A Survey On Fuzzy Mcdm Methods

Published Date: 02 Nov 2017

1Martin.A*, 1Prasanna Venkatesan.V, 2Girija.J, 2Jayashree Devi.E, 2Senbagavalli.S

1Dept. of Banking Technology, Pondicherry University,Pondicherry, India

2Department of IT, Sri ManakulaVinayagar Engineering College,Pondicherry, India

*[email protected],[email protected],[email protected]

Abstract

Fuzzy Multiple Criteria Decision Making (FMCDM) provides strong decision making in domains where selection of best choice is highly complex. This survey paper reviews the main streams of consideration in Fuzzy Multiple Criteria Decision Making theory and practice in detail. The main purpose is to identify various applications and the approaches, and to suggest approaches which are most robustly and effectively useable to identify best choice. This survey work also address the problem areas in FMCDM still requiring further research is also discussed. FuzzyMulti criteria decision making is used in fields like Banking, Performance evolution, Safety assessment and many other complex issues. This FMCDM method helps to choose the best alternatives where many criteria have come into existence. The different techniques used in FMCDM are TOPSIS, FAHP, Hybrid DEMATEL and ELECTRE. This survey finds various research opportunities in FMCDM and its further development.

Keywords: Multi criteria decision making, Fuzzy MCDM, TOPSIS, Best choice, decision making

Introduction

In our day today life, so many decisions are being made from various criteria’s, so the decision can be made by providing weights to different criteria’s , where all the weights are obtain from expert groups. It is important to determine the structure of the problem and explicitly evaluate multiple criteria. For example, in building a nuclear power plant, certain decisions are made different criteria. There are not only very complex issues involving multiple criteria, some criteria may have effect toward some problem, but over all to have an optimum solution, all the alternatives must have common criteria which clearly lead to more informed and better decisions.

MCDM is pertaining to structuring and solving decision and planning problems involving multiple criteria. The desire of this survey is to support decision makers where there are many choices for a problem to be solved. Typically, there is no unique optimal solution for these problems and it is necessary to use decision maker’s desire to differentiate between solutions [1].Solving can be interpreted in different ways. It could correspond to choosing the "best" alternative from a set of available alternatives (where "best" can be interpreted as "the most preferred alternative" of a decision maker). Another interpretation of "solving" could be choosing a small set of good alternatives, or grouping alternatives into different preference sets. An extreme interpretation could be to find all "efficient" or "non-dominated" alternatives.

The difficulty of the problem occurs when many criteria exists for the alternatives. There is no longer a unique optimal solution to an MCDM problem that can be obtained without incorporating the desired information. The idea of an optimal solution is often put back by the set of non-dominated solutions. A non-dominated solution has the property that it is not possible to move away from it to any other solution without sacrificing in at least one criterion. Therefore, it makes sense for the decision maker to choose a solution from the non-dominated set. Otherwise, he could do better in terms of some or all of the criteria, and not do worse in any of them. Generally, however, the set of non-dominated solutions is too large to be presented to the decision maker for his final choice.

This survey on Multi Criteria Decision attained the need for MCDM, many works are done on MCDM in determining the best optimal solution for a problem using different methods in it, and each method in it has its own unique purpose. Many applications uses MCDM in determining the flaws in the system, these flaws can be managed by using appropriate method for solving the problem.

The rest of the paper, in section 2 we discuss about the prior research work, the methods and the applications of MCDM and section 3, the research opportunities and section 4 concludes the paper.

2. Prior Research

Decision making in effect has shown that fuzzy logic allows decision making with estimated values in spite of incomplete information. It should be noted, however, that a decision may not be correct and can be improved later when additional information is available. Of course, a complete lack of information will not support any decision making using any form of logic. For difficult problems, conventional (non-fuzzy) methods are usually expensive and depend on mathematical approximations (e.g. linearization of nonlinear problems), which may lead to poor performance. Under such circumstances, fuzzy systems often outperform conventional MCDM methods. Many works have been done in various fields like banking, general purpose, student and teacher performances, water resource location and many. Here the alternatives and criteria have been collected and the evaluation of the criteria has been done to choose the best alternatives. Structuring complex problems well and considering multiple criteria explicitly lead to more informed and better decisions.

2.1Methods of MCDM

MCDM methods are applied to different applications and find the best solution to choose the best alternative.

The fig 1 gives the clear view of all MCDM methods and its types. The widely used MCDM methods have been described in following headings,

Figure 1: Hierarchical structure of MCDM Methods

2.1.1 Analytic Hierarchy Process

The basic idea of AHP is to capture experts’ knowledge of phenomena under study. These methods are systematic approaches to the alternative selection and justification problem by using the concepts of fuzzy set theory and hierarchical structure analysis. Decision-makers usually find that it is more confident to give interval judgments than fixed value judgments. When a user preference is not defined explicitly due to fuzzy nature this method can be applied. AHP includes the opinions of experts and makes a multiple criteria evaluation; it is not capableof reflecting human’s vague thoughts. The classical AHP takes into consideration the definite judgments of decision makers, thus the fuzzy set theory makes the comparison process more flexible and capable to explain experts’ preferences. The Analytic Hierarchy Process (AHP) decomposes a complex MCDM problem into a system of hierarchies [2]. The final step in the AHP deals with the structure of an m*n matrix (Where m is the number of alternatives and n is the number of criteria’s have been considered). The matrix is constructed by using the relative importance of the alternatives in terms of each criterion. Analytic Hierarchy Process (AHP) is an MCMD method based on priority theory. It deals with complex problems which involve the consideration of multiple criteria/alternatives simultaneously.

2.1.2 Fuzzy Analytic Hierarchy Process

Fuzzy AHP is Fuzzification of the AHP (analytic hierarchy process) used in conventionalmarket surveys, etc. In AHP, several products and alternatives are evaluated, and bymeans of pair comparisons, the weight of each evaluation item and the evaluation valuesfor each product and alternatives are found for each evaluation item, but the results ofpair comparisons are not 0,1, but rather the degree is given by a numerical value[4]. In fuzzyAHP, the weight is expressed by possibility measure or necessary measure, and inaddition, the conventional condition that the total of various weights be 1 is relaxed.

2.1.3 TOPSIS

The TOPSIS method assumes that each criterion has a tendency of monotonicallyincreasing or decreasing utility. Therefore, it is easy to define the ideal and negative-idealsolutions. The Euclidean distance approach was proposed to evaluate the relativecloseness of the alternatives to the ideal solution. Thus, the preference order of thealternatives can be derived by a series of comparisons of these relative distances.The TOPSIS method first converts the various criteria dimensions into non-dimensional criteria as was the case with the ELECTRE method [15].The concept of TOPSIS is that thechosen alternative should have the shortest distance from the positive ideal solution (PIS) and the farthest from the negative ideal solution (NIS). This method is used for ranking purpose and so to get the best performance bank we go for TOPSIS method where we can be able to choose the best one. FUZZY TOPSIS method is used to evaluate the criteria in each region and then all the criteria have been ranked based on the region.

2.1.4 ELECTRE

ELECTRE (Elimination EtChoix Traduisant la REalite´) is one of the MCDM methods. This method allows decision makers to select the best choice with utmost advantage and least conflict in the function of various criteria. The ELECTRE method is used for choosing the best action from a given set of actions and was later referred to as ELECTRE I. Different versions of ELECTRE have been developed including ELECTRE I, II, III, IV and TRI. All methods are based on the same fundamental concepts but differ both operationally and according to the type of the decision problem. Specifically, ELECTRE I is intended for selection problems, ELECTRE TRI for assignment problems and ELECTRE II, III and IV for ranking problems.The main idea is the proper utilization of "outranking relations". ELECTRE creates the possibility to model a decision process by using coordination indices. These indices are concordance and discordance matrices. The decision maker uses concordance and discordance indices to analyze outranking relations among different alternatives and to choose the best alternative using the crisp data.

2.1.5 Grey Theory

Grey Theory, extremely high mathematical analysis of the systems that are partly known and partly unknown and defined as ‘‘weak knowledge’’ and ‘‘insufficient data’’. Grey Theory examines the interactional analysis when the decision-making process is not clear, there are a great number of input data and it is discrete and insufficient. In recent years, Grey Theory has been successfully applied in many decision making problems, such as supplier selection facility layout selection, financial performance evaluation, demand forecasting and material selection.

These are MCDM methods, which are generally applied to find best alternative when choices and criteria are high. There is a need for each and every method, for any selection purpose we go for ELECTRE method where the purpose gets fulfilled, for any other ranking purpose we choose TOPSIS where it is used to obtain the best among all, for any clarification purpose we go for Grey Theory where all those in complete data can be rectified. The next section discusses about applications of these Fuzzy MCDM methods.

Apart from the methods have been used here, many other methods are available in MCDM which are listed below with its description. The suitability of each method and problem in which it can be applied is discussed in following division. The merits and demerits of various MCDM methods are given in Table 1 as follows,

Table 1: MCDM methods with its merits and demerits

Sl.

No

MCDM Methods

Description

Advantages

Disadvantages

1.

Analytic hierarchy process (AHP)

It also includes pair wise comparison of different alternatives for different criterion.

1. Flexible, intuitive and checks inconsistencies

2. Since problem is constructed into a hierarchical structure, the importance of each element becomes clear.

3. No bias in decision making

1. Irregularities in ranking

2. Additive aggregation is used. So important information may be lost.

3. More number of pair wise comparisons are needed

2

Analytic Network Process(ANP)

AHP builds the decision problem from arrangement of different goals, criteria and alternatives and pair wise comparison of the criteria to obtain the best alternative

1. Independence among elements is not required.

2. Prediction is accurate because priorities are improved by feedback.

1. Time consuming

2. Uncertainty

3. Hard to convince decision making

3.

Data envelopment analysis

DAE is a method where it is used to find the efficiency of combination of multi inputs and multi outputs of the problem.

1. Multiple inputs and outputs can be handled.

2. Relation between inputs and outputs are not necessary.

3. Comparisons are directly against peers

4.Inputs and outputs can have very different units

1.Measurement error can cause significant problems

2. Absolute efficiency cannot be measured.

3. Statistical tests are not applicable.

4. Large problems can be demanding.

4.

Aggregated Indices Randomization method (AIRM)

This method solves the complex problem where uncertainty occurs which has incomplete information for the problem to be solved.

1. Non-numeric, non-exact and non-complete expert information can be used to solve multiple criteria decision making problems.

2. Transparent mathematical foundation assures exactness and reliability of results.

It aims only at complex objects multi-criteria estimation under uncertainty.

5.

Weighted Product model(WPM)

Here the alternatives are being compared with the other by the weights and ratio of one for each criterion.

1. Can remove any unit of measure.

2. Relative values are used rather than actual ones.

No solution with equal weight of DMs

6.

Weighted Sum Model (WSM)

It is used for evaluating a number of alternatives in accordance to the different criteria which are expressed in the same unit.

Strong in a single dimensional problems

Difficulty emerges on multi-dimensional problems

7.

Goal Programming

Goal programming is a division where it has more than one objective which conflicts with each other, and by arranging the goals or target have to be achieved by minimizing the irrelevant information.

1. Can handle large numbers of variables, constraints and objectives.2. Simplicity and ease of use

1. Setting of appropriate weights.

2. Solutions are not Pareto efficient.

8.

ELECTRE

It is used to select the best choice with maximum advantage and least conflict in the function of various criteria

Outranking is used

Time consuming

9.

Grey analysis 

This methods deal with all incomplete data and to overcome the deficiencies of other methods.

Perfect information has a unique solution

Does not provide optimal solution.

10.

PROMETHEE

Allows the user to directly make use of the data such as alternatives an, criteria fk, of the problem in a simple evaluative multi criteria table

Outranking is used.

Uncertainty.

2.2 Applications of FMCDM

FMCDM is used in various domains such as banking, performance elevation, decision making in different organization, safety assessment, multi choice general purpose problems,and etc. This section discuss about the various FMCDMS methods and its application domains.

2.2.1Fuzzy MCDM Applications

Fuzzy occurs in various business organizations when multiple choices available, to take a best decision for the development of the business organization fuzzy multi criteria techniques are applied. For example for supplier selection in the organization is one of the multi criteria decision making problem which includes both quantitative and qualitative factors [19]. In order to choose the best suppliers it is essential to make a trade-off between these tangible and intangible factors some of which may conflict. The process of determining the suitable suppliers who are able to provide the buyers with the right quality product or services at the right price at the right time and in the right quantities, is one of the most critical activities for establishing an effective supply chain. To solve this various methods such as TOPSIS, ELECTRE and AHP are applied.ELECTRE is used to reach close to the positive and get move off from negative points.

Fuzzy MCDM can also be applied for safety assessment where the urge of safety is prevailing everywhere. The safety measures can be obtained by having good alternatives and criteria’s. Once the criteria have been formed the evaluation of the criteria takes places in order to find the approximate solution for it. Safety issues are really at the core of marine engineering,the safety comes on, how the crew members understand the urge of risk and how the members effectively manage it to bring more reliable [21]. These are ensuring safety by mitigating risks and increasing the reliability of the system. For this we use fuzzy techniques such as TOPSIS, ELECTRE, and AHP. Fuzzy MCDM methods also applied in areas such as location planning [8], revision of OWA operator problems [24], etc., which are described in Table 2.

Table 2:FMCDM applications in other domains

Application

Author and year

Alternatives

Criteria

Problem

Techniques

Best alternative

Location planning for

urban distribution centers under uncertainty[8]

(Anjali et al.,

2011)

3 Different

Areas

A1

A2

A3

1.Accessibility

2.Security

3.Connectivity to multimodal

transport

4. Costs

5.Environmental impact

6. Proximity to customers

7. Proximity to suppliers

8. Resource availability

Location planning for urban distribution centers is vital in saving distribution costs and minimizing traffic congestion arising from goods movement in urban areas.

TOPSIS

A1 >A3 >A2.

A1 is the best area

Revising the OWA operator problems under uncertainty(case study)[24]

(Mahdi et al., 2009)

1. Sahand

2. Shahriar

3. Kalghan

4.Germichai

5. Givi

6. Taleghan

7. Talvar

8 .Galabar

9. Sanghsiah

10. Soral

11. Siazakh

12. Bijar

1.Allocation of water to prior usages

2.Number of

beneficiaries

3.Supporting other

projects

4.Benefit/cost

5.Range of environmental

impacts

6.Publicparticipation

7.Jobcreation

In finding the most robust alternative among these seven criteria

FSROWA

Fuzzy-Stochastic-Revised

Ordered Weighted Averaging (FSROWA) method is applied.

Germ chai project is the most preferred project

Enhancing information delivery in extended enterprise networks [31]

(Lau et al., 2003)

P1, P2, P3, P4, P5 (information receivers)

1. partner’s price range

2. partner’s interest to information

3. partner’s product range

To find the best supplier for mold and die manufacturing concern, the product price range, the information receiver’s interest and the product range are often considered by enterprises.

FMCDS

P2

Evaluating anti-armor weapon using ranking fuzzy numbers [12]

(Shu-Hsien and Kuo-Chung, 2000)

1. Dragon

2. Milan and

3. Sword

(weapon systems)

1. basic capability

2. fight capability

3.logisticmaintenance

4. electronic system

Fuzzy multi criteria decision support procedure is applied to non-quantitative factors where decision making is complex.

Fuzzy multiple attribute decision making

Sword

Evaluation suppliersinsupplychainmanagement[19]

(Mohammad, et al.,2010)

1.Suppiler 1

2.Suppiler 2

3.Suppiler 3

4.Suppiler 4

5.Suppiler 5

1.Urgent delivery

2.On time delivery

3.Ordering cost

4.Warranty period

5.Product price

6.Financial stability

7.Delivery lead time

8.Accessibility

9.Reliability

10.Transportation cost

11.Rejection of defective product

12.Cost of support service

13.Testability

Supplierselection,the processof determiningthe suitablesupplierswhoareabletoprovidethe buyer withtherightqualityproductsand/orservicesatthe rightprice,attherighttimeandintheright quantities.

TOPSIS

Supplier 3

A fuzzy multi-criteria decision making model for supplier selection[14]

(DoraidDalalah et al, 2011)

1.Saudi Arabian for Packaging

Industry (SAPIN), 2.Arabian Can Industry (ACI),

3, ZA Turkish Supplier

And

4. Al-Watonga for Containers Manufacturing (CMC).

1.unit price and payment terms

2.delivery terms

3.supplier factory capacity 4.shipping method

5.lead time 6.location of can supplier 7.technical specifications

8.Services and communications with the supplier

9.compensationfor waste

10,major customers with the same business 11.certificate of

Supplier

For the selection of cans supplier/Suppliers at Nitrides Factory in Amman-Jordan to demonstrate the proposed model.

1.Modified fuzzy DEMATEL model,

2.A modified TOPSIS model

SAPIN

Examine the use and application of MCDM techniques in safety assessment [21].

(Orestis et al., 2010)

3 DIFFERENT

COMPNANY

1.C1

2.C2

3.C3

1.Cost-control

2.Detailed information about the crewmembers and their behavior

3.availability of presenting data per ship

4.comparsion with industry

5. Planning, preview and scenarios of risk management.

To enhance safety by mitigating risks and increasing the reliability of a system.

1.TOPSIS

2.ELECTRE

3.AHP

C2

Multi-criteria decision making approach based on immune co-evolutionary

algorithm with application to garment matching problem[33]

(Yong-Sheng Ding et al, 2011)

65 trousers with the same color, style and material

for female are studied

waist girth (W),

hip girth (H), and

trousers length (L)

To solve the large scale garment matching problem where Size fitting problem

is a main obstacle to large scale garment sales and online sales because it is difficult to find the fit

Garments by the general size information

co-evolutionary immune algorithm

for the MCDM model

The product which satisfies the "CUSTOMER SATISFACTION and SERVICE QUALITY " the most

An incident information management framework based on data integration, data

mining, and multi-criteria decision making[35]

(Yi Peng et al,2011)

1.Beijing

2.Tianjin

3.Hebei

4.Chongqing

5.Xinjiang( 31 provinces)

1.Percentage of areas covered to total areas

2. Percentage of areas affected to total areas

(Drought

Flood

Hailstorm

Frost)

A case study on agro meteorological disasters that

Occurred in China between 1997 and 2001. The case study demonstrates that the combination of data mining

and MCDM methods can provide objective and comprehensive assessments of incident risks.

TOWA operator,

cluster analysis, grey relational analysis, and TOPSIS

Chongqing

Assessment of health-care waste treatment alternatives using fuzzy

multi-criteria decision making approaches[34]

(Mehtapetal,2011)

1.Incineration

2. Steam sterilization

3.Microwave

4. Landfill

1.Economic

2.Environmental

3.Technical

4.Social

The objective of this research is to propose multi-criteria decision making techniques for

conducting an analysis based on multi-level hierarchical structure and fuzzy logic for the evaluation of

HCW treatment alternatives.

fuzzy MCDM methodology,

hierarchical

distance-based fuzzy MCDM algorithm

Landfill

Comparative analysis of multi-criteria decision making methodologies

and implementation of a warehouse location selection problem[37]

(Tuncay al.,

2011)

1. Warehouse A

2. Warehouse B

3. Warehouse C

4. Warehouse D

1.Unit price

2.Stock

holding

capacity

3.Average

Distance to shops

4.Average distance to main

suppliers

5.Movement

Flexibility

To compare the MCDM methods and implementation of a warehouse location selection problem

AHP,

TOPSIS,

ELECTRE and

Grey Theory

WAREHOUSE D

Health- Safety and Environmental Risk Assessment of

Refineries Using of Multi Criteria Decision Making Method[39]

(Saharet al.,

2012)

Power plant

1. location 1

2. location 2

3. location 3

4. location 4

1.environment of the power plant,

2.health-safety risks, 3.technological risks,

4.the affected environment risks

To find the best location for the implementation of the power plant using the AHP

AHP

Location 3

Mathematical analysis of fuel cell strategic technologies

development solutions in the automotive industry[40]

(Keivan

et al., 2011)

1.Professional manpower on industrial &

semi-industrial scale

2.Professional manpower on laboratory scale

3.Know-how on industrial & semi-industrial scale

4.Know-how on laboratory scale

5.Hardware on industrial & semi-industrial scale

6.Hardware on laboratory scale

Power density

2. Efficiency system of fuel cells

3. Fuel type (Including the effect on fuel cells operation,

process stages, availability, cost, safety and environment

considerations)

4. Life time and preserving fuel cells

5. Operational heat, start-up period, reaction period and

response of fuel cells

6. Security and confidence

The analysis of fuel cell strategic technology in the automotive industry using TOPSIS

TOPSIS

Professional manpower on laboratory scale

[32]Unit price and payment terms (C1), delivery terms (C2), supplier factory capacity (C3), shipping method (C4), lead time (C5), location of can supplier (C6), technical specifications (C7), certifications (Regular and International) (C8), services and communications with the supplier (C9), compensation for waste (C10), printing complies to design and color (C11), easy open and spoon leveling (C12), testing methods for packaging materials and available tests from supplier (C13), variation of dimensions (C14), stretch wrapping and clean separators, pallet size and height (C15), major customers with the same business (C16), certificate of supplier materials (C17), SAPIN - Saudi Arabian Packaging Industry, ACI - Arabian Can Industry, CMC - Containers Manufacturing,

[35]Incident information management framework consists of three major components. The first component is a high-level data integration module in which heterogeneous data sources are integrated and presented in a uniform format. The second component is a data mining module that uses data mining methods to identify useful patterns and presents a process to provide differentiated services for pre-incident and post-incident information management. The third component is a multi-criteria decision-making (MCDM) module that utilizes MCDM methods to assess the current situation, find the satisfactory solutions, and take appropriate responses in a timely manner

[34]Sub criteria: Economic: Capital cost, Operating cost, Environmental: Solid residuals and environmental impacts, Water residuals and environmental impacts, Air residuals and environmental impacts, Release with health effects.Technical: Reliability, Volume reduction, Need for skilled operators, Occupational hazards occurrence impact, Treatment effectiveness, Level of automation, Occupational hazards occurrence frequency.Social: Adaptability to environmental policy, Land requirement, Public acceptance obstacles

Table 2 describes some the application of Fuzzy MCDM in various disciplines. In some applications uncertainty in decision making arises, so fuzzy multi criteria decision making is chosen to solve this issue.The criteria used in urban distribution centers such as security, accessibility, cost, and environment [8]. The sensitivity analysis is performed to determine the influence of criteria and weights on location planning is applied to find the suitable locations. The selection of location for placing the watershed which is using the new method FSROWA is introduced to combine the Fuzzy and Stochastic features into a revised OWA operator, for choosing the effective place for the location of the water shed [24].To search the best place for urban Centre distribution all the places are ranked based on criteria.

The co-evolutionary immune algorithm for the multi-criteria decision making (MCDM) model, is used for the model to solve the large scale garment matching problem. Size fitting problem is a main obstacle to large scale garment sales and online sales because it is difficult to find the fit garments by the general size information. This study regards the fit garment matching problem as a MCDM model with the constraints of size satisfaction. An immune co-evolutionary algorithm is used to search the fit garments from the candidate garments in the stock [33]. Health-care waste (HCW) management is a high priority environmental, public and health concern in developing countries. The management and treatment of HCW are gaining more attention with the rising awareness. The proposed decision approaches enable the decision-makers to use linguistic terms, and thus, reduce their cognitive burden in the evaluation process. By using MCDM, the evaluation of multi-level hierarchical structure and fuzzy logic forHCW treatment can be obtained [34].

An effective incident information management system deals with several challenges. Decision makers (DMs) have to detect variance and extract useful knowledge, make DMs in evaluating the risks and selecting an appropriate alternative during an incident, and provide differentiated services to satisfy the requirements of different incident management phases. Multi-criteria decision-making (MCDM) module that utilizes MCDM methods to assess the current situation, finds the satisfactory solutions, and takes appropriate responses in a timely manner [35]. The compare of different methods such as AHP, TOPSIS,ELECTRE (I, II, IS, III, IV and A), Grey theory and the case study for warehouse selection are done using this methods and the different characteristic of each method is discussed [37].

AHP method is used in the analysis of the health - Safety and Environmental Risk Assessment of Refineries for the location of the power plant, the risk factor such as health-safety risk, technology risk etc[39].Analysis of the fuel cell strategic technology development solution in the automotive industry using TOPSIS methods, which is used to select best strategic technology for the fuel cell[40].

From all these works, different methods have been used for different applications where each of the method has its own characteristics in finding the best alternatives. The applications which are developed to solve multi choice problems and FMCDS methods which are chosen provides better performance in cases such as supplier chain management in business applications, safety assessment in marine engineering , watershed location and urban distribution centers in public sectors.

2.2.2FMCDM in Banking

Traditionally banks have a fixed set of criteria to process the mortgage or loan application to sanction the loan. The decisions are made rigidly by bank officers after going through the criteria. With fuzzy logic, this process can be made easier and more efficient. Nowadays, banks are increasingly turning to intelligent banking solutions like artificial intelligence to screen out the many loan applications before human officers make the final recommendation and approval. With these approaches, banks can save valuable man-hours on such menial task and dedicate the resources to other productive one. Therefore it brings efficiency to the bank processes and lowers the operating cost for the bank. Table 3 describes some of the bank applications in which FMCDS have been applied to solve the multi criteria problem.

Table 3:FMCDM applications in banking domain

Application

Author and year

Alternatives

Criteria

Problem

Techniques

Best alternative

Banking performance based on

Balanced Scorecard[16]

(Hung-Yi et al., 2009)

Three banks

1.C Bank,

2.S Bank, and

3.U Bank

1. Finance

2. Customer

3.Internal Process

4. Learning and Growth

To rank the banking performance and improve the gaps with three banks as an empirical example.

The three MCDM analytical tools of

1. SAW,

2. TOPSIS,

3. VIKOR

‘‘U Bank"

Fuzzy performance evaluation in Turkish Banking Sector using Analytic Hierarchy Process and TOPSIS[20]

(Nes_eet al., 2009)

The largest five commercial banks of Turkish Banking Sector are examined and these banks are evaluated in terms of several financial and non-financial indicators

Financial criteria:

1. Asset quality

2. Capital adequacy

3. Liquidity

4. Profitability

5. Income and expenditure

Non Financial criteria:

1.Pricing

2.Marketing

3.Productivity

4.Delivery services

To maintain the performance of the banking system since the economy is changing rapidly.

1.Fuzzy sets and fuzzy numbers

2.FAHP

3.TOPSIS

Customer satisfaction and Service quality have been evaluated for commercial banks.

Intelligent phishing detection system for e-banking using fuzzy data mining[22]

(Maher et al., 2010)

The different banks that provide e-banking system

1.URL & Domain Identity

2.Security & Encryption

3. Source Code & Java script

4. Page Style & Contents

5. Web Address Bar

To detect and identify any phishing websites in real-time, particularly for e-banking, is really a complex and dynamic problem involving many factors and criteria.

Fuzzy data mining algorithms and techniques

The sites which posses a good URL and encryption methods holds the best

The impact of 3D e-readiness on e-banking development in Iran: A fuzzy AHP

Analysis[23]

(Mahammadet al., 2008)

1.Human resource readiness

2.Top management readiness

3.Strategy readiness

4.Structure readiness

5. Technology readiness

1.organizational e-readiness

2. industry e-

3. macro environmental e-readiness

New information technologies and emerging business forces have triggered a new wave of financial innovation–electronic banking (e-banking).

Fuzzy AHP

Top management readiness and Strategy

Readiness

To evaluate Taiwan’s commercial bank efficiency allocation model for banks[6]

(Bo Hsiao

et al., 2011)

The 24 Taiwanese commercial bank efficiency evaluation

1.Assets

2.Expense

3.Deposit

4. Fees

5.Amount of loans

Amount of investments

Data envelopment analysis (DEA) mainly utilizes envelopment technology to replace production function in microeconomics. The input and output of decision making units (DMUs) are projected into the attributes to evaluate or measure their performance.

1.Fuzzy BCC

2.Fuzzy SBM

BANK 12

Table 3 describes the various applications of Fuzzy MCDMin banking sector. However, intelligent banking systems has seen its usefulness enhanced with breakthroughs in technology such as fuzzy logic, there is still a need of human interpretation that must be used in dealing with sensitive transactions. It is a still a long way before intelligent banking system can do away with human interaction at all levels. Fuzzy logic allows a computer to reach a decision based on a myriad of factors with different levels of importance [6]. Rather than a yes or no answer, fuzzy logic application reaches a decision based on the weight given to the factors. The artificial intelligence in the application will compare all the potential results both positive and negative before coming to a final conclusion. Fuzzy logic applications using artificial intelligence often make use of neural networks to process the task.

Banking is the sector where fuzzy may occur many times, to overcome this fuzzy MCDM is applied. The fuzzy multi criteria decision making is very much useful in banking application and the performance evaluation of banks has important results for creditors, investors and stakeholder’s since it determines banks’ capabilities to compete in the sector and has a critical importance for the development of the sector [20].

The threat for E-Banking is identifying any phishing websites in real-timeis really a complex and dynamic problem involving many factors and criteria [22].Thebanking and financial industry is transforming itself in unpredictable ways powered in an important way by advances in information technology. Methods like TOPSIS, AHP, FAHP, FBCC and FSBM have been applied in e-banking.

In credit limit allocation model for banks all the criteria have been identified and each criteria assign weight by the experts group, and then criteria have been grouped in some regionwise [23]. The FUZZY TOPSIS method is used to evaluate the criteria in each region and then all the criteria have been ranked. Liner programming assigns credit risk concentrationlimits to the regional heads such that the total value of capital from all location (TVCA) becomes maximum.

The studied works gives an overview of applications of FMCDS where the different methods have been applied and used. Fuzzy is a techniques which is widely used where uncertainty occurs, where the judgment of the result is not clear and optimal, the fuzzy weights have been assigned to each criteria and they have been evaluated. In banking sector FMCDM is used to overcome the uncertainty which was the drawback of the system. It is also being used in E- Banking where users often tend to have problem or dilemma in selecting the links where there is a threat of hacking the passwords through spam mails and hence fuzzy have been applied to identify the phishing web sites and links. The below sections explains about the performance evaluation of MCDM applications.

2.2.3 Fuzzy MCDM in Performance evaluation

Not only general domains, the Fuzzy MCDM methods also applied to evaluate the performance of organization. Table 4 describes FMCDS methods which are applied to evaluate the performance of organizations. Performance of a teacher has been computed by applying COPRAS-G method. The method has been adapted because it can utilize numerical scores in the form of interval marking. Common methodologies reported in past research can handle quantitative numerical score. These methods cannot consider interval making assigned to a particular item whereas COPRAS-G method overcomes thisdrawback [13].

In Evaluation of training performanceof administrative instructors fuzzy set theory is applied to measurement the performance. AHP is applied to obtain criteria weight and for ranking TOPSIS is applied. A Fuzzy MCDM is an approach for evaluating decision alternatives involving subjective judgments made by a group of decision makers. A pair-wise comparison process is used to help individual decision makers make comparative judgments, and a linguistic rating method is used for making absolute judgments [15].

Table 4: FMCDS in Performance evaluation

Application

Author and year

Alternatives

Criteria

Problem

Techniques

Best alternative

Application of MCDM approaches on teachers’ performance evaluation and appraisal[13]

(Sauravdatta, et al., 2008)

5 teachers’

T1

T2

T3

T4

T5

1.Interaction with students

2.Time taken for Problem solving (decision making)

3.Depth of knowledge in own field

4.Dedication, Punctuality and involvement

5.Pedagogy of teaching

To find the best teachers using MCDM technique. The performance and appraisal of each teacher are done separately.

COPRAS-G

T3

Training Performance Evaluation of Administration

Sciences Instructors by Fuzzy MCDM Approach[15]

(Nikoomaram

et al., 2011)

4 Instructor

Instructor A

Instructor B

Instructor C

Instructor D

1.Teaching style, 2.Individual features and social relation, 3.Knowledge level,

4.observance

of educational regulations and 5.Educational tools.

To find the best trainee and the performance od the administrative

science instructors

FMCDM

Instructor A

Power customer satisfaction and profitability analysis using multi-criteria

decision making methods[36]

( RabahMedjoudj et al, 2013)

A1,

A2

A3

A4

cost, reliability, availability, maintainability

and power quality

The aim of this paper is to investigate

Appropriate tools (multi-criteria decision making methods) aiding decision makers to achieve these goals.

Analytic

hierarchy process (AHP) method.

A2

Multi-criteria decision-making method based on interval-valued intuitionistic

fuzzy sets[38]

(V. LakshmanaGomathiNayagamet al., 2011)

1. a car company;

2. a food company;

3. a computer company;

4.an arms company.

1.the risk analysis;

2. the growth analysis;

3. the environmental impact analysis

To find the best company for investment of money in the 4 company using the interval valued intuitionistic fuzzy sets.

interval-valued intuitionistic

fuzzy information

A2 >A4 >A3 >A1

[36]Alternative 1 (A1): Corresponds to the actual state of the electric power system under study,

Alternative 2 (A2): Faults detectors are installed at each substation; consequently the time to fault research is reduced.

Alternative 3 (A3): To alternative 2 (A2), are added remote control switches on outgoing MV lines to reduce the number of customers concerned by a failure.

Alternative 4 (A4): Some overhead circuits are undergrounded and sections of the aging cables are replaced by new ones (are concerned the sections with a number of joints exceeding the threshold value).

The performance evaluation is which is used to measure the performance of the employee in the organization. Evaluations are utilized to determine whether the employee meets the certain criteria and to recommend appropriate follow-up actions. During the evaluation of performance uncertainty occurs, so MCDM approach is applied to measure the performance issues. In Teachers performance evolutionmany alternatives and criteria are applied to analyze the performance of teachers and best teacher is identified using COPRAS-G. In the same way to analyze the training administrative instructor’s performancevarious criteria such as the knowledge level, problem solving skills, cognitive abilities and so [15] have been considered.

Consumer demands for electrical energy are increasingly growing, because this energy is present in all the fields of human activity. The alternatives are technical and the organizational measures often taken in planning and operation phases of electrical power systems is to investigate appropriate tools (multi-criteria decision making methods) aiding decision makers to achieve the goals like customer satisfaction and profit making [36]. Multi-criteria decision-making method based on interval-valued intuitionistic fuzzy sets which is used for determining the best company(a car company; a food company; a computer company; an arms company) to invest the money to obtain more profit[38].

3. Finding of Survey

Multi criteria decision making is discussed in various application in the survey works. The multi criteria decision making is one of the powerful tool for obtaining the best choice for a complex decision making situations using various methods such as Fuzzy AHP, ELECTRE, TOPSIS, Grey theory etc. The evaluation of the criteria and ranking which is best alternative is done using this various technique. The outcome of this survey has been described below,

MCDM is the powerful technique for decision making

The MCDM is used in many application such as performance elevation , warehouse location, supplier selection , information delivery , supply chain management , Assessment of health-care waste treatment , Banking performance, e-banking , teachers’ performance, selection process. The decision making in all these application is efficient and are success. Table discuss about various

The performance of the MCDM is very high in the business organization which is used to solve the complexity of the problem. MCDM is used in all real world application such as warehouse location, environment assessment. The performance of the organization is increased by better solution can be obtained by MCDM.

In the business, the collections of relevant information are done, to provide the better solution for the problem. The relevant information is very useful in the making the decision in the complex problem which occurs in the organization.

Banking

Performance mgmt.

Selection process

Business

Partner selection

Risk mgmt.

Information delivery

Environment assessment

Voice of customer

Education

Health care

Marine egg.

financial investment decisions

financial ratios

manufacturing systems

demand forecasting

material selection

bioinformatics

The methods of MCDM are unique in there characteristic, which can be used in the certain problem that suits there characteristic. For example, the TOPSIS method, that has chosen the best alternative based on a maximization of the distance from the negative ideal point and minimization of the distance from the positive ideal point. Grey theory methods, examines the interactional analysis when the decision-making process is not clear, there are a great number of input data and it is discrete and insufficient data.

COMPARSION OF AHP AND FUZZY AHP

Analytic hierarchy process AHP is a method used for ranking purpose in selecting the best one when the decision maker has multiple criteria. Here this method helps the decision makers to select a better alternative from all by satisfying the minimal score to rank each decision alternative based on how well each alternative meets them.

Fuzzy AHP, where it helps the human to make quantitative predictions as they are not well versed, but they are equally better in making quantitative forecasting. The uncertainty occurs during the judgments where in turn in consistency arises in between the alternatives.

Fuzzy pair wise comparisons here states that there are many criteria’s but if any criteria has a less important among all then it can be weighed ass zero unlike other methods. Though that criterion is handled for the decision making process, if it has no importance when compared to all others. In the classic AHP method, deterministic values and operations do not permits such a situation "having zero weighed", but if a criterion is evaluated as less than all of the others, then the numerical weight of the criteria will be near to zero. Fuzzy AHP can merely ignore the criteria that have less importance whereas the classic AHP where it will be given with so weight. This can also be an advantage for fuzzy-AHP presenting additional information for decision maker that there is no difference between the existence or nonexistence of such a criterion. Therefore, the decision maker can focus on the more important criteria.

Classical and fuzzy methods are not the rivals with each other at same conditions. The important point is that if the information / evaluations are certain, classical method should be chosen; if the information / evaluations are not certain, fuzzy method should be chosen. In recent years, because of the uniqueness of information and decision makers, probable deviation should be integrated to the decision making processes, and because of that for each decision making method, a fuzzy version is developed. Fuzzy AHP method is a natural result of this necessity.

Linguistic and subjective evaluations take place in questionnaire form. Each linguistic variable has its own numerical value in the predefined scale. In classical AHP these numerical values are exact numbers whereas in fuzzy AHP method they are intervals between two numbers.

Comparison of ELECTRE, TOPSIS and GREY THEORY

TOPSIS method, that selects the best alternative by the process of maximization of the distance from the negative ideal point and minimization of the distance from the positive ideal point, was not only applied to areas such as performance evaluation with the use of financial investment decisions and financial ratios but also applied to problems such as flexible manufacturing systems and selection of production processes, within the scope of operation management. Similarly, ELECTRE methods (Electre I, IS, II, III, IV,A) that selects the best alternative by means of pairwise comparison of all alternatives; within the decision problems , especially has been applied to solve the issue of environmental management, environmental valuation

Grey Theory, extremely has a high mathematical analysis of the systems that are partly known and partly Unknown and defined as ‘‘weak knowledge’’ and ‘‘insufficient data’’. Grey Theory examines the interactional analysis when the decision-making process is not obvious, there are a great number of input data and it is distinct and insufficient. In recent years, Grey Theory has been successfully applied in many decision making problems.

THE CORE (MAIN) PROCESS

In TOPSIS method, the calculation of each alternative distance from the positive ideal and the negative ideal solutions draws attention. While ELECTRE I and ELECTRE II methods are separated from the other methods through the determination of concordance and discordance matrices for each criteria’s and alternative pair, ELECTRE III method, different from the other methods, is based on the principle of fuzzy logic and uses the preference and indifference thresholds while determining the concordance and discordance indexes

THE TYPE AND NUMBER OF OUTRANKING RELATIONSHIP

The number of pairwise comparison matrix can be too many which can be a disadvantage of AHP and in the situations like when the number of alternative and criteria are a lot, the opportunity of carrying out the methodology substantially prevented. TOPSIS and ELECTRE I methodologies need less input compared to AHP and eliminate the necessity of comparisons of pairs.

THE CONTROL OF CONSISTENCY

The limitation of consistency is one of the most important advantages of AHP. The consistency is not controlled in TOPSIS, ELECTRE I and ELECTRE II methods.

Furthermore, since it is necessary to make pairwise comparisons in all the levels of hierarchy, it gets harder to perform AHP as the number of alternatives and criteria gets increased for more complex problems. On the other hand, AHP can be performed easily disregarding the data applying evaluation of alternatives based on criteria is quantitative or qualitative. TOPSIS method gets attention as its simplicity in perception and use. TOPSIS and ELECTRE methods can be performed easily for a problem with more numbers of alternatives and criteria’s.

PROBLEM DEFITION

Form the survey we have definition a problem using the Fuzzy MCDM methods, which is a comparative analysis of the Fuzzy MCDM methods for information delivery in banking sector. In this we are comparing the three different methods such as TOPSIS, ELECTRE and Grey Theory. The decision making is used to produce the right information to right user in right time. Banking is the sector where more number user and more information. There arise uncertainties to which user which information should be produced. To solve is uncertainty, three methods are compare and the right information for the right user is given.

4. Research Opportunities

The true goal in integrated decision-making support is to provide the decision-maker with the ability to look into the future, and to make the best possible decision based on past and present information and future predictions. In the case of sustainable development, this means that to be able to predict in advance the risk and vulnerability of populations and infrastructure to hazards, both natural and man-induced. This requires that data be transformed into knowledge, and that the consequences of information use, as well as decision-making and participatory processes, be analyzed carefully. The conclusion obtained from the survey works are, the use of fuzzy will give only an approximate solution for problem. The use of fuzzy is to analyze the quantitative and qualitative data for any application. The different methods under FMCDM help us to perform may subtasks between where evaluation and ranking are done by different methods. Each method has its own uniqueness. This is how fuzzy in analyzing an application. In previous works the mapping of information has been done where what information is needed for which users, for e.g., Government needs a lots of information when compared to other users like customers, management etc.,. So the further works can be enhanced by sending information to the users via correct medium and right time. The work to be done is to customize the correct information, where a student as a customer can get enough information regarding the educational loans. The visualization is mainly used to attract the users to get accessed often.

4.1 Fuzzy MCDM Application and Fuzzy MCDM Methods

The use of fuzzy multi criteria techniques are applied in various fields such as Banking sectors, issues such as urban distribution centers, water shed allocation, safety assessment, and performance evolution of business organizations. The statistical report for the some of the areas in which multi criteria decision making is used is described in Table5.

S.no

Banking

Business

Environment assessment

Performance evaluation

1

To evaluate Banking performance based on

Balanced Scorecard

To find the best supplier for mold and die manufacturing concern in the enterprises

Location planning for urban distribution centers

To find the best teachers using MCDM technique. The performance and appraisal of each teacher are

2

To analysis performance of the banking system during economy is changing rapidly

Finding the best supplier who is able to provide the right quality products and/or services at the right price with the right quantities and at the right time.

In finding the most robust alternative among these seven criteria for water planet location

To find the best trainee and the performance of the administrative

science instructors

3

To detect the phishing mails

Supplier selection, in selecting the best suppliers who are able to provide the buyer with the right quality products

To enhance safety by mitigating risks and increasing the reliability of a system.

To find the best company for investment of money in the 4 company using the interval valued intuitionist fuzzy sets.

4

New information technologies and emerging business forces in banking

For the selection of cans supplier/Suppliers at Nitrides Factory in Amman-Jordan

Evaluation of

HCW treatment alternatives

to investigate

Appropriate tools (multi-criteria decision making methods) aiding decision makers to achieve these goals

5

Data envelopment analysis (DEA) mainly utilizes envelopment technology to replace production function in microeconomics

Implementation of a warehouse location selection problem

6

To find the best location for the implementation of the power plant using the AHP

Contribution

5

4

6

4

Table 5.List of application in FMCDS

Table 5 describes the analysis report of the multi criteria techniques which is widely used in various applications. Table 5, also describes the clear essence of the domains in which MCDM is applied. Most of the multi criteria based problems fuzzy MCDM approach is applied due to its capability of solving uncertainty issues and it gives the best determination for the decision makers, so that MCDM method is used in many domains. Each MCDM method is chosen according to difficulty of the problem. Table 6 describes about most widely applied methods in multi criteria decision making and these methods are ranked based on its applicability and usage in various domains.

S. No

MCDM Methods

Usage

1

FMCDM

5

2

TOPSIS

9

3

FAHP

6

4

VIKOR

2

5

ELECTRE

5

6

Others

3

Table 7. MCDM Methods and its usage

Based on these input a graph is plotted which is depicted in figure 1 it shows most widely applied techniques of MCDM. The ranking of the MCDM methods based on this survey is given in the figure 1 according to its usage in various applications.

Figure 1.Usage of MCDM methods

From the figure 1 it known that TOPSIS method is applied mostly in many applications. The next is FMCDM method that has been used in the fuzzy application for solving the uncertainty. A Fuzzy MCDM is an approach for evaluating decision alternatives involving subjective judgments made by a group of decision makers. A pair-wise comparison process is used to help individual decision makers to make comparative judgments, and a linguistic rating method is used for making absolute judgments. The other methods are Fuzzy BCC, Fuzzy SBM, FSROWA and COPRAS-G. This survey outlines research opportunities in MCDM, the features of MCMD can be applied to any domain when multiple choices are available for decision making.

5. Conclusion

This survey finds opportunities in multi criteria decision making where decision making involves multiple choices. Fuzzy multi criteria decision making is used in many applications like Banking, performance evaluation, safety assessment, and etc. FMCDS is applied to domains in which we need to evaluate more alternatives and multiple criteria and from that select the best alternative. According to the problem and its domain the MCDM methods have been selected. Very limited work has been applied using multi criteria decision making. This survey finds more work on banking where fuzzy occur most often in decision making. Fuzzy based MCDM is suitable for approximate problem spaces. Thus FMCDS can be applied to analyze quantitative and qualitative data of any application to arrive the solution. The different methods under FMCDM help us to perform tasks between where evaluation and ranking are done by different methods. Each method has its own uniqueness.