Int. J. Services and Operations Management, Vol. 7, No. 4, 2010

465

An AHP framework of supplier evaluation with reference to high-value and critical items: a case study Debadyuti Das*
Faculty of Management Studies University of Delhi 110 007 Delhi, India E-mail: debadyuti_das@yahoo.co.in *Corresponding author

Deepak Barman
Faculty of Management Studies Banaras Hindu University 221 005 Varanasi, India E-mail: dbarmanbhu@rediffmail.com
Abstract: This study developed a simple two-stage decision framework for evaluating suppliers of high-value and critical items with reference to a heavy engineering organisation by employing the Analytic Hierarchy Process (AHP). The first stage involved examining the qualifying criteria of the items on quality, while the second stage was concerned with identifying all other relevant attributes, including quality concerning high-value and critical items applicable to the organisation under study, and with finding out the relative importance of the same. The attributes were organised in a two-tier hierarchy showing four major criteria in the first tier and 17 subcriteria in the second tier. This enabled us to employ AHP to find out the relative importance of each individual subcriterion through pair-wise comparison between all major criteria and subcriteria by eliciting opinions from three experts. The findings indicated the quality/reliability of the items to be the most important criterion, followed by price and then technological capability of the suppliers. A group of five suppliers was evaluated on all 17 subcriteria with regard to their performance and the overall performance of the suppliers was compared on the factor scores obtained by each supplier. Keywords: supplier evaluation; analytic hierarchy process; AHP; high-value and critical items. Reference to this paper should be made as follows: Das, D. and Barman, D. (2010) ‘An AHP framework of supplier evaluation with reference to high-value and critical items: a case study’, Int. J. Services and Operations Management, Vol. 7, No. 4, pp.465–482. Biographical notes: Debadyuti Das is a Reader in the Faculty of Management Studies, University of Delhi, India. His current research interests include the contemporary issues of supplier relationships in a supply chain environment, supply risk management, reverse logistics and supply chain issues in tourism.

Copyright © 2010 Inderscience Enterprises Ltd.

466

D. Das and D. Barman
His research papers have appeared in various journals such as Industrial Engineering Journal, International Journal of Tourism Policy, Journal of Services Research, etc. Deepak Barman is a Professor and Dean in the Faculty of Management Studies, Banaras Hindu University, Varnasi, India. He has more than 30 years of teaching and research experience in the field of statistics and operations research. His current research interests include the contemporary issues of operations and supply chain management. His research papers have been published in various journals such as Serbian Journal of Management (Serbia); Theoretical Biology (USA); Demography India (India); and Marketing Mastermind (India).

1

Introduction

Supplier evaluation has become a strategic issue in today’s competitive business environment in view of its overwhelming potential for improving the overall performance of a supply chain. This is because, in most of the manufacturing industries, the manufacturer’s cost of raw materials and component parts account for almost 70% of the product cost (Ghodsypour and O’Brien, 1998), and in high-technology firms, it can go up to 80% of the total product cost (Weber et al., 1991). Thus there exists tremendous potential for improving the cost effectiveness of the whole organisation and consequently the whole supply chain through proper supplier evaluation. In addition, supplier evaluation also helps in reducing the risk and uncertainty associated with the purchase, maximising the overall value to the organisation and building long-term relationships between buyers and suppliers (Monczka and Trecha, 1998). Studies have revealed that supplier evaluation and selection is a typical Multiple Criteria Decision Making (MCDM) problem involving criteria that can be both qualitative and quantitative. Over the years, researchers have developed a large number of supplier evaluation methods, such as the linear weighting method (De Boer et al.,1998), Analytic Hierarchy Process (AHP) (Schoenherr et al., 2008; Xia and Wu, 2007; Gaudenzi and Borghesi, 2006; Yang and Chen, 2006; Ounnar and Pujo, 2005), Analytic Network Process (ANP) (Wu et al., 2009; Sarkis and Talluri, 2002), Total Cost of Ownership (TCO) approaches (Monczka and Trecha, 1998), Mathematical Programming (Kokangul and Susuz, 2009; Wu et al., 2009; Demirtas and Ustun, 2008), Data Envelopment Analysis (DEA) (Narasimhan et al., 2001), Fuzzy approach (Chen et al., 2006) and Fuzzy AHP (Chan and Kumar, 2007). All these methods have their merits, demerits and uniqueness. The relative advantages and disadvantages of these methods are discussed in the literature review section in a very concise manner. For example, linear weighting is a very simple method, but weighs all the attributes equally. On the other hand, mathematical programming techniques cannot incorporate qualitative factors. AHP can incorporate both qualitative and quantitative factors, but cannot account for the risk and uncertainty that might be present in the decision environment. It is found that no single method can consistently outperform the other methods. The problem considered in the present paper involves procurement of a large number of high-value and critical items by a heavy engineering organisation from a number of suppliers which are located at dispersed geographical regions. The existing method being

An AHP framework of supplier evaluation

467

followed by the organisation does not guarantee the selection of good quality items. Essentially it involves two steps. The first step requires the suppliers to meet the minimum qualifying criteria pertaining to quality. Once the supplier gets clearance on adherence to the prescribed minimum standards on quality, it is no longer evaluated on quality in the subsequent stage. In the second step, maximum importance is attached to the price of items for the purchase of high-value and critical items. This leads to the selection of those suppliers which are concerned only with maintaining the qualifying standard of quality and sometimes rejection of those suppliers which provide very good quality at a slightly higher price. This practice has a significant bearing on the quality of the final products manufactured by the firm, and sometimes becomes visible through production of goods of second-grade quality, generation of customer complaints, etc. This particular phenomenon has motivated us to take it up as a research problem with a view to developing a suitable framework of supplier evaluation which will enable an organisation select suppliers of high-value and critical items with all desirable attributes. This requires the organisation to examine the quality of the items also in the second stage in order to see how far the suppliers have contributed towards improving the quality of the final products. The organisation under study requires an easily understandable and easy to implement method for capturing and evaluating the multiple attributes of suppliers with a view to finally shortlisting a few suppliers. The nature of the problem discussed above makes it an appropriate candidate for the application of AHP in the second stage of the problem. With this backdrop, the present study attempts to develop a simple two-stage framework for the evaluation of suppliers of high-value and critical items with reference to a heavy engineering organisation by employing AHP. This process, developed by Saaty (1980), is an established MCDM approach that employs a unique method of hierarchical structuring of a problem and subsequent ranking of alternative solutions by a paired comparison technique. The advantage of AHP lies in its robust and well-tested solution method and its capability of incorporating both quantitative and qualitative elements in evaluating alternatives. Section 2 provides a brief review of relevant literature concerning the factors influencing supplier evaluation and the methods thereof. In addition, it presents an overview of AHP along with its application in different contexts. Section 3 shows the case study in detail, wherein the hierarchical structure of criteria and subcriteria, the procedure for eliciting responses from experts, the supplier evaluation process, data analysis and the findings are discussed. The paper concludes with a brief summary, limitations and the applicability of the approach elsewhere.

2

Review of literature

The purchasing and supply management literature is replete with studies concerning the factors influencing the performance evaluation of suppliers, the methods adopted for supplier evaluation and selection, and so on. Research works pertaining to supplier evaluation can be systematically analysed under two headings: 1 2 supplier selection criteria supplier evaluation techniques.

468

D. Das and D. Barman

Accordingly the literature review is presented under these two headings. In addition, a third heading on AHP has also been included which describes, in brief, the methodology of AHP.

2.1 Supplier selection criteria
Dickson (1966) identified 23 supplier selection criteria through his survey of 170 purchasing managers. He concluded that cost, quality and delivery performance were the three most important criteria in supplier evaluation. The review of supplier selection criteria carried out by Weber et al. (1991) revealed that quality, cost and delivery criteria are still basic. Other important criteria, like the technological and financial capability of the suppliers, and the quality system adopted at suppliers’ premises, were not given due importance in traditional models. However, with the increasing demands of customers coupled with the complexity of the supply environment, the less tangible factors have also become important along with the most visible factors for supplier evaluation. Some of the less visible criteria deemed important for supplier evaluation are process flexibility/sensitivity of suppliers to buyers’ requirements (Kokangul and Susuz, 2009; Wu et al., 2009; Demirtas and Ustun, 2008; Chan and Kumar, 2007; Weber et al., 1991), support to design processes/support in new product development (Demirtas and Ustun, 2008; Toni and Nassimbeni, 2001), problem-solving capability (Wu et al., 2009; Liu and Hai, 2005) and ease of communication (Kokangul and Susuz, 2009; Demirtas and Ustun, 2008; Chan and Kumar, 2007). Table 1 systematically shows a number of factors in supplier evaluation considered important from the perspective of different researchers.
Table 1 Criteria Cost Criteria for the performance evaluation of suppliers References Wu et al. (2009), Chan and Kumar (2007), Xia and Wu (2007), Katsikeas et al. (2004), Sarkis and Talluri (2002), Narasimhan and Talluri (2001), Ghodsypour and O’Brien (1998), Choi and Hartley (1996), Swift (1995), Weber et al. (1991) Wu et al. (2009), Kokangul and Susuz (2009), Chan and Kumar (2007), Xia and Wu (2007), Chen et al. (2006), Liu and Hai (2005), Sarkis and Talluri (2002), Narasimhan and Talluri (2001), Ghodsypour and O’Brien (1998), Choi and Hartley (1996), Swift (1995), Weber et al. (1991) Chan and Kumar (2007), Katsikeas et al. (2004), Choi and Hartley (1996), Swift (1995) Wu et al. (2009), Kokangul and Susuz (2009), Demirtas and Ustun (2008), Chan and Kumar (2007), Xia and Wu (2007), Liu and Hai (2005), Katsikeas et al. (2004), Sarkis and Talluri (2002), Narasimhan and Talluri (2001), Ghodsypour and O’Brien (1998), Choi and Hartley (1996), Swift (1995), Weber et al. (1991) Sarkis and Talluri (2002), Narasimhan and Talluri (2001), Choi and Hartley (1996)

Quality/Reliability of the product

Technical/After-sales support Ability to meet delivery schedule/ Delivery lead time

Quality system at suppliers’ place/ Quality policy/Quality philosophy

An AHP framework of supplier evaluation
Table 1 Criteria Technological capability/Innovation capability/R&D capability Criteria for the performance evaluation of suppliers (continued) References

469

Chen et al. (2006), Liu and Hai (2005), Katsikeas et al. (2004), Sarkis and Talluri (2002), Narasimhan and Talluri (2001), Choi and Hartley (1996), Weber et al. (1991), Toni and Nassimbeni (2001) Swift (1995) Kokangul and Susuz (2009), Wu et al. (2009), Kokangul et al. (2009), Demirtas and Ustun (2008), Chan and Kumar (2007), Liu and Hai (2005), Ghodsypour and O’Brien (1998), Weber et al. (1991) Kokangul and Susuz (2009), Demirtas and Ustun (2008), Chan and Kumar (2007), Sarkis and Talluri (2002), Ghodsypour and O’Brien (1998), Choi and Hartley (1996) Katsikeas et al. (2004), Weber et al. (1991) Liu and Hai (2005), Katsikeas et al. (2004), Choi and Hartley (1996), Swift (1995), Weber et al. (1991) Kokangul and Susuz (2009), Chan and Kumar (2007), Chen et al. (2006), Liu and Hai (2005), Choi and Hartley (1996), Swift (1995), Weber et al. (1991) Weber et al. (1991) Chan and Kumar (2007), Swift (1995), Weber et al. (1991) Demirtas and Ustun (2008), Toni and Nassimbeni (2001)

Breadth of product line/Ability of a supplier to supply a number of items Process flexibility/Sensitivity of suppliers to buyers’ requirements/Responsiveness Willingness of suppliers to share information/Communication openness

Existence of IT/ Communication system Integrity of vendor/Honesty/ Vendor’s image Financial capability of the supplier

Business volume/Amount of past business Geographic proximity of suppliers Support to design processes/Support in new product development

2.2 Supplier evaluation techniques
As already mentioned in the introductory section of the paper, researchers have employed numerous methods of supplier evaluation, for example, linear weighting methods, AHP, ANP, TCO models, mathematical programming models, fuzzy approach, fuzzy AHP and DEA. Table 2 shows, in brief, the relative advantages and limitations of these techniques in a concise form. Table 2 reveals that the field of supplier evaluation is quite rich in respect of research studies and still continues to attract the attention of both researchers and practitioners due to its huge capacity to improve the whole supply chain performance. Despite the development of a number of sophisticated supplier evaluation techniques over the last two decades, AHP still has not lost its appeal and relevance due to its inherent simplicity and ease of use, as some recent works bear testimony to this fact. Schoenherr et al. (2008) employed AHP to assess supply chain risks within the context of an offshore sourcing decision. In this particular problem, AHP was used to evaluate the importance of each risk factor and to determine the best alternative. Gaudenzi and Borghesi (2006) used AHP to prioritise risk factors that may jeopardise the objective of perfect customer order fulfilment. Wu et al. (2006) developed a comprehensive list of supply risk factors

470

D. Das and D. Barman

for vendor selection via a structured review of literature and ranked them based on AHP. Ounnar and Pujo (2005) employed AHP for evaluating suppliers within a self-organised logistical network.
Table 2 Techniques Linear weighting methods Supplier evaluation techniques Approach followed It uses a simple scoring method which is heavily dependent on human judgement. It depicts the supplier evaluation criteria in a hierarchical fashion. The weights of the criteria and subcriteria are determined through pair-wise comparison of the same at each level by eliciting opinions from the experts. While AHP employs a unidirectional hierarchical relationship among criteria and subcriteria at different levels, ANP recognises that there exist interrelationships between the elements in different levels of the hierarchy and also between elements in the same level. Thus the decision elements are organised into networks of clusters and nodes. It evaluates the efficiency of suppliers by utilising a set of input measures and another set of output measures of the suppliers. It requires the business units to define and measure various cost components associated with a purchased item, which includes cost associated with late delivery, poor quality or other forms of supplier nonperformance. Advantages/Limitations The approach is very simple to understand and practise. However, it weighs all the criteria equally, which seldom happens in practice. It is a simple but robust and powerful tool of the MCDM problem, which can incorporate both qualitative as well as quantitative criteria. However, it may give rise to the problem of rank reversal if a new alternative is introduced or an old one is eliminated. The problem of rank reversal has been overcome in ANP. However, it can become quite complex as the number of evaluation criteria and interrelationships among the criteria increases. References De Boer et al. (1998)

Analytic Hierarchy Process (AHP)

Kokangul and Susuz (2009), Schoenherr et al. (2008), Xia and Wu (2007), Liu and Hai (2005), Ghodsypour and O’Brien (1998), Barbarosoglu and Yazgac (1997) Wu et al. (2009), Demirtas and Ustun (2008; 2009), Sarkis and Talluri (2002)

Analytic Network Process (ANP)

Data Envelopment Analysis (DEA) Total Cost of Ownership (TCO) approaches

DEA requires an enormous amount of data pertaining to both input and output measures. It is difficult for the purchasers to identify the components of cost and assign a quantitative value to each component.

Narasimhan et al. (2001)

Monczka and Trecha (1998)

An AHP framework of supplier evaluation
Table 2 Techniques Mathematical programming Supplier evaluation techniques (continued) Approach followed It helps the decision maker formulate the supplier evaluation problem in terms of an objective function that subsequently needs to be maximised or minimised by varying the values of the variables in the objective function. Decision making in a supplier selection problem may involve a high degree of fuzziness and uncertainty and as such quantifying human perceptions with the help of crisp number may not be very appropriate. FST is one of the effective tools to handle uncertainty and vagueness. The imprecision and the uncertainty associated with the supplier evaluation problem is addressed by the fuzzy set theory while eliciting opinions from the experts. AHP helps in converting the problem into a hierarchy of criteria and subcriteria. Advantages/Limitations Linear and nonlinear single or multi-objective programming models help in achieving goals like minimum cost, on-time delivery and adherence to quality. However, these models cannot account for the intangible factors associated with supplier performance. The FST approach involves a lengthy process of fuzzification and defuzzification of data. References

471

Kokangul and Susuz (2009), Wu et al. (2009), Demirtas and Ustun (2008; 2009), Xia and Wu (2007)

Fuzzy Set Theory (FST) approach

Chen et al. (2006), Sarkar and Mohapatra (2006)

Fuzzy AHP

It takes care of both quantitative as well as qualitative criteria. However, the process of fuzzy set theory in fuzzy AHP involves quite lengthy fuzzification and defuzzification of data.

Chan and Kumar (2007), Kahraman et al. (2003)

The nature of the present work, in essence, reflects an application-oriented case study which requires an easy-to-use and implementable method in practice. AHP, being very simple and easily understandable by the practitioners, satisfies the above criteria. Moreover, the limitation of AHP in respect of rank reversal is of no concern in the present study since the number of alternatives remained the same throughout the work. All the participants were properly guided in the approach. Moreover, the number of criteria was kept to a manageable level, thus reducing the number of pair-wise comparisons.

2.3 Analytic Hierarchy Process (AHP)
The strength of AHP lies in its ability to structure a complex and multi-attribute problem hierarchically and then to investigate each level of the hierarchy separately (Saaty, 1980). The AHP methodology considers a major objective and shows all the criteria in a hierarchical chart for the achievement of the objective. The criteria are again subdivided into subcriteria in the next level and the process is repeated till such divisions are found possible and necessary. AHP requires the experts to make a pair-wise comparison of

472

D. Das and D. Barman

the criteria (or subcriteria) at each level and to mention their opinion on a criterion’s importance compared to another criterion in terms of a linguistic scale (nine-point scale) suggested by Saaty (1990). This is explained in Table 3.
Table 3 Scale serial no. 1 2 3 4 5 6 7 8 9 Linguistic scale used by the experts and its conversion to Saaty’s scale Saaty’s scale value 9 7 5 3 1 1/3 1/5 1/7 1/9

Linguistic scale Very High (VH) Between Very High and High (VH and H) High (H) Between High and Medium (H and M) Exactly equal (EQ) Between Medium and Low (M and L) Low (L) Between Low and Very Low (L and VL) Very Low (VL)

Definition Extremely high importance of one element over another Very high importance of one element over another High importance of one element over another Moderately high importance of one element over another Equal importance of both the elements Moderately low importance of one element over another Low importance of one element over another Very low importance of one element over another Extremely low importance of one element over another

The pair-wise comparison matrix values for all the criteria and subcriteria are normalised using Saaty’s (1980) eigenvalue method and accordingly priority weights for criteria and subcriteria are calculated. Saaty proposed some computational methods for converting local priorities into global priorities as per the problem hierarchy. For the details of the AHP methodology, the readers are referred to the works of Saaty (1980; 1990). The application of AHP has been reported in diverse areas such as the evaluation of mobile phone alternatives based on desirable features (Isiklar and Buyukozkan, 2007), identification of the relative importance of factors to improve passenger security checks at airports (Yoo and Choi, 2006) and classification of materials and components for different material planning and control systems (Razmi et al., 2006). Further, Benyoucef and Canbolat (2007) attempted to integrate the concepts of fuzzy logic with AHP for selection of suppliers in hospitals in electronic procurement. Das and Mukherjee (2008) employed AHP to find out the relative importance of touristic attributes of a tourist destination.

3

Case study

The present study was carried out in a large engineering organisation located in Varanasi, India that is primarily engaged in the manufacture of diesel locomotives and employs thousands of workers at different levels. The locomotives manufactured by the company are almost fully indigenous in nature. The company has regularly exported its products to several countries such as Vietnam, Sri Lanka, Bangladesh, Tanzania, Malaysia

An AHP framework of supplier evaluation

473

and Myanmar. It is credited with obtaining Integrated Quality, Environment and Health and Safety Management System certifications based on ISO-9001, ISO-14001 and OHSAS 18001 respectively. It has to procure a large number of different categories of components, spare parts, etc., from numerous suppliers situated both in the domestic market as well as abroad. However, till now, the company has not followed any systematic method of supplier evaluation which incorporates all important attributes of a high-value and critical item.

3.1 Existing supplier evaluation process in the organisation under study
As mentioned in the introductory section of the paper, the organisation essentially follows a two-step procedure for evaluating suppliers of high-value and critical items. The outcome of the first step results in the items either being rejected due to their failure to meet the qualifying criteria or the same being passed on for further investigation on other attributes like price, delivery schedule, reliability and service response time. In the second step, maximum importance is attached to the price of the items and little to moderate importance is given to the other attributes. No further investigation or evaluation of quality is carried out at this stage. Thus the suppliers, on the one hand, are concerned with achieving only the prescribed minimum level of quality and, on the other, attempt to reduce the prices of the items as much as possible in order to be finally selected by the organisation. This procedure, in essence, does not necessarily result in good quality suppliers getting selected. The authors feel that the first step of the existing supplier evaluation process is absolutely fine. However, the second step requires thorough modification in view of the growing importance of a range of attributes of high-value and critical items.

3.2 Generation of supplier evaluation criteria relevant to the organisation under study
The issue concerning the second step of the supplier evaluation process was discussed with the professionals of the Stores and Purchase department and with a few Production and Maintenance personnel. The professionals feel that, in the changing scenario, selecting suppliers merely on the basis of price (after meeting the minimum level of quality criteria) will not instil enough confidence in the minds of the customers. Rather all relevant attributes including quality need to be properly taken into consideration for final evaluation and selection of suppliers in the second stage. Keeping this in mind, the details of the criteria pertaining to supplier evaluation were identified through a literature review and shown to the said professionals with a view to making the list of criteria relevant to the organisation, especially with reference to high-value and critical items. At this juncture, further discussions took place with the professionals and the list of criteria was further modified. Broadly four criteria were identified which were broken down into 17 subcriteria. Figure 1 reveals the hierarchical structure of criteria and subcriteria in detail. This structuring of the criteria and subcriteria will have a very important bearing on the selection of suppliers because the suppliers will have to pay appropriate attention to improving the performance on all these subcriteria in addition to meeting the prescribed minimum qualifying criteria on quality.

474
Figure 1

D. Das and D. Barman
Hierarchy of supplier evaluation criteria relevant to the organisation

Objective

Identification of supplier evaluation criteria with reference to high-value and critical items

Major criteria

C1

C2

C3

C4

C11
Subcriteria

C12

C13

C14

C15

C41

C42

C43

C21

C22

C23

C24

C31

C32

C33

C34

C35

Major criteria: C1: Technical factors; C2: Commercial factors; C3: Communication & miscellaneous support provided by suppliers; C4: Financial performance of suppliers Subcriteria: C11: Quality/Reliability; C12: Quality system at suppliers’ place; C13: Technological capability and innovation of suppliers; C14: Breadth of product line; C15: Technical support/After-sales support C21: Price; C22: Ability to meet delivery schedule; C23: Payment terms and conditions; C24: Geographic proximity C31: Service response time; C32: Sensitivity of suppliers to buyer’s requirement; C33: Willingness of suppliers to share vital information; C34: Support in value engineering C35: Support in developing new product specification; C41: Financial capability of suppliers; C42: Supplier’s volume of business with the buyer with respect to supplier’s turnover; C43: Supplier’s volume of business with respect to buyer’s total annual turnover

3.3 Determining the importance of evaluation criteria by eliciting experts’ opinions
Saaty’s pair-wise comparison method of the AHP is employed in the present study in the second stage of supplier evaluation to find out the relative importance of supplier evaluation criteria. The three top executives from the Stores and Purchase department were identified as experts in the present study with a view to securing responses from them. The purpose of the study and the pair-wise comparison method of the AHP were first thoroughly explained to all the three experts. The experts, while comparing the criteria, showed their preferences in terms of linguistic variables as suggested by Saaty, such as Very High and High. Pair-wise comparison data of the evaluation criteria are given in a linguistic scale (as per Table 3) in Tables 4 through 8. Each table contains the opinions of all three experts.

An AHP framework of supplier evaluation
Table 4 Relative importance of major criteria at the first level C1 (5,5,5) (6,6,6) (7,6,7) (8,7,7) C2 (4,4,4) (5,5,5) (6,6,6) (7,7,7) C3 (3,4,3) (4,4,4) (5,5,5) (6,6,6) C4 (2,5,3) (3,3,3) (4,4,4) (5,5,5)

475

Major criteria C1 C2 C3 C4 Table 5 Technical subcriteria C11 C12 C13 C14 C15 Table 6 Commercial subcriteria C21 C22 C23 C24 Table 7

Relative importance of subcriteria at the second level (technological factors) C11 (5,5,5) (8,7,8) (6,6,6) (7,7,7) (6,6,7) C12 (2,3,2) (5,5,5) (3,3,3) (4,4,4) (3,3,4) C13 (4,4,4) (7,7,7) (5,5,5) (6,6,7) (6,6,6) C14 (3,3,3) (6,6,6) (4,4,3) (5,5,5) (4,4,4) C15 (4,4,3) (7,7,6) (4,4,4) (6,6,6) (5,5,5)

Relative importance of subcriteria at the second level (commercial factors) C21 (5,5,5) (6,6,6) (7,8,7) (7,8,8) C22 (4,4,4) (5,5,5) (6,7,6) (6,7,7) C23 (3,2,3) (4,3,4) (5,5,5) (6,6,6) C24 (3,2,2) (4,3,3) (4,4,4) (5,5,5)

Relative importance of subcriteria at the second level (communication and support) C31 (5,5,5) (5,6,5) (6,6,6) (7,7,8) (8,7,8) C32 (5,4,5) (5,5,5) (6,6,6) (7,7,7) (7,7,8) C33 (4,4,4) (4,4,4) (5,5,5) (6,6,7) (7,6,7) C34 (3,3,2) (3,3,3) (4,4,3) (5,5,5) (6,6,6) C35 (2,3,2) (3,3,2) (3,4,3) (4,4,4) (5,5,5)

Communication subcriteria C31 C32 C33 C34 C35 Table 8

Relative importance of subcriteria at the second level (financial performance) C41 (5,5,5) (4,3,4) (6,6,7) C42 (6,7,6) (5,5,5) (7,8,8) C43 (4,4,3) (3,2,2) (5,5,5)

Financial subcriteria C41 C42 C43

476

D. Das and D. Barman

For illustration purposes, let us consider Table 8. While comparing C41 with C42, the three experts found the relative importance of C41 over C42 to be ‘between medium and low’, ‘low’ and ‘between medium and low’ respectively. In terms of the scale serial number as explained in Table 3, the corresponding values are 6, 7 and 6 respectively. These are shown in the C41-to-C42 matrix cell as (6,7,6).

3.4 Verifying the consistency of opinions of the individual experts
Data of relative importance on the above-mentioned evaluation criteria were converted to Saaty’s scale using the conversion table given in Table 3. The eigenvalue (λmax), consistency index (IC) and consistency ratio (RC) were calculated to check the consistency of the individual experts using Saaty’s consistency rule. To exemplify the procedure, let us take the relative importance data of the financial performance subcriteria as depicted in Table 8. The data are converted to Saaty’s scale as shown in Tables 9(a) through 9(c). The eigenvalue, consistency index and consistency ratio are also mentioned along with each of these tables. It may be mentioned here that a consistency ratio of 0.10 or less is considered acceptable (Saaty, 1990).
Table 9
(a) Expert 1 C41 C41 C42 C43 1 3 1/3
C

Relative importance of financial performance converted to Saaty’s scale
(b) Expert 2 C42 1/3 1 1/5
C

(c) Expert 3 C43 3 7 1
C

C43 3 5 1 C41 C42 C43

C41 1 5 1/3
C

C42 1/5 1 1/7

C41 C41 C42 C43 1 3 1/5
C

C42 1/3 1 1/7
C

C43 5 7 1

λmax = 3.038; I = .019; R = .033

λmax = 3.065; I = .033; R = .056

λmax = 3.065; I = .032; R = .056

To illustrate the conversion of the scale serial number assigned by the experts into its corresponding Saaty’s scale value, let us consider Table 9(c). While comparing C41 with C42, expert 3 assigned the relative importance of C41 over C42 to ‘between medium and low’. In terms of the scale serial number, the corresponding value is 6 and in terms of Saaty’s scale, the same value reduces to 1/3 as demonstrated in Table 3. Likewise all the values expressed in terms of scale serial number in Tables 4 to 8 were converted into the corresponding Saaty’s scale value. The tables further reveal that the opinions of the experts with regard to the financial performance of suppliers (C41, C42 and C43) are highly consistent, as in all the cases, the values of the consistency ratio are well below 0.10. The three experts are also consistent in giving their opinions for the major evaluation criteria at the first level as well as for the subcriteria at the second level, as shown in the summary in Table 10.

Table 10

Major evaluation criteria (first level) IC RC

Subcriteria (technical factors, second level)

Subcriteria (commercial factors, second level)

Subcriteria (communication and support, second level)

Subcriteria (financial factors, second level)

Consistency of individual experts

λmax
0.039 0.066 0.067 0.074 5.321 0.080 0.072 4.118 0.039 0.043 0.074 5.359 0.089 0.080 4.237 0.079 0.019 5.362 5.256 0.044 5.281 0.070 0.063 4.204 0.068 0.075 5.20 0.050 0.090 0.064

λmax

IC

RC

λmax

IC

RC

λmax

IC

RC 0.044 0.080 0.057

λmax
3.038 3.065 3.065

IC 0.019 0.033 0.032

RC 0.033 0.056 0.056

An AHP framework of supplier evaluation

Expert 1

4.118

Expert 2

4.199

Expert 3

4.201

477

478

D. Das and D. Barman

3.5 Determination of weights and composite weights of the supplier evaluation criteria
The individual opinions of the experts were pooled together after converting the same into Saaty’s scale in order to obtain the group opinion values of the relative importance of the supplier evaluation criteria. This was carried out by taking the geometric mean of the converted scale values. This helped in resolving the lack of consensus among the individual experts in respect of their individual opinions. The group opinion values of the relative importance (weights) of evaluation criteria at the first and second levels are shown in Tables 11(a) through 11(e). These are level-wise weights (local priorities).
Table 11 (a) Group opinion values of the relative importance of evaluation criteria at the first level; (b–e) Group opinion values of the relative importance of evaluation criteria at the second level
(a) Major criteria Weights C1 0.5275 (b) Subcriteria Weights C11 .4548 C12 .0465 (d) Subcriteria Weights C31 .3880 C32 .3243 C33 .1631 C34 .0780 C35 .0466 Subcriteria Weights C13 .2559 C14 .0863 C15 .1565 Subcriteria Weights C21 .5567 C2 0.2757 C3 0.1328 (c) C22 .2628 (e) C41 .2437 C42 .6692 C43 .0871 C23 .1182 C24 .0623 C4 0.0640

Composite weights (global priorities) are computed by the AHP technique and are shown for all 17 criteria in Table 12.
Table 12
Subcriteria Composite weights

Composite weights of supplier evaluation criteria
C11 C12 C13 C14 C15 C21 C22 C23 C24 C31 C32 C33 C34 C35 C41 C42 C43

.239 .0245 .135 .045 .082 .153 .072 .032 .017 .051 .043 .021 .010 .0062 .015 .043 .005

For explanation purposes, the composite weight of C11 is determined by multiplying the weight of C1 at the first level with that of C11 at the second level. It is evident from Table 12 that the experts have assigned the highest importance to the quality/reliability criterion, followed by price and then the technological capability of the suppliers. The relative importance of the remaining criteria is also clearly shown in Table 12.

3.6 Supplier evaluation process
In order to demonstrate the applicability of the supplier evaluation method in practice, five established suppliers of high-value and critical items were considered. These suppliers have already met the prescribed qualifying criteria relating to quality and therefore they have been directly passed on to the second step of the supplier evaluation

An AHP framework of supplier evaluation

479

process. The same three experts were requested to assess the performance of these suppliers on the above 17 subcriteria on a five-point rating scale through consensus. Score 1 indicated a very poor performance, score 5 very good performance, while the intermediate values implied a level of performance ranging from poor to good. Achieving a consensus among the experts on the performance of suppliers pertaining to all 17 subcriteria was a very time-consuming exercise. We had to resort to two rounds of interviews with the experts. Subsequently the individual score obtained by a supplier on a subcriterion was multiplied by its corresponding composite weight, which enabled us to find out the factor score of a supplier on a single subcriterion. Likewise, the factor scores of all the suppliers were determined on all the remaining subcriteria. Thereafter, the factor scores obtained by a single supplier on all the 17 subcriteria were summed up in order to determine the total factor score of each supplier. Finally the factor scores secured by all the five suppliers were compared. The supplier securing the highest total factor score was ranked 1, the second highest ranked 2 and so on. The entire exercise and results are shown in Table 13.
Table 13 Evaluation of suppliers
Performance rating of the suppliers S1 5 5 5 4 5 4 4 5 3 5 5 4 5 5 5 5 5 S2 4 4 4 4 4 3 5 4 4 5 3 4 3 4 5 3 4 S3 3 4 4 4 4 5 5 5 5 3 2 3 2 3 3 2 3 S4 4 4 5 3 3 4 4 4 4 5 4 5 3 4 4 3 4 S5 5 5 4 4 4 4 5 4 4 5 5 4 5 5 5 4 5 S1 1.195 0.1225 0.675 0.180 0.410 0.612 0.288 0.160 0.051 0.255 0.215 0.084 0.050 0.031 0.075 0.215 0.025 4.6435 1 Factor scores of the suppliers S2 0.956 0.098 0.540 0.180 0.328 0.459 0.360 0.128 0.068 0.255 0.129 0.084 0.030 0.0248 0.075 0.129 0.020 3.8638 4 S3 0.717 0.098 0.540 0.180 0.328 0.765 0.360 0.160 0.085 0.153 0.086 0.063 0.020 0.0186 0.045 0.086 0.015 3.7196 5 S4 0.956 0.098 0.675 0.135 0.246 0.612 0.288 0.128 0.068 0.255 0.172 0.105 0.030 0.0248 0.060 0.129 0.020 4.0018 3 S5 1.195 0.1225 0.540 0.180 0.328 0.612 0.360 0.128 0.068 0.255 0.215 0.084 0.050 0.031 0.075 0.172 0.025 4.4405 2

Evaluation criteria C11 C12 C13 C14 C15 C21 C22 C23 C24 C31 C32 C33 C34 C35 C41 C42 C43

Importance weight 0.239 0.0245 0.135 0.045 0.082 0.153 0.072 0.032 0.017 0.051 0.043 0.021 0.010 0.0062 0.015 0.043 0.005

Total factor scores secured by the suppliers Overall rank of the suppliers

480

D. Das and D. Barman

4

Conclusion

The present study has developed a simple two-stage decision framework of the supplier evaluation process focusing on high-value and critical items by employing AHP with reference to a heavy engineering organisation. In addition, the study has also demonstrated the applicability of the supplier evaluation process by considering five established suppliers of high-value and critical items. The study has revealed that the supplier evaluation process essentially involves two steps: the first step examines the qualifying criteria on quality and the second step investigates the other essential attributes, including quality in detail. These attributes or supplier evaluation criteria of high-value and critical items specific to the organisation under study were identified by conducting a literature review and by eliciting opinions from the experts. Subsequently the relative importance of the criteria was found by seeking opinions from three experts through the pair-wise comparison technique of AHP. These criteria along with their relative importance gave valuable insights to the Purchase managers. A group of five established suppliers was assessed with regard to their performance on 17 subcriteria on a five-point Likert scale, which enabled us to determine the factor scores of all the suppliers on all subcriteria. Finally the performance of the suppliers was compared on the basis of the total factor scores secured by the suppliers. This would enable the managers to select a particular supplier or a group of suppliers. The study suffers from the limitation of eliciting opinions from only three experts while carrying out the pair-wise comparison of supplier evaluation criteria. One more expert could have been approached for the same purpose. This probably would have produced a more realistic picture with regard to the relative importance of the supplier evaluation criteria. Nevertheless it adds to the existing body of supply management literature by demonstrating the applicability of the framework in a real-life situation. The same kind of study can be replicated elsewhere with suitable modifications/adaptations.

Acknowledgement
The authors would like to thank the anonymous reviewers for their insightful comments on improving the content and structure of the paper.

References
Barbarosoglu, G. and Yazgac, T. (1997) ‘An application of the analytic hierarchy process to the supplier selection problem’, Production and Inventory Management Journal, Vol. 38, 1st quarter, pp.14–21. Benyoucef, M. and Canbolat, M. (2007) ‘Fuzzy AHP-based supplier selection in e-procurement’, Int. J. Services and Operations Management, Vol. 3, No. 2, pp.172–192. Chan, F.T.S. and Kumar, N. (2007) ‘Global supplier development considering risk factors using fuzzy extended AHP-based approach’, Omega, Vol. 35, pp.417–431. Chen, C.T., Lin, C.T. and Huang, S.F. (2006) ‘A fuzzy approach for supplier evaluation and selection in supply chain management’, International Journal of Production Economics, Vol. 102, pp.289–301.

An AHP framework of supplier evaluation

481

Choi, T.Y. and Hartley, J.L. (1996) ‘An evaluation of supplier selection practices across supply chain’, Journal of Operations Management, Vol. 14, pp.333–343. Das, D. and Mukherjee, K. (2008) ‘Development of an AHP-QFD framework for designing a tourism product’, Int. J. Services and Operations Management, Vol. 4, No. 3, pp.321–344. De Boer, L., Van der Wagen, A. and Telgen, J. (1998) ‘A review of methods supporting supplier selection’, European Journal of Purchasing and Supply Management, Vol. 4, Nos. 2–3, pp.109–118. Demirtas, E.A. and Ustun, O. (2008) ‘An integrated multi-objective decision making process for supplier selection and order allocation’, Omega, Vol. 36, pp.76–90. Demirtas, E.A. and Ustun, O. (2009) ‘Analytic network process and multi-period goal programming integration in purchasing decisions’, Computers & Industrial Engineering, Vol. 56, pp.677–690. Dickson, G.W. (1966) ‘An analysis of vendor selection systems and decisions’, Journal of Purchasing, Vol. 2, No. 1, pp.257–267. Gaudenzi, B. and Borghesi, A. (2006) ‘Managing risks in the supply chain using the AHP method’, International Journal of Logistics Management, Vol. 17, No. 1, pp.114–136. Ghodsypour, S.H. and O’Brien, C. (1998) ‘A decision support system for supplier selection using an integrated analytic hierarchy process and linear programming’, International Journal of Production Economics, Vols. 56–57, pp.199–212. Isiklar, G. and Buyukozkan, G. (2007) ‘Using a multi-criteria decision making approach to evaluate mobile phone alternatives’, Computer Standards & Interfaces, Vol. 29, No. 2, pp.265–274. Kahraman, C., Cebeci, U. and Ulukan, Z. (2003) ‘Multi-criteria supplier selection using fuzzy AHP’, Logistics Information Management, Vol. 16, No. 6, pp.382–394. Katsikeas, C.S., Paparoidamis, N.G. and Katsikeas, E. (2004) ‘Supplier source selection criteria: the impact of supplier performance on distributor performance’, Industrial Marketing Management, Vol. 33, pp.755–764. Kokangul, A. and Susuz, Z. (2009) ‘Integrated analytical hierarchy process and mathematical programming to supplier selection problem with quantity discount’, Applied Mathematical Modelling, Vol. 33, pp.1417–1429. Liu, F-H.F. and Hai, H.L. (2005) ‘The voting analytic hierarchy process method for selecting supplier’, International Journal of Production Economics, Vol. 97, pp.308–317. Monczka, R.M. and Trecha, S.J. (1998) ‘Cost-based supplier performance evaluation’, Journal of Purchasing and Materials Management, Vol. 24, No. 2, pp.2–7. Narasimhan, R., Talluri, S. and Mendez, D. (2001) ‘Supplier evaluation and rationalization via data envelopment analysis: an empirical examination’, The Journal of Supply Chain Management, Summer, pp.28–37. Ounnar, F. and Pujo, P. (2005) ‘Evaluating suppliers within a self-organized logistical network’, The International Journal of Logistics Management, Vol. 16, No. 1, pp.159–172. Razmi, J., Zairi, M. and Gunasekaran, A. (2006) ‘The application of analytic hierarchy process in classification of material planning and control systems’, Int. J. Services and Operations Management, Vol. 2, No. 4, pp.352–366. Saaty, T.L. (1980) The Analytic Hierarchy Process, New York: McGraw Hill. Saaty, T.L. (1990) Multi-Criteria Decision Making – The Analytic Hierarchy Process, USA: RWS Publications. Sarkar, A. and Mohapatra, P.K.J. (2006) ‘Evaluation of supplier capability and performance: a method for supply base reduction’, Journal of Purchasing & Supply Management, Vol. 12, pp.148–163. Sarkis, J. and Talluri, S. (2002) ‘A model for strategic supplier selection’, The Journal of Supply Chain Management, Vol. 38, pp.18–28.

482

D. Das and D. Barman

Schoenherr, T., Tummala, V.M.R. and Harrison, T.P. (2008) ‘Assessing supply chain risks with the analytic hierarchy process: providing decision support for the offshoring decision by a US manufacturing company’, Journal of Purchasing & Supply Management, Vol. 14, pp.100–111. Swift, C.O. (1995) ‘Preferences for single sourcing and supplier selection criteria’, Journal of Business Research, Vol. 32, pp.105–111. Toni, A.D. and Nassimbeni, G. (2001) ‘A method for the evaluation of suppliers’ co-design effort’, International Journal of Production Economics, Vol. 72, pp.169–180. Weber, C.A., Current, J.R. and Benton, W.C. (1991) ‘Vendor selection criteria and methods’, European Journal of Operational Research, Vol. 50, pp.2–18. Wu, T., Blackhurst, J. and Chidambaram, V. (2006) ‘A model for inbound supply risk analysis’, Computers in Industry, Vol. 57, No. 4, pp.350–365. Wu, W-Y., Sukoco, B.M., Li, C-Y. and Chen, S.H. (2009) ‘An integrated multi-objective decision-making process for supplier selection with bundling problem’, Expert Systems with Applications, Vol. 36, pp.2327–2337. Xia, W. and Wu, Z. (2007) ‘Supplier selection with multiple criteria in volume discount environments’, Omega, Vol. 35, pp.494–504. Yang, C-C. and Chen, B-S (2006) ‘Supplier selection using combined analytical hierarchy process and grey relational analysis’, Journal of Manufacturing Technology Management, Vol. 17, No. 7, pp.926–941. Yoo, K.E. and Choi, Y.C. (2006) ‘Analytic hierarchy process approach for identifying relative importance of factors to improve passenger security checks at airports’, Journal of Air Transport Management, Vol. 12, pp.135–142.