate a crisp regression line demonstrated the general pattern of the information. In the second stage, the blunder term of the fuzzy regression model was speaks to the fluffiness of the information and furthermore resolved to given the regression model was best logical power for the information. At long last they proposed that two-organize strategy would be wise to execution than the Kim–Bishu's and Diamond fuzzy linear regression model. Kyung. Kim, et. al. [2005][22] proposed fuzzy slightest outright deviation technique to build fuzzy linear regression model with fuzzy input and fuzzy output. They had proposed that the fuzzy minimum total deviation technique was more compelling than the slightest square strategy utilized the in the fuzzy regression examination. Rajan Alex [2006] [23] introduced fuzzy regression and fuzzy deduction. He was connected the two sorts of data assets, quantitative and qualitative data and furthermore utilized all the while in reasonable expectation. At long last he proposed that fuzzy regression and fuzzy induction would be advised to execution than immaculate regression or unadulterated deduction model. Hye-Young Jung, et. al. [2015] [24], Proposed a rank change strategy. They researched a strategy to acquire an anticipated output as for particular target esteem. They proposed that the rank change strategy in fuzzy regression model was preferable performs over the Chen and Hsuehand Diamond fuzzy regression models.

This review has attempted on fuzzy regression model utilizing linear and non linear programming issue, fuzzy regression examination utilizing minimum square strategy, fuzzy regression investigation utilized left and right fuzzy numbers and fuzzy regression model utilized as a part of interval estimation. A few authors were recommended that fuzzy regression model in various fields for forecast object was preferable execution over conventional regression examination. The analysts were created different calculations for foreseeing fluffiness and ambiguous wonder with linear programming, fuzzy

1. Ping-Teng Chang, E., Stanley, Lee (1996). “A Generalized Fuzzy Weighted Least-Squares Regression,” Fuzzy Sets And Systems, pp. 289-298. 2. Yen, K.K., Ghoshray, S., Roig, G. (1999). “A Linear Regression Model Using Triangular Fuzzy Number Coefficients”, Fuzzy Sets and Systems, Vol. 106, pp. 167–177. 3. Hsiao-Fan Wang, Ruey-Chyn Tsaur (2000). “Resolution of Fuzzy Regression Model”, European Journal of Operational Research, vol. 126, pp. 637- 650. 4. Jing-Rung Yu, Gwo-Hshiung, Tzeng, Han-Lin Li (2001). “General Fuzzy Piecewise Regression Analysis With Automatic Change-Point Detection”, Fuzzy Sets And Systems,vol. 119, pp. 247-257. 5. Miin-ShenYang, Tzu-Shun Lin (2002). “Fuzzy Least-Squares Linear Regression Analysis for Fuzzy Input–Output Data”, Fuzzy Sets and Systems.vol. 126, pp. 389–399. 6. Liem Tran, Lucien Duckstein (2002). “Multiobjective fuzzy regression with central tendency and possibilistic properties”, Fuzzy Sets and Systems, vol. 130, pp.21–31. 7. Chen, Y., Tang, J., Fung,R. Y. K., Ren, Z. (2004). “Fuzzy Regression-Based Mathematical Programming Model For Quality Function Deployment”, International Journal Of Production Research, vol. 42, No.5, pp. 1009-1027. 8. Modarres M, Nasrabadi E, Nasrabadi, M. M. (2005). “Fuzzy Linear Regression Models With Least Square Errors”, Applied Mathematics And Computation, Vol. 163, No.2, pp. 977-989. 9. Azadeh, O., Seraj, M. (2011). “An Integrated Fuzzy Regression–Analysis of Variance Algorithm For Improvement Of Electricity Consumption Estimation In Uncertain Environments”, Int J Adv Manuf Technol, Vol. 53, pp. 645–660. 10. Taghizadeh, M. R., et. al. (2011). “Design of A Multi-Level Fuzzy Linear Regression Model for Forecasting: A Case Study of Iran,” Scientific Research and Essays, Vol. 6 No. 23, pp. 5072-5081.

“Road Accident Models with Two Threshold Levels of Fuzzy Linear Regression,” Journal of Emerging Trends in Computing and Information Sciences vol.3,No.2, pp.225-230. 12. Ruoning Xu, Chulin Li (2001). “Multidimensional Least-Squares Fitting With A Fuzzy Model”, Fuzzy Sets And Systems, Vol. 119, pp. 215-223. 13. Kyung-Bin Song, Young-SikBaek, Dug Hun Hong, Gilsoo Jang (2005). “Short-Term Load Forecasting For the Holidays Using Fuzzy Linear Regression Method”, IEEE Transactions on Power Systems. Vol. 20, No. 1, pp. 96-101. 14. Volker Krätschmer (2006). “Least-Squares Estimation In Linear Regression Models With Vague Concepts”, Fuzzy Sets And Systems, vol. 157, pp. 2579 – 2592. 15. Choi, S. H., Buckley, J. J. (2008). “Fuzzy Regression Using Least Absolute Deviation Estimators”, Soft Computing, vol.12,No.3, pp.257-263. 16. Furkan Baser, Aysen Apaydin (2010). “Calculating Insurance Claim Reserves With Hybrid Fuzzy Least Squares Regression Analysis,” Gazi University Journal Of Science, Vol. 23, No.2, pp.163-170. 17. Fatemeh Torfi1, Reza Zanjirani Farahani, Iraj Mahdavi (2011). “Fuzzy Least-Squares Linear Regression Approach To Ascertain Stochastic Demand in The Vehicle Routing Problem”,Applied Mathematics, Vol.2 , pp. 64-73. 18. Pushpa B., Vasuki, R. (2013). “A Least Absolute Approach to Multiple fuzzy Regression Using TW-Norm Based operations”, International Journal of fuzzy Logic systems, Vol. 3, pp.73-84. 19. Ralf Korner, Wolf Gang Nfither (1998). “Linear Regression With Random Fuzzy Variables: Extended Classical Estimates, Best Linear Estimates, Least Squares Estimates”, Journal of Information Sciences. Vol.109, pp. 95-118. 20. Hassanpour, H., Maleki, H. R., Yaghoobi, M. A. (2009). “A Note On Evaluation Of Fuzzy Linea Regression Models By Comparing Membership Functions,” Iranian Journal Of Fuzzy Systems, Vol. 6, No. 2, pp.1-6. 21. Boreux, J.J., Gadbin-Henry, C., Guiot, J., Tessier, L. (1998). “Radial Tree-Growth For. Res, Vol. 28,pp.1249–1260. 22. Kyung, Kim, et. al. [2005] proposed fuzzy least absolute deviation method to construct fuzzy linear regression model with fuzzy input and fuzzy output. They had suggested that the fuzzy least absolute deviation method was more effective than the least square method used the in the fuzzy regression analysis 23. Rajan Alex., “A new kind of fuzzy regression modeling and its combination with fuzzy inference”, Soft Comput, pp. 618-622. 24. Hye-Youngjunget Jin Hee Yoon, Seung Hoe Choi (2015). “Fuzzy Linear Regression Using Rank Transform Method”, Fuzzy Sets and Systems, vol.274, pp.97–108.

Department of Mathematics, OPJS University, Churu, Rajasthan, India