Decision Making Model for Electronic Tender Evaluation (eTE) using Fuzzy AHP with Extent Analysis Method 1

Decision Making Model for Electronic Tender Evaluation (eTE) using

Fuzzy AHP with Extent Analysis Method

1Ros Haslinda Alias,2Noor Maizura Mohamad Noor,, 3Ali Selamat, 4Md Yazid Mohd Saman, 5Mohd Lazim Abdullah

1,2,4Computer Science Department,

5Department of Mathematics, Faculty of Science and

Technology, Universiti Malaysia Terengganu, [email protected],

[email protected],

[email protected],

[email protected]

3Software Engineering Department, Faculty of Computer Science and Information

System,Universiti Teknologi Malaysia, [email protected]

Abstract Transformation in procurement technique aims to create a sustainability development and

innovative outsourcing strategy. It can increase the probability of project success with quality and give

an added value in money and contract times. Wide range of issues should be considered in producing

a sincere and trusted decision within respective agencies. Our research proposed a procurement

transformation model for selecting the optimal result of contractor selection under multi criteria

environments. The corresponding data are centralizing, easy to manage and can be updated

accordingly. Fuzzy comparing judgment is used as the method to overcome the ambiguity aspects in

selecting significant preferences by decision maker regarding to the subjective opinion. Decision

making process is represented by the integration of evaluation model to produce a final result. Finally,

the model was tested in tender evaluation processes to determine the most beneficial contractor to

perform the construction project.

Keywords: Contractor Selection, Fuzzy Sets, Analytic Hierarchy Process, Extent Analysis

1. Introduction

Tender evaluations and contractor selections continues to be an area of significant importance and

interest to organizations responsible for delivering project outcomes [1]. There were a few evolutions in tender management’s procedures. The methods have been transformed from the traditional manual paper-based tender processes until web based tendering processes. This process is aim to select the most suitable contractor with the competitive price and high capability to ensure a smooth and successful project. However the need to create an appropriate evaluation technique is urgently required, since various issues must be accounted to tackle the complexity situations along the processes. Some of the issues are companies conducting the project do not have an approval certificate, corruption elements within organizations, weak monitoring systems, bureaucracy, lack of staffs and in some cases, the company fails to follow the design standards, due to the financial constraints caused by the high price of materials. These issues need to be reviewed and drastic actions should be taken to tackle the problems. This is to prevent incidents such as building collapse; projects did not follow the schedule, abandoned projects and low quality of constructions. Therefore, the tender committees play an important role in making selection of the most eligible contractor to avoid corruption and illegal action in project handling.

There are complicated phases throughout the evaluation stages. The processes deal with multi-criteria, multi-user problem and engage with handling of imprecise and vague information. Therefore, it needs a combination of methods to structure all the processes in more effective way. Fuzzy sets theory has been proved to be a very convenient way for searching solutions of the problems containing elements of human subjectivity, such as making decisions in order to choose construction contractors [2]. In our research, the tender board from top level management need to identify entire project requirements as stated in the organizational standard produced by committee based on the project specification. First, they need to clarify evaluation criteria and also specify the quantitative or qualitative value for each project demand. Then, they also need to nominate which assessors deserve to make the evaluation. On behalf of the assessors, they need to judge the alternatives as defined by the filtering process in the earlier stage. Thus, selecting the most beneficial contractor takes into

Decision Making Model for Electronic Tender Evaluation (eTE) using Fuzzy AHP with Extent Analysis Method Ros Haslinda Alias, Noor Maizura Mohamad Noor, Ali Selamat, Md Yazid Mohd Saman, Mohd Lazim Abdullah

International Journal of Digital Content Technology and its Applications(JDCTA) Volume6,Number22,December 2012 doi:10.4156/jdcta.vol6.issue22.8

79

consideration of the various points of view from different areas of expertise. This paper presents a model of using one of the multi criteria decision making methods known as the Analytic Hierarchy Process (AHP) using fuzzy comparison judgment based on specification provided by Jabatan Kerja Raya Malaysia (JKRM). Section two represents related works on method used in contractor selection followed by proposed evaluation model in section 3. The next section introduces fuzzy AHP and computation of extent analysis method with some modifications. Then, a brief scenario about contractor selection can be found in section four. Numerical example is provided correspondingly in the last section before the conclusion.

2. Related works

A reliable and effective evaluation requires a decision maker to analyze a large amount of data as well as consider many dimensions of multi criteria decision making issues and attributes [3]. Making a right selection is a very subjective opinion and it is depends on many characteristics. Selection is the process of aggregating the results of evaluation to identify the optimum choice through the process of investigating or measuring contractor attributes [4]. Processes of identifying relevant criteria with the appropriate weights by the assessors are being discussed by many researchers. Ref. [1] uses an experimental design approach to quantify the importance of nine common criteria used in an actual evaluation and selection of a contractor. Results shows that, the past project performances, technical expertise and cost are the most important indicators to determine that the contractor have both organizational experiences and reputations. Various techniques are available in contractor selection. Most applied method is known as AHP [5, 6]. Fuzzy characteristic are applied in cases of multi attribute decision making where there are many qualitative criteria need to be measured. Ref. [7] have initiated the use of systematic multi-criteria decision analysis techniques for contractor selection and bid evaluation based on utility theory which permits different types of contractor capabilities to be evaluated. Ref. [8] purposed a multiple-layer fuzzy pattern recognition approach to solve contractor selection problem by integrating judgments, experience and preferences of decision-makers. They used paired comparison method to decide relative membership degrees of qualitative criteria as well as weights of the criteria set. Ref. [9] suggested the application of an extended fuzzy AHP method to the process of group decision making to determines criteria weights for bidder assessment. Result showed that the proposed fuzzy AHP method was superior to the traditional AHP in terms of improved quality of criteria prioritization. Ref. [10] proposed a web-based sub-contractor evaluation system called WEBSES by which it can be evaluated based on a combined criterion. Ref. [5] introduced an integrated method in contractor selection by statistical and enhanced analytic hierarchy process (AHP) model into guided AHP, guided ranked AHP and AHP-magic to improve the usability features in contractor evaluation. Ref. [11] presented an approach which used a graph theory and matrix as the decision analysis tools for contractor selection to identify an eligible contractor. Ref. [12] proposed a fuzzy set theory combined with quality function deployment as a hybrid solution for contractor selection in housing refurbishment and the result revealed that the proposed hybrid fuzzy-QFD approach has potential for handling multiple criteria decision making problem. Although various procedures for a contractor’s selection techniques including multi criteria [7, 9, 10], hybrid solution [12], and fuzzy combination approaches [8, 9], have been practically applied, the concept of actuality in contractor evaluation need to be re-considered due to the diversity of factors throughout the process that may affect the project outcome. There are many phases involved in the selection processes and they are complicated to be modeled in the required structure. Therefore, all ambiguity issues occur during the processes must be properly engaged to produce a comprehensive result for each project within a given time.

3. Proposed evaluation model

Wide range of criteria has to be particularly judged to form a listing of selected tenders for each phase. Normal selection processes are related to the specific characteristics of the project and the requirements of the contractor or consultant involved. Important aspect in model building is to capture system abstraction and simplification by organizing theoretical beliefs and empirical observations


80

about the system. Figure 1 shows our proposed decision process model for evaluation techniques purposely used in construction site as the assessment approach in making a decision. Our approach involves of integrated components with relevant process model. Clients/consultants are part of crucial agents in performing the decision making process for this evaluation model. The interaction between decision making agents and two main modules are explained in the next section.

Figure 1. The integration of decision making process of evaluation model

3.1 Filtering module

Filtering module aims to define the number of eligible alternatives deserve for evaluation. It consists of statistical model and prequalification model. Cut off price (COP) is a minimum price level which is reasonable to be accepted by tender committees. Price below than cut off is considered as too low and does not promise for the successful project. Statistical method on cut off price depends on mean and standard deviation of the overall bidding price, after freak price was declared. Freak price is the value that is considered either too low or too high. Contractor in this category must be rejected from the list for the next module. Figure 2 shows the flow chart for filtering process.

Figure 2. Flowchart of filtering for alternatives

3.2 Ranking module

Based on empirical evaluation carried out by [5], 90% respondents have agreed that performing

tender evaluation based on statistical model alone cannot give the best and fair results. This is due to

Decision Making

1. Filtering: (Statistical model)

2. Ranking: (MCDM model)


81

the fact that, complexity in decision making situations does not only involves quantitative elements, but also include qualitative criteria, which occupies multiple scales and comparisons. Method of ranking tenderers with weighting and scoring of evaluation is the most popular technique used in tender evaluation [13]. The ability as well as simplicity to assign a preference rank for general decision making situations is simultaneously needed [14]. We have implemented the multi criteria analysis to determine the attributes and value on the process structure, and outcome from decision making. Figure 3 depicts the structure to define the ranking based on scoring for each attributes. In AHP, a composite problem must be structured into hierarchical levels to represent all the features in the problem domain. It starts with the optimal solution or the main goal, followed by the defined criteria and sub-criteria and the last are alternatives or choices. In order to evaluate relative importance of alternatives, pair-wise comparisons are performed for various criteria, which are difficult to be quantified. Pair-wise weighting among n elements in each level leads to an approximation to the ratio of aij=wi/wj which is the weight of element i to element j. Using pair wise comparison, the relative importance of one criterion over another can be expressed.

Figure 3. Hierarchical structure of selection criteria

In our numerical example three eligible candidates are selected (alternatives) A = (��, ��, ��). Based on evaluation standard by JKRM, tender opening committee establishes a required specification used to evaluate all the candidates. Four important criteria with a relevant sub-criteria are considered: (C1:financial, C2:performance, C3:staff, C4:equipment; C1-1:asset, C1-2:liability, C2-1:current, C2-2:previous, C3-1:experience and C3-2:qualification). To complete the task decision maker needs to assign their preference in pair-wise comparison judgment. 3.2.1 TFN fundamental

Human judgments, including preferences, are often vague and preferences cannot be estimated in exact numerical values [15, 16]. Basically human judgments are influenced by intuition and emotion which causes the uncertainties in making preferences. Therefore it is inadequate to model real-life situations. Fuzzy set theory offers a possibility of handling data and information involving subjective characteristics of human nature in the decision-making process was pioneered by Zadeh [17]. In this study, membership function, µA(x) is defined by triangular fuzzy number (TFN) to express decision-makers opinion as depicted in figure 4.

Goal

C1 C2 C3 C4

C1-1 C1-2 C2-1 C2-2 C3-1 C3-2

A1 A2 A3 A4


82

Figure 4. Triangular fuzzy membership function

The reason for using TFN by many researchers is because of its ease to be applied and could be intuitively represented. A fuzzy subset Ã of X to be TFN if its membership function ~(X) that map each element x in X to real number in the interval [18]. The basic features of TFN is defined as [19]:

( x – l ) / ( m – l ), l ≤ x ≤ m, ~(X) = ( u – x ) / ( u – m ), m ≤ x ≤ u, (1) 0, otherwise

The TFN is donated by Ã = ( l, m, u ) where l is for lower bound and u is for upper bound of fuzzy number Ã. The main operational laws of two TFNs Ã1 = ( l1, m1, u1 ) and Ã2 = (l2, m2, u2 ) are as follows [20]:

Ã1 + Ã2 = ( l1 + l2 , m1 + m2 , u1 + u2 ) (2)

Ã1 – Ã2 =( l1 – l2 , m1 – m2 , u1 – u2 ) (3)

Ã1 * Ã2 = ( l1 * l2 , m1 * m2 , u1 * u2 ) for li > 0, mi > 0, ui > 0 (4)

Ã1 / Ã2 = ( l1 / u2 , m1 / m2 , u1 / l2 ) for li > 0, mi > 0, ui > 0 (5)

(Ã)-1 = ( l, m, u )-1 for l,m,u > 0 (6)

3.2.2 Fuzzy AHP with extent analysis method

The fuzzy AHP allows the user to analyze a decision problem in detail in terms of precision of

estimation and assessment consistencies [9]. This section describes an empirical approach to evaluate preferences on selected alternatives. Fuzzy comparing judgment is used to tackle the ambiguity aspect in choosing significant preferences which may affect the selection of most beneficial solution. Besides Buckley method [21] and Van Laarhoven and Pedrycz method [19], extent analysis method [22] is one of the most popular and preferred methods in the fuzzy AHP field. The reason is based on the simplicity of this approach, where the steps involved exhibits a similarity to the conventional AHP method. Therefore, it appears to be easier than the other fuzzy-AHP techniques. This method have been adopted by many researchers [23-26], however based on empirical research by [27], they found that the extent analysis method still cannot estimate the true weights from a fuzzy comparison matrix and has led to quite a number of misapplications in the literature. Arising from those researches, we used total integral value, together with an index of optimism [28] as a method for ranking fuzzy numbers; to reflect the decision maker's optimistic attitude. In this paper, the method is based on fuzzy AHP with the use of extent analysis and slightly modified version of Chang’s method was applied by reflecting the discussions in [25]. Below are the outlines of the extent analysis method of fuzzy AHP:

Let E be an object set: X = {x1, x2, …. , xi, …, xm} and a goal set is denoted by U = {u1, u2, ... , uj, … , un}. According to the method of extent analysis proposed by Chang (1996), take each object and perform extent analysis for each goal respectively. Therefore, we can get m extent analysis values obtained for each object, with the following procedures:

l m u

1.0 ~(X)


83

� �� , � �� , … . . , � �� , � = 1, 2, … . , �,(7)

Where all the� �� , (j = 1,2,…,m) are TFN. Equation (8) is used to calculate the value of fuzzy synthetic extent with respect to the i-th object is defined as:

Si= � Mgi

j ⨀ �� Mgi

j

m

j=1

n

i=1

�-1m

j=1

(8)

In order to obtain∑ �� , perform the fuzzy addition operation of m extent analysis value such that:

� �� =� �, � !��

�� , � "��

�� (9)��

��

and to obtain $∑ ∑ ��%&�� '(�, perform the fuzzy addition operation of ��

(j=1,2,…,m) values such

that:

� � ��

��%

&�� =� �, � !�%

�� , � "�%

��%

�� (10)

The following steps is modified from [22] by using some modification from [27], compute the inverse vector in Eq. (8) such that:

*� � ��&�

��%

&�� +(� = , 1∑ &%&�� + ∑ "&%&�� , 1∑ !&%&�� , 1∑ "&%&�� + ∑ &%&�� .(11)

To represent the relative important, the last step is using total integral value with an index of optimization introduced by [28]. It is used to derive the priorities vector of the synthetic extent value by the following equation: /0123456 = 12 7(!5 + "5) + 12(1 − 7)( 5 + !5)

= 1297"5 +!5 + (1 − 7) 5:(12)

Where value 7;90,1: is an index of optimism represents optimism degree of decision maker. The final

step is to calculate weight vector < = (=�, =�, … . . , =%)0 of fuzzy judgment matrix using following equation: =5 = /0(>4?)1∑ /01234@6%@�� A = 1,2, … . , �.(13)

Where =5 is important weight vector normalized in crisp value.


84

3.3 Design of the tender evaluation system

Based on regulation practice in Malaysia, consultant collaborates with a client to prepare tender documents, publish and make the evaluations on the contractors. Client in-house staffs only responsible to manage the appointed consultants and verify whether their job is fulfill. Consultant will hire architect, civil, structure, mechanical and electrical engineers, as well as quantity surveyor to prepare the tender documents. Consultant collaborates with client to prepare tender documents, publish and make the evaluations on the contractor. Tenders are evaluated by more than two committee members. The compulsory person is the quantity surveyor or technical officer in the related area. There are four guidelines stated by JKRM for basic tender evaluations:

i. Comply with the tender condition such as register with authority in associated class/field, tender signature; attach compulsory documents and other relevant conditions.

ii. Technical ability of contractor. iii. Financial ability of tenderer in capital encircling, nett worth, credit capability, balance

work in progress and so on. iv. Analysis on tender price with time duration of work accomplishment. If time duration is

categorized as important factor, it must be clearly declared in tender document.

Therefore, there is the need for a comprehensive system to improve communications among all individuals involved in tendering processes so that, an enhanced productivity for all parties; consultants/clients and contractors could be obtained. Figure 5 shows the framework of eTE on the centralized database.

Figure 5. Evaluation framework of eTE

4. Numerical example

The proposed method currently applied to solve the problem with appropriate computation has been discussed earlier. Decision makers use linguistic terms to describe the relative important of their preferences as presented in Table 1.

CONTRACTORS

Mod 2: Registration

- Register - Tender searching - Purchase tender document - Upload document

Mod 3: Evaluation

- Select project - Filtering and analysis - Select the preference - Populate overall result for project

CLIENT/CONSULTANT

Mod 1: Setup

- Prepare advertisement - Prepare tender document - Set the criteria’s, -Assign Tender Opening Committee

e-Tender

Evaluation

(eTE)

b. Retrieve Mod 1

c. Submit Mod 2

a. Upload & store Mod 1

Mod 4: Notification

- Tender result

d. Access eTE & evaluate e. Submit Mod 3

f. Submit Mod 3


85

Table 1. Linguistic terms used for selecting user preferences

Linguistic Terms Triangular fuzzy scale Reciprocal scale Just equal, (JE) (1,1,1) (1,1,1) Equally important, (EI) (1/2, 1, 3/2) (2/3, 1, 2) Weakly more important, (WMI) (1, 3/2, 2) (1/2, 2/3, 1) Strongly more important, (SMI) (3/2, 2, 5/2 (2/5, 1/2, 2/3) Very Strongly more important,(VSMI) (2, 5/2, 3) (1/3, 2/5, 1/2) Absolutely more important, (AMI) (5/2, 3, 7/2) (2/7, 1/3, 2/5)

The algorithm presented to achieve the goal can be done according using the following steps: Step 1: Define goal, criteria and list of decision-makers pertinent to project specification. Step 2: Generate the alternatives (filtering module). Step 3: Identify weight of criteria and sub-criteria. Rating example of criteria and sub-criteria are

given as Table 2.

Table 2. Local weight of criteria and sub-criteria

Criteria Local weight, Wx

C1 0.476 C1-1 0.833 C1-2 0.167 C2 0.333 C2-1 0.875 C2-2 0.125 C3 0.143 C3-1 0.125 C3-2 0.875 C4 0.048

Step 4: Establish fuzzy decision matrix of pair-wise comparison C4 = {�D&�} of the type:

C4 = F 1 �D�� ⋯ �D�%�D�� 1 ⋯ �D�%⋯ ⋯ 1 ⋯�D%� �D%� ⋯ 1 H

Where �D�& = 1/�D&� = (1/"D&� , 1/!J &� , 1/"D&�). Decision makers make their own judgments expressing their individual opinions for a set of alternatives. There are an of alternatives for evaluator to rank them in order of preferences. Table 3 shows example for fuzzy comparison judgment.

Table 3. Local weight of criteria example of alternatives with respect to C1-1

A1 A2 A3 A4

A1 (1,1,1) (5/2,3,7/2) (2/7, 1/3, 2/5) (1, 3/2, 2) A2 (2/7, 1/3, 2/5) (1,1,1) (3/2, 2, 5/2) (1, 3/2, 2) A3 (5/2, 3, 7/2) (2/5, 1/2, 2/3) (1,1,1) (2/5, 1/2, 2/3) A4 (1/2, 2/3, 1) (1/2, 2/3, 1) (3/2, 2, 5/2) (1,1,1)

Step 5: Assign preferences using linguistic term to pair wise comparisons for each alternative regarding to its criteria. This scale has been used by [23, 24] for solving fuzzy decision making problem.

Step 6: Calculate extent analysis method for each parameter being evaluated using Eq. (8) – (11). By applying Eq. (8), we have

S1 = (0.17,0.23,0.30)


86

S2 = (0.14,0.19,0.26) S3 = (0.15,0.20,0.26) S4 = (0.13,0.18,0.25)

By applying Eq.(11), we have

I1 = 0.228 I2 = 0.197 I3 = 0.203 I4 = 0.184

Step 7: Derive the priorities vector of the synthetic extent value using Eq. (12). Step 8: Obtain the weighting vector assessed by each decision maker, as expressed by Eq. (13). Using

index of optimization α = 0.5 we get normalized weight vector for alternative with respect to C1-1 WC1-1= (0.28, 0.24, 0.25, 0.23)T

The same process is applied to the rest of criteria to get the weight with respect to each alternative as shown in Table 4.

Table 4. Weights for alternatives with respect to each criterion

C1-1 C1-2 C2-1 C2-2 C3-1 C3-2 C4

A1 0.28 0.28 0.30 0.31 0.31 0.30 0.29 A2 0.24 0.24 0.26 0.24 0.19 0.24 0.26 A3 0.25 0.25 0.29 0.25 0.28 0.21 0.22 A4 0.23 0.23 0.15 0.21 0.22 0.25 0.23

From the assessment made by decision-maker, overall weights is derived by adding all levels in the hierarchy and multiply all the weight of criteria with the weight for each alternative to obtain the global weight as defined in Table 5.

Table 5. Weight assessment result from a decision maker

A1 A2 A3 A4 Weight 0.29 0.25 0.26 0.21 Rank 1 3 2 4

The alternative represents the highest score would be considered at the first rank while the lowest represents the last rank. As the result, contractor A1 is the most preferred choice made by the decision makers.

5. Conclusion and future works

Success procurement at the best price possible with minimum conflicts and disputes becomes the purposed of our research. To put in place procurement solution, one standard method or measurement need to be applied where there are problems bound with devastating consequences. Evaluation is the most important segment in the whole selection domain. Therefore, implementing the effective evaluation technique becomes an essential element to achieve the decision quality. This study presented a MCDM technique based on fuzzy comparison judgment to perform selection of the most beneficial contractor. Subjective judgment encourages of ambiguity and imprecise assessment from decision makers are well defined using fuzzy comparison judgment. Linguistic terms used for the assessment are more convenient and intuitive for a decision maker during the evaluation process. In future, we may apply a recommender agent selection to save time and to deliver a cost-effective solution to respective agencies. In automated selection process, the approach used by the agent is based on rule-based expert system which is generally applied for recommendation systems. Agent retrieves and analyses the scores in the knowledge base and provides the benefits of learning tendencies to give


87

a suggestion. In particular, evaluation panels can take the modeling agent’s advices or proceed independently by reviewing the score.

6. References

[1] D. J. Watt, B. Kayis, and K. Willey, "The Relative Importance of Tender Evaluation and

Contractor Selection Criteria," International Journal of Project Management, vol. 28, pp. 51–60, 2010.

[2] E. Plebankiewicz, "Contractor Prequalification Model using Fuzzy Sets," Civil Engineering and

Management, vol. 15, pp. 377–385, 2009. [3] S. Chen, R. Li, and B. Tang, "Group-buying Website Evaluation Based on Combination of

TOPSIS,Entropy Weight and FAHP " Journal of Convergence Information Technology, vol. 7, p. 130 ~ 139, 2012.

[4] G. D. Holt, "Which Contractor Selection Methodology?," International Journal of Project

Management, vol. 16, pp. 153-164, 1998. [5] F. Ahmad and M. Y. M. Saman, "Integrated Decision Support Model for Construction Project

Tendering," Journal of Computing, vol. 3, pp. 82-88, 2011. [6] K. P. Anagnostopoulos and A. P. Vavatsikos, "An AHP Model for Construction Contractor

Prequalification," Operational Research. An International Journal, vol. 6, pp. 333-346, 2006. [7] Z. Hatush and M. Skitmore, "Contractor Selection using Multicriteria Utility Theory: An

Additive Model," Building and Environment, vol. 33, pp. 105–115, 1998. [8] L. Yawei, C. Shouyu, and N. Xiangtian, "Fuzzy Pattern Recognition Approach to Construction

Contractor Selection," Fuzzy Optimization and Decision Making, vol. 4, pp. 103-118, 2005. [9] P. Jaskowski, S. Biruk, and R. Bucon, "Assessing Contractor Selection Criteria Weights with

Fuzzy AHP Method Application in Group Decision Environment," Automation in Construction,

vol. 19, pp. 120-126, 2010. [10] G. Arslan, S. Kivrak, M. T. Birgonul, and I. Dikmen, "Improving Sub-contractor Selection

Process in Construction Projects: Web-based Sub-contractor Evaluation System (WEBSES)," Automation in Construction, vol. 17, pp. 480-488, 2008.

[11] M. Darvish, M. Yasaei, and A. Saeedi, "Application of the Graph Theory and Matrix Methods to Contractor Ranking," International Journal of Project Management, vol. 27, pp. 610-619, 2009.

[12] Y. K. Juan, Y. H. Perng, D. C. Lacouture, and K. S. Lu, "Housing Refurbishment Contractors Selection based on a Hybrid Fuzzy-QFD Approach," Automation in Construction, vol. 18, pp. 139-144, 2009.

[13] N. Aissaouia, M. Haouaria, and E. Hassinib, "Supplier Selection and Order Lot Sizing Modeling: A Review," Computers & Operations Research, vol. 34, pp. 3516 – 3540, 2007.

[14] T. L. Saaty, "How to Make a Decision: The Analytic Hierarchy Process," European Journal of

Operational Research vol. 48, pp. 9-26, 1990. [15] D. Zhang, J. Zhang, K.-K. Lai, and Y. Lu, "An Novel Approach to Supplier Selection based on

Vague Sets Group Decision," Expert Systems with Applications, vol. 36, pp. 9557–9563, 2009. [16] S. Khorshid, "Soft Consensus Model based on Coincidence Between Positive and Negative Ideal

Degrees of Agreement under a Group Decision-making Fuzzy Environment," Expert Systems

with Applications, vol. 37, pp. 3977-3985, 2010. [17] L. Zadeh, "Fuzzy Sets," Information and Control, vol. 8, pp. 338-353, 1965. [18] A. K. Choudhurya, R. Shankarb, and M. K. Tiwaria, "Consensus-based Intelligent Group

Decision Making Model for the Selection of Advanced Technology," Decision Support Systems

archive, vol. 42, pp. 1776-1799, 2006. [19] P. J. M. V. Laarhoven and W. Pedrycz, "A fuzzy Extension of Saaty’s Priority Theory," Fuzzy

Sets and Systems, vol. 11, pp. 229–241, 1983. [20] A. Kaufman and M. M. Gupta, Introduction to Fuzzy Arithmetic: Theory and Application. New

York: Van Nostrand Reinhold, 1991. [21] J. J. Buckley, "Fuzzy Hierarchical Analysis," Fuzzy Sets and Systems, vol. 17, pp. 233-247,

1985. [22] D. Y. Chang, "Applications of the Extent Analysis Method on Fuzzy AHP," European Journal

of Operational Research, vol. 95, pp. 649-655, 1996.


88

[23] C. Kahraman, U. Cebeci, and Z. Ulukan, "Multi-criteria Supplier Selection using Fuzzy AHP," Logistics Information Management, vol. 16, pp. 382 - 394, 2003.

[24] M. Dagdeviren and I. Yuksel, "Developing a Fuzzy Analytic Hierarchy Process (AHP) Model for Behavior-based Safety Management," Information Sciences 178, vol. 168, pp. 1717–1733, 2008.

[25] C. G. Sen and G. Cına, "Evaluation and Pre-allocation of Operators With Multiple Skills: A Combined Fuzzy AHP and Max–Min Approach," Expert Systems with Applications 37, vol. 37, pp. 2043–2053, 2010.

[26] S. Wang, S. Li, J. Zhou, Q. Li, and L. Kang, "Reliability Allocation for CNC Machine Based on Improved Fuzzy Analytic Hierarchy Process," AISS: Advances in Information Sciences and

Service Sciences, vol. 4, p. 320 ~ 327, 2012. [27] Y.-M. Wang, Y. Luo, and Z. Hua, "On the Extent Analysis Method for Fuzzy AHP and its

Applications," European Journal of Operational Research, vol. 186, pp. 735–747, 2008. [28] T.-S. Liou and M.-J. J. Wang, "Ranking Fuzzy Numbers with Integral Value " Fuzzy Sets and

Systems, vol. 50, pp. 247-255 1992.


89

Decision Making Model for Electronic Tender Evaluation (eTE) using Fuzzy AHP with Extent Analysis Method 1

Documents