A Study of Fuzzy Parameter with Queue Model

INTRODUCTION

Queueing models have been turned out to be extremely helpful in numerous reasonable applications in territories, for example, e.g., generation systems, stock systems and correspondence systems. The inconsistency of the arrival and service processes in many applications is fundamental to the performance of the system. Queueing models help us to understand the impact of inconsistencies and to assess them. Kendall had a brief documentation depicting a range of these queueing models. It's a third part a/b/c code. The main letter defines the time distribution of the interarrival and the second the distribution of the time of service. For example, the letter G, M in the exponential distribution (M stands for the Memoryless) and D in deterministic times is used for a general distribution. For example. The number of servers is defined in the third and last letters. M/M/1, M/M/c, M/G/1 and M/D/1 are just a few models. With additional letter to cover other queueing templates the documentation can be reached. For example, the four letter codes M/M/1/N abridge a system with exponential intercom and service times, a single server, and lounge for N customers only (counting the one in service). Models of queuing are also called line models holding up. Here we study what happens when a person or many people go to queues. In our daily life, we are very familiar with queues. Normal cases of queuing models we experience go to a specialist or go to a hairpiece. In essence, a queuing system occurs when elements or people call arrivals who need some kind of service from another element. A service exists and the person connects to that system on a line or a queue. A stochastic process is a collection of time-listed random variables. One alternative view is that it is a distribution of probabilities across a number of ways: it regularly portrays a random appreciation or system after a while. There is a settled direction (route) in a deterministic process that the process goes but in a stochastic process we don't know how. This should not be regarded as information of the way, as the probability distribution gives the information on the way. This can be likened to a determinist process, for example, given that the probability distribution is given as a probability one. Furthermore, the process usually develops after a while. Nevertheless, from a formal numerical point of view, a higher priority image is that we make them hidden (obscure) and look just at the bottom

takes true values. We start by studying discrete processes of stochastic time. These processes can be explicitly communicated and are gradually 'substantial' or 'simple to understand.' A collection system of customers/things in some place, including people getting the service, is known as queue or line-ups. Queuing theory is the principal part of time patterns, where stochastic exchange is involved. The basic motivation behind the theories is to explore rapidly those queuing issues in companies, transport and business that concern times due to people and machines, as well as to complete the creation of a thing by using tandem, bi-tandem queues, at least two distinct progressive phases of the stochastic process. For researchers who need to carry out basic research on stochastic processes, including numerical models, the trend towards analytical research has been developed and queuing theory has been shown. The theory of probability gives the establishment theory and stochastic model theories of tele-traffic, so it is important for the experts in the understudies to develop the probability ideas for the material they use. We aim to include perusers with some probability foundation adequately in this section. Although the cover is complete here, because it speaks of all the likely ideas and methods used in later parts, it excludes the numerous precedents and activities regularly incorporated into a probability course book, which allows readers to get a better handle on the material. The Queuing models can be used for a wide range of applications, including machine fixes, toll booths, taxi stands, boat stationing and vacuuming, patient bookings in healing centers, PC planning, timing sharing and systems structures, media transmission, and service associations. The traditional theory of queuing expects certain preset probability distributions to be followed by interval time and service times. Ethymological articulations like fast, moderate, moderate, suitable or missing, as opposed to probability distributions, can be all the more practical in numerous practical conditions for arrival and services examples and times. All the qualities of queuing along these lines can be better determined by fuzzy numbers, such as the arrival rate, service and the excursion rate. This gives the extent to which the queues of flouted theory are studied. In areas such as assembly system, media transmission and data processing, fluid queues can be adequately connected.

Fuzzy logic is best defined as a mathematical logic in which true values between 0 and 11 can be assumed. It is known as bivalent logic that each statement must either be true or false. The central concept of fluid logic is that each proposition can be a system that mathematically expresses partial truth. Fuzzy's logic is understood by examples, as are many theories. "Georgia Tech is a great school," take the statement. This statement can either be true or false by means of the bivalent logic. Fuzzy Logic says that this is 100% true if student registration is 10,000 or higher, but only 50% true if registration is three,000 and 0% true if the registration is fifty per cent. The first introduction of the Fuzzy Logic theory in a 1965 paper titled "Fuzzy Sets" was by Lotfi Zadeh, a professor at Berkeley University California. There can be four major contributions to the philosophical basis of the Fuzzy logic. In 200 B.C., Aristotle suggested the "Law of the Excluded Middle," which held that all statements must be false or true. Plato was one of the first to suggest the dichotomy did not describe reality fully. He theored that between true and false there was a state. Logicist Jan Lukaisiewicz suggested mathematics in 1920 for a tri-valued logic that incorporated the concept of fractional reality. This third logic value was referred to as "possible," beyond true and false. The ideas of Lukaisiewicz formed the basis of a wide range of research on what became known as highly valued logic.

Many fuzzy linguistic designs, like 'low,' 'medium,' 'high,' etc., are used to define a variable status. The relevancy of a fluid variable is that it facilitates gradual transitions between states and therefore has a neutral ability to express and manage incertitudes. Computing involves the handling of numbers and symbols in the traditional sense. But humans, by contrast, use mainly words in computing, reasoning, and natural language, or mental perceptions. One key aspect of word computation is that natural languages are merged and calculated with fluctuating variables. In computing words, the concept of a granule plays an essential part. "Study plays a key role in human cognition," Zadeh says. This is a way to achieve comparison of data for humans. Fuzzy sets, defined in the set R of real numbers, have a particular significance. Membership functions  : R  [0,1] clearly possess a quantitative meaning and may be viewed as fuzzy numbers provided they satisfy certain conditions. Initiative concepts of approximate numbers or intervals such as near 5' numbers or numbers around the actual numbers. Such concepts are essential to characterize state of fluctuating variables. In many applications, these fugitive numbers play an important role, including fugitive control, decision-making, approximate reasoning and optimization. A fuszy number is the fuszy subset of the real line, where a given real number is the highest membership values. The membership function on both sides of the key values is monotonous for a fuzzy number. The

A fuzzy subset A of the real line R with membership function A : R [0, 1] is called a fuzzy number if

A triangular fuzzy number A ~ is a fuzzy number specified by 3-tuples (a1, a2, a3) such that a1  a2  a3, with membership function defined as This is represented diagrammatically as

A trapezoidal fuzzy number A ~ is a fuzzy number fully specified by 4-tuples (a1, a2, a3, a4) such that a1  a2  a3  a4, with membership function defined as

A piecewise – quadratic fuzzy number (PQFN) is defined by the membership function as The PQFN is a bell shaped function and symmetric about the line x = a3, has a supporting interval a [a ,a ] ~  1 5 . Moreover, and . -cut at level between the points (a2, a4) called cross over points. Also the interval of confidence at level  is . This is diagrammatically

treat the fuzzy diametrical operations with triangular, trapezoidal and piecewise quadratic fuzzy number.

Queuing models are broader for both the services organization and the manufacturing companies, where different clients are serviced according to the specific queue discipline through different types of servers. The intervals and the service times of certain distributions must be followed within the context of traditional queuing theory. In practice, service rate is often described as fast, slow or moderate by linguistic terms, which can best be described by the fluency sets. Li and Lee had analysis findings based upon a Zadeh extension principle for two fuzzy queuing systems. As Negi and Lee commented, their approach is however rather complicated and not usually computational. Therefore, Negi and Lee propose two approaches: the rest of the formulations and two variable. Sadly, their approach only provides possible numbers rather than intervals; in other words, the performance measurements' membership functions are not fully described. This model follows the cut-off approach of breaking a fluffy queue into a family of crisp queues. The parametric programming technique for describing the crisp queues family is used when T-varies. The solution derives the membership of the crisp queues from the parametric programs. In order for a fluid queue FM/FM/1 to show the validity of the proposed approach, FM indicates fluid exponential time.

Consider a general queuing system with one server. The inter arrival time and service time are approximately known and are represented by the following fuzzy sets. Where x and y are the crisp universal sets of the inter arrival time and service time, and and are the corresponding membership functions. Where A() and S() are crisp sets. Using -cuts, the inter arrival time and service time can be represented by different levels of confidence intervals. Consequently, a fuzzy queue can be between ordinary sets and fuzzy sets. Let the confidence intervals of the fuzzy sets are respectively. When both inter arrival time and service time are fuzzy numbers, based on Zadeh‘s extension principle, the membership function of the performance measure P(x, y) is defined as Our approach is to construct the membership function which is based on deriving the -cuts of The corresponding, Parametric programming technique for finding lower and upper bounds of the -cut of are If both are invertible with respect to , then a left shape function and a right shape function can be obtained from which the membership function is constructed.

This queue adopts a first-come first-served queue discipline and consider an infinite source population where both the inter arrival time and the service time follow exponential distributions with ratio respectively, which are fuzzy variables rather than crisp values.

Consider a FM/FM/1 queue, where both the arrival rate and service rate are fuzzy numbers represented by = [3,4,5,6] and = [19,20,21,22]. The interval of confidence at possibility level  as [3 + , 6 – ] and [19 + , 22 – ].

When x reaches its lower bound and y reaches its upper bound, attains its minimum. Consequently, the optional solution for (1) is . On the contrary, to maximize increases to its upper bound and y decreases to its lower bound. In this case the optimal solution for (2) is

CONCLUSION

Queueing models can be isolated into two general gatherings. On one hand, those that depict genuine circumstances and, for other, those regularizing that report a remedy of what the genuine circumstance ought to be or, said in another way, the ideal perspective to which ought yearn for. Graphic models give normal values and the probabilities of execution estimates that portray the system when examples of arrivals and services, the number of servers, system limit and queue discipline have all been set. As opposed to these models, the second gathering, which is regularly called queuing choice models (plan and control models), endeavors to compute what the parameters ought to be to improve the models. When there is vulnerability with regard to data incorporation, models considered in this work must be improved. The use of fuzzy subset theory is used to resolve vulnerability. The count of questionable parameters and furious data in the queueing models involves that the capacity to be improved contains furious coefficients. Service actions, for instance the service rate, the number of servers and the queue discipline or a mixture of components, are generally controllable. Some checks can sometimes be applied when customers

constraints on physical space and require specific Parameters which would be wild from the beginning. The study of classical tape models with fluid data that this study tries to use on both previously described routes. In this line, the paper manages different models that show the real circumstances of the queue when one or more parameters are uncertainly known and illuminates plan and control models to streamline fluid queueing systems. Basic queueing systems include composed queues where units are managed by their arrival request. This "holding up control" is often found in queueing models, yet for reasons of productivity and picking order, favorite classes of units with a specific need are characterized by the status of message transmission within a media communication system. Many genuine queuing systems are more careful than any other models that are possible to access this disciplinary requirement model.

REFERENCES

[1] C. Kao, Chang-Chung Li, Shih-Pin Chen (1999), Parametric Programming to the Analysis of Fuzzy Queues, Fuzzy Sets and Systems, 107: pp. 93- 100. [2] S.H. Chen (1985), Operations on Fuzzy Numbers with the Function Principle, Tamkang Journal of Management Sciences, 6(1): pp. 13-25. [3] S.J. Chen, S.M. Chen (2003), A New Method for Handling Multicriteria Fuzzy Decision Making Problems using FN-IOWA operators. Cybernatics and Systems, 34: pp. 109-137. [4] S.J. Chen and S.M. Chen (2007), Fuzzy Risk Analysis Based on the Ranking of Generalized Trapezoidal Fuzzy Numbers. Applied Intelligence, 26(1): pp. 1-11. [5] C.H Cheng (1998), A New Approach for Ranking Fuzzy Numbers by Distance Method. Fuzzy Sets and Systems, 95: pp. 307-317. [6] S.P. Chen (2005), Parametric Non – linear Programming Approach to Fuzzy Queues with Bulk Service, European Journal of Operational Research, 163: pp. 434–444. [7] S.J. Chen, L. Chen and C. Hwang (1992), Fuzzy Multiple Attribute Decision Making Methods and Applications. In Lecture Notes in Economics and Mathematics Systems. Springer, New York. Operational Research, 157: pp. 429-438. [9] G. Choudhury (1998), On a Batch Arrival Poisson Queue with a Random Setup Time and Vacation Period. Computers and Operations Research, 25, 12: pp. 1013-1026. [10] G. Choudhury (2000), An M[X]/G/1 Queuing System with a Setup Period and a Vacation Period, Queuing Systems, 36: pp. 23-28. [11] G. Choudhury (2002), A Batch Arrival Queue with a Vacation Time under Single Vacation Policy. Computers and Operations Research, 29: pp. 1941-1955.

PhD Student, Kalinga University, Raipur