Using interpretive structural modeling and fuzzy analytical process to identify and prioritize the interactive barriers of e-commerce implementation

Accepted Manuscript Title: Using interpretive structural modeling and fuzzy analytical process to identify and prioritize the interactive barriers of e-commerce implementation Author: Changiz Valmohammadi Shahrbanoo Dashti PII: DOI: Reference:

S0378-7206(15)00104-4 http://dx.doi.org/doi:10.1016/j.im.2015.09.006 INFMAN 2843

To appear in:

INFMAN

Received date: Revised date: Accepted date:

17-10-2014 26-8-2015 21-9-2015

Please cite this article as: C. Valmohammadi, S. Dashti, Using interpretive structural modeling and fuzzy analytical process to identify and prioritize the interactive barriers of e-commerce implementation, Information and Management (2015), http://dx.doi.org/10.1016/j.im.2015.09.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

cr

ip t

Using interpretive structural modeling and fuzzy analytical process to identify and prioritize the interactive barriers of e-commerce implementation

us

Changiz Valmohammadi, Ph.D, Assistant Professor

South Tehran Branch- Islamic Azad University, Tehran, Iran

an

Faculty of Management and Accounting

Department of Information Technology Management

M

Shariati Ave.

d

[email protected]

te

Tel: +98-21-22895783

Ac ce p

Fax: +98-21- 22894243

Shahrbanoo Dashti

MSCs. Graduate of Industrial Engineering

South Tehran Branch- Islamic Azad University, Tehran, Iran

1 Page 1 of 38

Using interpretive structural modeling and fuzzy analytical process to identify and prioritize the interactive barriers of e-commerce implementation Abstract The main purpose of this study is to present a novel and useful application of the specific

ip t

analytic technique to indicate the interactions and calculate the ranking of the barriers of electronic commerce (EC) in Iran Khodro industrial group, an Iranian leading automotive

cr

company using the combination of two techniques i.e. interpretive structural modeling (ISM) and fuzzy analytical network process (FANP). Based on an in-depth review of the relevant literature

us

and interview with managers and experts of the company, thirteen barriers and challenges to the implementation of e-commerce were determined and categorized into the four main factors,

an

namely technical, organizational, individual, and environmental. In the next step ISM technique is applied to construct a structural graph and identify inherent interactions among these barriers. Then FANP is used to quantify the relationships and weigh the significance of these barriers. The

M

results obtained from the proposed model reveal that “lack of awareness regarding the benefits and nature of electronic commerce” is the most important barrier to the implementation of e-

d

commerce. Such a modeling approach can be of great value for companies to prioritize their implementation of EC.

te

efforts and resources on removing the most important barriers and challenges towards successful

Ac ce p

Keywords: Electronic commerce (EC), Barriers and challenges, Interpretive Structural Modeling (ISM), Fuzzy Analytical Network Process (FANP)

1. Introduction:

Electronic commerce can be considered as the interaction among one company and another companies or customers (Shamsafar & Sharbafazari, 2008). E-commerce is the process of buying, selling, or exchanging products, services, and information via computer networks, including the internet (Ma & Wei, 2012; Thulani et al., 2010). However, e-commerce is not restricted to purchasing and selling products over internet but also it contains all supporting activities during the business process. E-commerce can be a source that improves domestic economic and rapid globalization of production, and development of available technology (Sheth 2 Page 2 of 38

& Sharma, 2005). Considering highly complex business and intensely competitive environment, and variability of customers (Lombardi et al., 2013), e-commerce is well-accepted in the developed world and is playing a vital role in economic development ( Mohanna et al., 2011) that can lead to more effective marketing actions, efficient processes, higher levels of customer

ip t

satisfaction, and higher returns on investments (Lombardi et al., 2013). Investment in the implementation of e-commerce within the organization can increase the efficiency and reduce

cr

the costs. Studies show that adoption of e-commerce leads to 21 to 70 percent saving on the costs of various activities. Statistics published on the top 500 companies around the world indicate that

us

34 percent in 1995 and nearly 80 percent in 1996 have applied this method to promote their products. In 2006, the value of on-line transactions was estimated at 12.8 trillion dollars

an

(Khanzade & khani, 2008). Iran with a population of more than 70 million and the GDP of

about $270 billion is the second largest economy in the Middle East. With a relatively strong economic growth (about 4.8 percent) and diverse industries, Iran offers a great potential in e-

M

commerce in the future (Yasin et al., 2014). Also, as Al-Somali et al. (2015) note Ecommerce adoption in developing countries is more complex due to challenges such as

d

insufficient regulatory environments and inadequate infrastructure; and so there is a

te

particular need to study less developed countries. According to the forecasts, the implementation of e-commerce in Iran will approximately cost

Ac ce p

210 billion dollars during third development plan period. Estimates show that these costs have high economic justification and are challenged due to e-commerce implementation (Khanzade & khani, 2008). Ever-increasing growth of e-commerce in developed countries means that business policy and strategy should be fundamentally reconsidered in developing countries (Moghaddasi, 2007). Considering the newness of e-commerce field in Iran, this country has been facing the various barriers and challenges in e-commerce implementation. For example Valmohammadi (2012) in his study points out to barriers encountered by Iranian companies including poor IT infrastructure and low speed of internet in this country. Numerous researchers have extensively studied and identified challenges and barriers to implementation of e-commerce (see table 1). However, to the best knowledge of the authors no research have studied and analyzed the interactions between these barriers. Therefore, the currents study aims to fill this gap in the literature by offering a comprehensive model involving dimensions of barriers and their relationships in EC domain. Owing to the above discussion and 3 Page 3 of 38

the importance of EC successful implementation for Iran, the main purpose of this study is to present a comprehensive novel model to analyze interaction among barriers of electronic commerce and calculate the ranking of the barriers in Iran Khodro industrial group, an Iranian leading automotive company using the combination of two techniques i.e. interpretive structural

ip t

modeling (ISM) and fuzzy analytical network process (FANP). Accordingly, we believe that it would be useful to investigate how individual barriers identified in the literature and prioritized

cr

by the experts would influence EC implementation.

This study can particularly serve as a guideline to the senior management team of the survey

us

organization and policy maker of the country in general to do necessary measures to remove or at least reduce these inhibitors and provide suitable ground for successful implementation of EC in the country’s companies based on the importance and priority of the barriers and challenges

an

identified in this study. More importantly, as the novel method proposed in this study is applicable to other contexts, firms can use this method to solve their problems in other areas.

M

The remainder of this paper is organized as follows; literature review of barriers to e-commerce implementation and identifying the most important barriers is presented in section

d

2. Methodology of the study which covers ISM and FANP techniques is presented in section 3. Section 4 presents the results obtained from the application of this hybrid approach to decision

te

making in the case study. Finally, Section 5 wraps up the paper by conclusion and

Ac ce p

recommendations for future research studies. 2.

Literature review of barriers to EC implementation

Many studies have been performed on the challenges and barriers facing e-commerce in developed countries. For instance, in 2000, Commerce Net have identified ten major barriers to e-commerce implementation in USA including: security, reliability and risk, lack of trained staff, lack of business models, cultural issues, lack of public infrastructures of organizations, fraud, slow speed of internet, and legal issues. Ihlstrum et al. (2003) classified barriers to e-commerce implementation as internal or external barriers to the organization. Internal factors include lack of organizational knowledge and awareness on e-commerce and organizational resource limitations. However, external factors mainly include technical concerns, external stakeholders, and support and maintenance. Rao et al. (2003) in their study presented multi-stage development model of e-commerce and investigated facility factors and barriers within each of these stages. This multi-stage model can illustrate gradual evolution of e-commerce during different 4 Page 4 of 38

development stages. In addition, it will help companies to provide a road map for their improvement. Javalgy & Ramsey (2001) investigated factors affecting the growth of ecommerce. As we know, e-commerce is one of the widely-used applications of the internet and its growth depends highly on Information and Communication Technology (ICT), social,

ip t

cultural, commercial, legal and governmental infrastructures. This research indicates that the lack of each aforesaid infrastructure can be a significant barrier to the implementation of e-commerce.

cr

Those infrastructures are essential in terms of supporting the growth of e-commerce and utilizing its strategic advantages. Salman (2004) identified key hurdles to implementing e-commerce and

us

information technology in developing countries with specific focus on three crucial vectors of an organization i.e. people, process, and technology, and classified main challenges of e-commerce implementation such as human condition, political aspects and environmental issues.

an

Jones et al., (2004) recognized the important barriers to e-commerce adoption considering by both internet users and businesses through a series of interviews in the South China region.

M

These barriers include lack of a nationwide credit card system and high internet access costs, concerns about government policy, security and privacy. MacGregor & Vrazalic (2005) studied

d

477 Australian and Swedish small to-medium sizes enterprises (SMEs). In this research, barriers and hindrances to e-commerce are classified in a basic model. Barriers identified by these

te

researchers include: high cost of e-commerce implementation, high complexity of e-commerce implementation, long term of capital return rate in e-commerce, organizational resistance to the

Ac ce p

new technology because of fear among employees, unsuitability of e-commerce to the way the SME does business, unsuitability of e-commerce to the products/services offered by the SME, lack of awareness about business opportunities/benefits that e-commerce can provide, concerns about security of e-commerce, lack of experience on e-commerce implementation among customers, providers and business partners, lack of reliance on external consultants to provide necessary expertise, and lack of e-commerce standards. Keshtri (2007) used three categories of feedback systems–economic, sociopolitical to provide a simple model of e-commerce barriers in the developing world. This paper attempts to present mechanisms in order to indentify and remove barriers and problem areas of EC implementation in developing countries. Lawrence & Tar (2010) examined the barriers to e-commerce adoption in developing countries. Results showed that the lack basic infrastructural, socio-economic and the lack of government national ICT strategies are the major barriers in the adoption and growth of e-commerce , and in order to 5 Page 5 of 38

achieve the adoption and diffusion of EC in developing countries, cultural issues is one of the most essential factors. Based on the above mentioned literature and for the purpose of summarizing and identifying research scope, a set of most important barriers and challenges of e-commerce were

ip t

identified based on the view of the company's managers and experts which is shown in Table 1.

2.1.

cr

A brief description of each of them is presented as follows; High cost of EC implementation

us

Many researchers have mentioned that Internet technologies are too expensive to implement (See table 1). For instance, MacGregor and Vrazalic (2005) by literature suggest that internet technologies are too expensive to implement. Cost of connection and hardware and cost of

an

maintenance of electronic application, reduces success of e-commerce implementation. In addition, e-commerce needs professional skills which are expensive either employ internally or

2.2.

M

outsource (Marsom & Shahbazirad, 2011).

Organizational resistance to change

d

Actually successful implementation of electronic commerce requires significant changes in the

te

organization . One of the biggest problems that organizations have to deal with is how to handle the changes related to e-business. Usually changes process in adopting a new technology, leads

Ac ce p

to widespread resistance and likely resentment in employees (Hanafizade and Rezaii, 2012).

2.3.

Complexity of EC implementation

Implementation of e-commerce is too complex. The technical and organizational complexity in the process of implementing in some organizations, can lead to refusing the application of ecommerce ( Hajkarimi, et al., 2009).

2.4.

Lack of technical infrastructure

The majority of developing countries are not ready for e-commerce, because of the lack of network infrastructure especially among individual users and entrepreneurs. Among the most pressing infrastructure limitations are access to technology (computers, connectivity, and gateway to internet), limited bandwidth, which reduces the capacity to handle audio and graphic 6 Page 6 of 38

data; poor telecommunications infrastructures (most of which are still analogue and can only transmit voice) and unreliable electricity supply (Lewrence &Tar, 2010).

2.5.

Lack of availability of specialists

ip t

Another challenge to implementation of EC is inadequate access to specialist. Generally, it can be expressed that skills shortages are threatening business growth prospects. Despite of scarcity

cr

of skilled specialists in e-commerce, implementation process will be facilitated if these experts

2.6.

us

be called from outside the organization (Hanafizade & Rezaii, 2012).

Lack of awareness regarding the benefits and nature of electronic-commerce

This issue by many researchers is known as one of the major barriers to successful

an

implementation of e-business, because customers or suppliers often have little awareness of the skills and commitment to e-commerce (Darch & Lucas, 2002). Therefore, lack of awareness

Lack of suitability of EC for products/services

d

2.7.

M

about nature and benefits of EC adoption, reduce the speed of the EC implementation process.

Inconsistency and lack of proportion between the products / services provided by the

te

organization and the concept of e-commerce, is an important parameter in the organization project to successful implementation of EC. On the basis of their business through the internet,

Ac ce p

corporates decide whether the products or services are suitable or not to offering on the internet (Moghaddasi, 2007). 2.8.

Lack of technical skills/ IT knowledge of personnel

Shortage of technical skills and IT knowledge of employees on information technology will lead to the prevention of adoption of e- commerce in organizations. In many organizations Information Technology sectors function unanimously, so it requires personnel of IT units be expert in this field (Shamsafar & Sharbafazari, 2008). It should be noted organizations require different expertise in this field because of multidisciplinary nature of e-commerce.

2.9.

Governmental policy, legal issues / standards

Legal and political environment of a country when there is no cooperation and coordination by the government can prevent the growth of e- commerce (Aljifri et.al, 2003). Governments need a 7 Page 7 of 38

legal structure and protective for doing business in the country boundaries. Otherwise, it will act as an inhibitor to the growth of e- commerce.

2.10.

Security concerns/lack of trust

ip t

Confidence and trust is an essential requirement for secure electronic trading. The question of trust is even more prominent in the virtual world than it is in the real world. The geographical

cr

separation of buyers and sellers, often coupled with a lack of real-time visual or oral interaction, creates a barrier to e-commerce adoption (Lewrence & Tar, 2010). Concerns regarding hacking

us

the accounts of customers during the online transaction decrease the likelihood of successful

2.11.

an

implementation of e-commerce in companies.

Lack of financial resources

EC implementation is normally an expensive project and requires a huge amount of money and

M

the budget allocation for e-commerce is very limited (Ihlstrum & Nilsson, 2001). The organization needs to have an adequate amount of funding, to back up unpredictable

d

circumstances. Such as “Running and maintenance more costly than expected” and ” the costs of

2.12.

te

human factors e.g. training “ (Jahanshahi et al., 2013).

Lack of top management support

Ac ce p

Top management is responsible for each and every activity at the all levels of the organizations (Singh & Kant, 2008). Top management should provide a clear direction or vision in order to help EC team members. EC implementation will succeed only if top-management is fully committed beyond public announcements. Top-management commitment is as an enabler, while lack of top-management commitment as a barrier too. Lack of top-management commitment may stem from various reasons like lack of experience and training, resistance to change, and hesitation in launching improvement programs (Talib et al., 2011).

2.13.

Lack of trust in supplier of technology

The fear of fraud by supplier of technology has commonly been cited as a significant barrier to B2C e-commerce (Vijayasarathy & Jones, 2000). Many organizations believe that buying

8 Page 8 of 38

technology for example, interface software, from software seller/supplier that also supports this type of software, increases the risk of losing security information. Table1: An overview of the barriers and challenges to e-commerce implementation

4

Lack of technical infrastructure

5

Lack of availability of specialists

6

Lack of awareness regarding the benefits and nature of e-commerce

7

Lack of suitability of EC for products/services Lack of technical skills/ IT knowledge of personnel

8

cr

Complexity of EC implementation

us

3

an

Organizational resistance to change

M

2

Rao et al., (2003); Lawrence and Tar, (2010); Darch and Lucas, (2002); Kaynak, et al. (2005); MacGregor and Vrazalic, (2005); Aljifri et al., (2003); Ihlstrom and Nilsson, (2001); Heung (2003), Hawking et al. (2004), Thulani et al., (2010); Kuzic et al., (2002). Ihlstrom et al., (2003); Rao et al. (2003); Lawrence and Tar, (2010); MacGregor and Vrazalic, (2005); Gunasekaran et al., (2009); Thulani et al., (2010). Ihlstrom et al., (2003); Kuzic et al., (2002); MacGregor and Vrazalic, (2005); Thulani et al. (2010). Darch and Lucas, (2002); Rao et al., (2003); Javalgi and Ramsey, (2001); Lawrence and Tar, (2010); Kaynak, et al., (2005); Aljifri et al., (2003); Ihlstrom and Nilsson, (2001); Heung (2003), Liao et al., (2003); Kshetri (2007); Moodley (2003). Ihlstrom et al., (2003); Javalgi and Ramsey, (2001); Darch and Lucas, (2002); Kuzic et al. (2002); Kaynak et al. (2005); Aljifri et al. (2003); Heung (2003); Mohanna et al. (2011); Liao et al. (2003); Thulani et al. (2010). Ihlstrom and Nilsson, (2001); Ihlstrom et al. (2003); Darch and Lucas, (2002); Lawrence and Tar, (2010); MacGregor and Vrazalic, (2005); Stockdale and Standing, (2004); Heung (2003), Kshetri (2007), Mohanna et al. (2011); Thulani et al. (2010); Moodley (2003). Ihlstrom et al., (2003); Rao et al. (2003); MacGregor and Vrazalic, (2005); Stockdale et al. (2004), Heung (2003); Thulani et al. (2010). Ihlstrom et al., (2003); Darch and Lucas, (2002); Heung (2003); Lawrence and Tar, (2010); MacGregor and Vrazalic, (2005); Aljifri et al. (2003); Kshetri (2007); Thulani et al. (2010); Kuzic et al. (2002). Rao et al., (2003); Javalgi and Ramsey, (2001); Kuzic et al., (2002); Kaynak, et al., (2005); MacGregor and Vrazalic, (2005); Aljifri et al., (2003); Ihlstrom and Nilsson, (2001); Stockdale and Standing, (2004); Hawking et al., (2004); Mohanna et al., (2011); Lawrence and Tar, (2010). Rao et al. (2003); Darch and Lucas, (2002); Lawrence and Tar, (2010); Kaynak, et al. (2005); MacGregor and Vrazalic, (2005); Aljifri et al., (2003); Heung (2003); Hawking et al., (2004); Liao et al., (2003); Mohanna et al., (2011); Thulani et al., (2010); Kuzic et al., (2002). Ihlstrom et al., (2003); Lawrence and Tar, (2010); Stockdale and Standing, (2004); Heung (2003); Gunasekaran et al., (2009); Kshetri, (2007); Moodley (2003); Thulani et al., (2010). Lawrence and Tar, (2010); Rao et al. (2003); Heung (2003); Hawking et al. (2004); Liao et al., (2003); Kshetri (2007); Thulani et al., (2010). Hajkarimi et al., (2009); MacGregor and Vrazalic, (2005); Hawking et al. (2004); Kuzic et al. (2002).

d

High cost of EC implementation

Ac ce p

1

References

ip t

E-Commerce barriers

te

No

9

Governmental policy, legal issues / standards

10

Security concerns/lack of trust

11

Lack of financial resources

12

Lack of top management support

13

Lack of trust in supplier of technology

9 Page 9 of 38

2.2. 1. Classification of the 13 selected barriers Following the identification of barriers and challenges, we have to classify them into a few set of principal criteria. Some have classified these barriers into the groups of education,

ip t

management time, economic issues, and technical knowledge and industrial infrastructure, education, governmental laws and regulations, and social and cultural issues. Keshtri (2007)

cr

classified the barriers into three groups as economic, social-political, and cognitive barriers. Ihlstrum et al. (2003) classified barriers to the implementation of e-commerce into internal or

us

external factors to the organization. Javalgi & Ramsey (2001) concentrated on four groups of infrastructure including commercial, social and cultural, governmental and legal, and ICT as the factors influencing the growth of e-commerce. Darch & Lucas (2002) based their classification

an

on the following factors: costs, lack of awareness of the nature of e-commerce, lack of skills essential for e-commerce, lack of awareness of how services are offered in e-commerce, and lack

M

of time. Salehnia (2008) classified barriers to the implementation of e-commerce into four groups: Organizational factor, relational factor, individual factor, technical factor. Kuzic et al. (2002) mention technology, management, and business challenges as the factors relevant to the

d

challenges facing e-commerce. Kaynak et al. (2005) group main limitations of e-commerce into

te

classes of cost limitations, range and number of users, and security concerns. MacGregor &Vrazalic (2005), depending on the type of complexity, mention barriers to the implementation

Ac ce p

of e-commerce as “very rough” and “unsuitable”. Aljifri at al. (2003) classified these barriers into technical and industrial infrastructure, education, governmental laws and regulations, and social and cultural issues”. Salman (2004) classifies main challenges of e-commerce implementation into human condition, political aspects and environmental issues. Based on the aforementioned classifications and considering the experts' opinions of the company, final classification of the main barriers and their relevant items is presented in table 2.

10 Page 10 of 38

Table2- classification of barriers and challenges to e-commerce implantation in Iran Khodro

Security concerns/lack of trust (C1-2) Complexity of EC implementation (C1_3)

Individual (C3)

Lack of financial resources (C2-1)

Organizational resistance to change (C3-1)

High cost of EC implementation (C2-2) Lack of awareness regarding nature/benefits of EC (C2-3)

Lack of top management support (C3-2) Lack of technical skills/ IT knowledge of personnel (C3-3)

Environmental (C4) Lack of availability specialists (C4-1) Lack of suitability of EC for products/services (C4-2)

ip t

Lack of technical infrastructure (C1-1)

Organizational (C2)

Governmental policy / standards (C4-3)

cr

Technical (C1)

3.

an

us

Lack of trust in supplier of technology (C4-4)

Methodology

Yin suggests that case studies are epistemologically justifiable when research questions focus on

M

reasons behind observed phenomena, when behavioral events are not controlled, and when the emphasis is on contemporary events (Kshetri, 2007). The case study research method was chosen

d

because of its ability to provide rich and detailed data (Yin, 2003). Valmohammadi & Servati

te

(2011) by literature mention that a case study methodology is best option when the objective is to build theory in the preliminary phases of a research study or to add new perspectives to previous

Ac ce p

research. The objective of a case study is not statistical generalisation, but an analytical one. This methodology tries to generalise from case to theory; it does not attempt to extrapolate facts from the sample to the population. Also, according to Kshetri (2007) researchers argue that case method is ‘‘appropriate and essential where theory exists but the environmental context is different or where cause and effect are in doubt or involve time lags.’’ This research satisfies these criteria. As Kshetri (2007) by literature argues e-commerce research is in an early stage of theoretical development, particularly in the developing countries. Therefore, the generalizability of research conducted in the developed countries is questionable in the context of developing world. Accordingly, based upon suggestion of Eisenhardt (1989) who argues that best practices models provide good candidates for a case research methodology, in this study a leading automotive car

11 Page 11 of 38

manufacturing company in Iran was selected which as previously noted, is called Iran Khodro Company. This paper proposes a method which combines the interpretive structural modeling (ISM) technique and the fuzzy analytic network process (FANP) procedures to deal with the

ip t

prioritization of barriers to EC adoption in Iran Khodro Co. Figure 1 shows the proposed

cr

methodology and main steps to obtain research objectives.

us

Identify and select main barriers, sub barriers by interview with experts, and literature review

M

an

Determine Contextual relationship

te

Check the consistency

Ac ce p

Rejection

d

Construct Pair-Wise comparison matrices

Admission

Defuzzify matrices using Converting Fuzzy data into Crisp Scores (CFCS) method and calculate weights

Construct unweighted, weighted and limited super matrix to find EC barrier final weight

Figure1: Flow diagram of the research

3.1.

ISM approach 12 Page 12 of 38

ISM was first presented by Warfield. He proposed that theory of ISM is based on discrete mathematics, graph theory, social sciences, group decision- making, and computer assistance (Warfield, 1974) ; and Malone, is the second one who conducted brief review of the ISM (Eswarlal et al., 2011). It is generally felt that individuals or groups encounter difficulties in

ip t

dealing with complex issues or systems. The complexity of the issues or systems is due to the presence of a large number of elements and interactions among these elements. The presence of

cr

directly or indirectly related elements complicates the structure of the system which may or may not be articulated in a clear fashion. It becomes difficult to deal with such a system in which

us

structure is not clearly defined. Hence, it necessitates the development of a methodology which aids in identifying a structure within a system. Interpretive structural modeling (ISM) is such a methodology (Attri et al., 2013). ISM is a powerful qualitative tool and suitable modeling

an

technique for analyzing the influence one element (Wang et al., 2008), and for developing insights into collective understandings of these relationship and their levels. ISM allows

M

researchers and managers to gain a deeper understanding of the relationship among key issues (Saxena et al., 1992). It can be stated the method is interpretive as the judgment of the group

d

decides whether and how the enablers are related. It is structural as on the basis of relationship, an overall structure is extracted from the complex set of items. And it is modeling in that the

te

specific relationship and overall structure are portrayed in a diagram model (Sage, 1977). Main steps of the ISM procedure are summarized as follows; List criteria (sub-criteria) considered for the problem, and define each criterion (sub-

Ac ce p

-

criterion) as ei , i=1,2,3,...,n ( Lee & Wang, 2010). -

After identify the criteria (sub-criteria) in step 1, establish relation matrix which shows the relationship among the criteria (sub-criteria) (Lee & Wang, 2010). A relation matrix is made by the opinion of the experts (Eswarlal et al., 2011), this can be done by asking questions, for

example “Does the variable ei influence the variable ej?” If the answer is “Yes” then ij = 1,

13 Page 13 of 38

otherwise ij = 0 (Cheng et al., 2007). The general matrix of the relation matrix is presented

an

us

cr

ip t

as below :

M

MATRIX D:

matrix (huang et al., 2005).

d

*Where ei is the i th element in the system, ij denotes the relation between i th and j th elements, D is the relation

Ac ce p

(1) and (2).

te

- After constructing the relation matrix, the reachability matrix can be calculated using equations

M=D+I

(1)

M*= Mk = Mk+1 k>1

(2)

*where I iis the unit matrix, k denotes the powers, and M* is the final reachability matrix.

Then, the reachability set is calculated and the priority is set based on equations (3),(4), as follows;

A (t i) = { t j | m’ ij = 1}

(3)

R (t i) = { t j | m’ ij = 1}

(4)

-

Where mij denotes the value of the ith row and the jth column. Then, from equation (5), the levels and relationships between the elements can be determined and the structure of the

14 Page 14 of 38

elements’ relationships can also be expressed using the graph (Cheng et al., 2007; shahbandarzade & Ghorbanpour, 2011). R (ti) ∩ A (ti) = R (ti)

Fuzzy set theory

ip t

3.2.

(5)

Fuzzy set theory was introduced by Zadeh (1965) to deal with uncertainty due to the

cr

inaccuracy and ambiguity, resulting from human language, human judgment, evaluation, and decisions. A fuzzy set A is a subset of a universe of discourse X, which is a set of ordered pairs

us

and is characterized by the membership function  A (x ) representing the mapping  A ( x )  [0,1] . The function value of  A (x ) for the fuzzy set A is called the membership value of x in A , which

an

represents the degree of truth that x is an element of the fuzzy set A . It is assumed

M

that  A ( x )  [0,1] , where  A ( x )  1 reveals that x completely belongs to A , while  A ( x )  0 indicates that x does not belong to the fuzzy set A where  A (x ) is the membership function and X

te

d

= {x} represents a collection of elements x.

Ac ce p

Triangular fuzzy number is defined by the triplet as (l, m, u); where l ≤m ≤u. the parameters l, m,

and u respectively, denote the smallest possible value, the most promising value , and the largest

possible value that describe a fuzzy event (sung, 2001; shahabandarzade & Ghorbanpour, 2011). The membership function of a triangular fuzzy number is shown in equation (6).

15 Page 15 of 38

(6)

ip t

Triangular fuzzy numbers are used as the membership function, which is illustrated in figure 2: A

0 m

r

A

an

l

us

cr

1

Figure 2: A membership function of the triangular fuzzy number

CFCS Defuzzification method

M

3.2.1.

The defuzzification method which is used in this paper is presented by Opricovic &Tzeng (2003). This method called CFCS method. If A k  ( l ijk , m ijk , rijk ) indicates the fuzzy assessments

d

ij

Where,

Ac ce p

Step 1: Normalization

te

of the kth evaluator. The CSCF method steps are as follow;

(7) (8) (9) (10)

Step 2: Computing lower (Ls) and upper (Us) normalized value: (11) (12)

Step 3: Computing total normalized crisp value: 16 Page 16 of 38

(13)

ip t

Step 4: Computing crisp value:

(14)

Fuzzy ANP

cr

3.3.

Many decision-making problems cannot be considered hierarchically because they have

us

interaction in various levels. The ANP allows for complex interrelationships among decision levels and attributes. The ANP feedback approach replaces hierarchies with networks

an

(Valmohammadi, 2010). Figure 3 shows the structural difference between hierarchy and network. Indeed, the elements within the hierarchy are often interdependent. The computation of local weights in ANP is exactly the same as analytic hierarchy process (AHP) method pair

M

wise comparisons among elements that need to be constructed. The result of computations or weights in ANP approach forms a supermatrix. By using initial supermatrix, it is possible to

d

derive the weights of priorities.

te

Figure 4 is a generalized supermatrix that is also introduced by Saaty (1996). A supermatrix is actually a blocked matrix, where each matrix block represents a relationship

Ac ce p

between two groups of nodes (clusters) in a network (Lee & Kim, 2000). Figure 3 also illustrates the structure and relative supermatrix in a network. (a) A hierarchy Goal

(b) A network

W21

Factors

Sub factors

W32

Goal

W21

Factors

W22

Sub factors

W32

Figure 3: (a) Hierarchy and (b) Network 17 Page 17 of 38

C1

e11 e12 



C2 e21e22  e2 m2

Cn en1en 2  enmn

W11

W12



W1n

W21

W22



W2 n

ip t

C1 e11e12  e1m1

e1m1

C2

 e2 m2





Cn

en1 en 2



Wn1









Wn 2

Wnn

enmn

us

W

cr

e11 e21

an

Figure 4: general form of the supermatrix

Because the real world is actually full of ambiguities or in one word is fuzzy, several research

M

studies have utilized advantages of the fuzzy theory in decision making techniques such as ANP. So, at this step of the research we employed FANP technique to determine the priority of the barriers and their relevant elements. Therefore, based on these barriers a questionnaire was

Ac ce p

te

demographics of the respondents.

d

designed and distributed among 17 managers and experts with linguistic scales. Table 3 shows

Table 3. Demographics of the respondents

Frequency

Level

99.4 0.6

Male Female Under graduate Graduate PhD Expert Master expert Manger 35-25 36-45 Above 46

58.5% 40% 1.5% 19.3% 21% 59.7% 15.7% 73.8% 10.5%

Demographics variables Gender Educational background

Position

Age

18 Page 18 of 38

18%

7-10

82%

10-15

Work experience(year)

It should be noted in order to confirm content validity of the questionnaire based on the work

ip t

done in the research of Jia & Reich (2013) three academics with specialty in EC, studied this questionnaire. Items that were either Ambiguous (fitting in more than one category),

cr

indeterminate (fitting in no category), or redundant (overlaps between items) were revised or eliminated.

us

The scales and related triangular fuzzy numbers are shown in Table 4. These scales are used in pairwise comparisons questionnaire to obtain the evaluation of relative importance of network

an

nodes. Definitions and descriptions in association with linguistic scales are shown in table 4. Table 4: The linguistic scale and corresponding triangular fuzzy numbers Linguistic scales

Scale of fuzzy number

1

Equally important Weakly important Essentially important Very strongly important Absolutely important Intermediate values ( x ) between two adjacent judgments

(1, 1, 1) (2, 3, 4) (4, 5, 6) (6, 7, 8) (7, 8, 9) (x- 1, x, x+1 ) (1/( x+ 1), 1/x, 1/ (x- 1))

9

    2,4,6,8

Ac ce p

1/ x

d

5 7

te

3

M

Fuzzy number

Pair wise comparison obtained from expert’s opinions in ANP should be consistent. Csutora (2001) presented a method to calculate consistency rate (C.R.) in fuzzy judgment matrixes. Based on the mentioned method if A  [ aij ] in which aij  (lij , mij , uij ) be a fuzzy

 is triangular fuzzy number judgment matrix, first construct A  [m ij ] , if A is consistent, then A consistent.

Based on Saaty (1980) in calculation of the C.R, Random index R.I. represents the average consistency index over numerous random entries of the same order reciprocal matrices. The value of R.I. depends on the value of n (the number of related criteria or alternative in decision matrices) which is shown in table 5.

19 Page 19 of 38

Table 5: random index used to compute consistency ratio (C.R.) n R.I.

1 0

2 0

3 0.52

4 0.89

5 1.11

6 1.25

7 1.35

8 1.4

9 1.45

10 1.49

C.I .   max  n  /  n  1

ip t

To calculate the C.R, the consistency index (C.I.) is formulated as follows: (15)

Where max the maximum eigenvalue and n is the dimension of matrix. Finally, the

cr

consistency ratio (C.R.) can be computed with the following Equation:

C.R.  CI

us

(16)

RI

If C .R .  0.1, the judgment is acceptable; otherwise, a new comparison matrix should be

an

established. After consistency test passed equations (7-14) of the CFCS method should be used for each evaluator (Csutora, 2001). Finally using equation (17) geometric average is applied to

M

integrate all consistent crisp values of k evaluators.

te

d

(17)

Ac ce p

(18)

*

*

A ij is an aggregated crisp judgment matrix and a ij is the aggregated crisp assessments

of criterion i and criterion j of k experts, i, j = 1, 2. . . n. and k is the number of experts. In the next step, the final weight of nodes from pair wise comparisons is calculated from equation (19)

4.

(19)

Applying the proposed methodology in case study

Based on the framework presented in figure 1, steps of data analysis are summarized as follows:

20 Page 20 of 38

Step 1: Identify main barriers and their relevant elements in EC As mentioned based on thorough research in literature and interview with the company's experts, 4 main barriers and 13 sub barriers are selected. These experts have more than 15 years of

ip t

experience in the area of business and well familiar with e-commerce (see table 3). Step 2: Determine Contextual Relationship matrix

After identifying and listing the barriers through literature review and experts' opinions, analysis

cr

is carried out. In this research, seventeen experts who were in charge of implementation of EC in Iran khodro identified contextual relationships among the barriers and their relevant elements

us

based on Matrix D. For example the relation matrix of main barriers is shown in table 6: Table 6: relation matrix of the main barriers C2 1

C2 C3 C4

1 1 1

1 1 1

C3 1

C4 0

1 1 1

0 0 1

an

C1 1

M

C1

d

For analyzing the barriers in developing ISM, the following four symbols have been used to

te

denote the direction of relationship between barriers (i and j): V - Barrier i will help to achieve barrier j;

Ac ce p

A - Barrier j will help to achieve barrier i;

X - Barriers i and j will help to achieve each other; and O - Barriers i and j are unrelated.

The SSIM (Structural self-interaction matrix) has been converted into a binary matrix, called the initial reachability matrix (see table 6) by substituting V, A, X and O by 1 and 0 as per given case. The substitution of 1s and 0s are as per the following rules: If the (i, j) entry in the SSIM is V, the (i, j) entry in the reachability matrix becomes 1 and the (j, i) entry becomes 0; If the (i, j) entry in the SSIM is A, the (i, j) entry in the reachability matrix becomes 0 and the (j, i) entry becomes 1; If the (i, j) entry in the SSIM is X, the (i, j) entry in the reachability matrix becomes 1 and the (j, i) entry also becomes 1; and 21 Page 21 of 38

If the (i, j) entry in the SSIM is O, the (i, j) entry in the reachability matrix becomes 0 and the (j, i) entry also becomes 0 (Singh & Kant, 2008). Reachability matrix was developed based on equations (1) and (2), as shown in table 7:

C1 C2 C3 C4

C2 1 1 1 1

C3 1 1 1 1

C4 0 0 0 1

cr

C1 1 1 1 1

ip t

Table 7: Reachability matrix of the main barriers

us

* Exceptionally, as can be seen, the values of the relation matrix and reachability matrix of the main barriers are the same. Reachability matrix presents clear relationship among all the main

an

barriers to EC. These relationships are show in figure 5: C1

C3

d

M

C2

C4

te

Figure 5: Relationship between the main barriers

Ac ce p

After constructing reachability matrix for main barriers, sub barriers and their relevant items, all the relationships of them, is depicted in figure 6. Note that network scheme is calculated based on the ISM results to evaluate the priority of e-commerce barriers.

22 Page 22 of 38

Goal

C2

C3

C4

ip t

C1

C4-1

C1-1

C2-1

us

cr

C3-1

C1-2

C4-2

C3-2

an

C2-2

C2-3

C3-3

C4-4

M

C1-3

C4-3

d

Figure 6: Relationship of the main barriers and their relevant items (the network scheme)

te

Step 3: Construct Pair-Wise comparison matrices In this step the pair-wise comparison matrices for the main barriers and their relevant

Ac ce p

items based on network relationships are gathered from a verbal questionnaire filled by seventeen experts of the company. And then converted to triangular fuzzy numbers based on the scales mentioned in table 4. For example the fuzzy comparison matrix for one evaluator of the main barriers with respect to the goal node, is shown in table 8. Table 8: The fuzzy pair wise comparisons with respect to Goal

Goal C1 C2 C3 C4

C1

C2

C3

C4

(1,1,1) (0.25,0.333,0.5) (2,3,4) (0.333,0.5,1) (2,3,4) (1,1,1) (4,5,6) (1,2,3) (0.25,0.333,0.5) (0.167,0.2,0.25) (1,1,1) (0.25,0.333,0.5) (1,2,3) (0.333,0.5,1) (2,3,4) (1,1,1)

23 Page 23 of 38

Step4: Check pair wise comparisons consistency Firstly, we should check the consistency. To check the consistency of judgments based on Csutora & Buckley (2001) table 9 which is the matrix obtained from the middle number of

ip t

triangular numbers in table 8, was constructed. Table 9: Middle number of triangular numbers in Table 3 C2

C3

C4

1 3 0.333 2

0.333 1 0.200 0.500

3 5 1 3

0.500 2 0.333 1

cr

C2 C3 C4

C1

us

Goal C1

an

In this step, we should check the consistency of judgments. To do this based on Csutora & Buckley (2001) recommendation, if the matrix including the middle number of the triangular fuzzy numbers is consistent, then the fuzzy matrix is consistent. Thus, to calculate the

M

consistency of fuzzy judgment matrix, based on table 8, equation (20) can be calculated as

(20)

Ac ce p

te

d

follows:

From equation (20), the max value, is equal to 4.0593 and using equations (15) and (16) and table 5 we have: C .I . 

max  n 4.0593  4 0.0198 =0.0198 R .I.=0.52 C . R .  =0.0222< 0.1  n 1 3 0.52 Because of C.R<0.1, we can conclude that the values of the table 8 are consistent, so the

judgment is acceptable. Similarly, this step is carried out for each 17 evaluators separately.

Step 5: Defuzzify matrices using CFCS method and calculate weights According to the results of the consistency test of the first evaluator’s judgment, comparisons of the four main barriers are passed. Then in order to aggregate the consistent judgment of all the 24 Page 24 of 38

experts and calculation of weights, we have to deffuzify the triangular fuzzy numbers of table 8 using CFCs method. Equations 7-14 are applied to deffuzification. Table 10 shows the result of defuzzification of the data of table 9.

ip t

Table 10: The final crisp value of one evaluator C1

C2

C3

C4

C1 C2 C3 C4

1 3.003 0.339 2.040

0.339 1 0.201 0.545

3.003 4.928 1 3.003

0.545 2.040 0.339 1

us

cr

Goal

To aggregate different opinions of decision makers, equation (17) is used. Then, the final weights will be calculated using geometric mean method (Saaty, 1980). Final weights and crisp

an

integrated values of fuzzy pair wise comparisons for the seventeen evaluators are shown in table 11.

C2

C3

C4

Weight

1 3.248815 0.366305 1.527189

0.310733 1 0.21863 0.429262

2.805091 4.516364 1 1.823417

0.736668 2.475607 0.56817 1

0.1864 0.5021 0.0959 0.2156

te

d

C1

Ac ce p

Goal C1 C2 C3 C4

M

Table 11: Final weights and Crisp integrated values for the main barriers with respect to Goal

Similarly, the final weights of all relevant items with respect to the main barriers, and final weights of interactions are calculated based on the opinions of the experts.

Step 6: Construct unweighted, weighted and limited super matrix to find EC barrier final weights

This step is the final step of proposed method. To evaluate EC barriers and finding priorities, first we must construct ANP unweighted supermatrix. The construction of ANP unweighted supermatrix is started by inserting the weights of the main barrier ratings which are shown in table10 under the goal node in supermatrix. Then, weights of the main barriers interactions and weights of the relevant items with respect to main barriers plus interaction weights of relevant items are inserted to supermatrix. Table 12 shows unweighted supermatrix. Weighted super matrix is obtained by multiplying the unweighted super matrix (Chen & Yang, 2011). To achieve 25 Page 25 of 38

weighted supermatrix, at first the columns of unweighted supermatrix must be normalized, it means, the sum of each column of the unweighted matrix should be equal to 1. The weighted supermatrix is shown in table 13. To obtain the limited supermatrix, weighted matrix is raised to the power of 2p +1; where p is an arbitrarily number until it reaches the convergence. The result

ip t

is provided in table 14 and is named the limited supermatrix. Finally, having the limited supermatrix, the global priorities of all relevant elements of the main barriers can be obtained.

cr

Table 12: The unweighted supermatrix C1

C2

C3

C4

Goal

C1-1

C1-2

C1-3

C2-1

C2-2

C2-3

C3-1

C3-2

C3-3

C4-1

C4-2

C4-3

C4-4

C1

0.37

0.41

0.28

0.12

0.19

0

0

0

0

0

0

0

C2

0.37

0.42

0.31

0.33

0.50

0

0

0

0

0

0

C3

0.26

0.18

0.40

0.24

0.096

0

0

0

0

0

0

0

0

0.30

0.22

0

0

0

0

0

0

0

0

0

0

0

0

C1-1

0.55

0

0

0

0

0.63

0

0.61

0.66

C1-2

0.28

0

0

0

0

0.23

1

0.39

0.34

C1-3

0.16

0

0

0

0

0.14

0

0

0

C2-1

0

0.40

0

0

0

0

0

0

C2-2

0

0.20

0

0

0

0

0

C2-3

0

0.39

0

0

0

0

C3-1

0

0

0.21

0

0

0

C3-2

0

0

0.68

0

0

C3-3

0

0

0.11

0

0

C4-1

0

0

0

0.123

0

C4-2

0

0

0

0.199

C4-3

0

0

0

C4-4

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0.62

0

0

0.48

0.46

1

0

0

0.75

0.38

0

0

0.52

0.53

0

0

1

0.25

0

1

1

0

0

0

0

0

0

0.67

0

0

0

1

0

0

0

0

0

1

0.33

1

0

1

0

0

0

0

0

0

1

0

0

0

1

0

0

1

1

1

0

0

0.37

1

0

0

0.22

0.66

0.21

0.49

0.47

0.34

0.39

0

d

te

0

0

0

us

0

0

0

0

an

0

M

C4 Goal

0

0

0.63

0

0.83

1

0.77

0

0.67

0

0

0.66

0.61

0.73

1

0

0

0.17

0

0

0.34

0.12

0.51

0.53

0

0

0.27

0

0

1

1

0

0

0

1

0

1

0

0

0

0

0

0

0

0

0

1

0

0

0

0

1

0.27

0.53

0.47

0

0

1

0

0

0

0

0

0

0

0

0

0.54

0

0.20

0

0

0

0

0

0

0

0

0

0

0

0

0.18

0.45

Ac ce p

0

26 Page 26 of 38

C1

C2

C3

C4

Goal

C1-1

C1-2

C1-3

C2-1

C2-2

C2-3

C3-1

C3-2

C3-3

C4-1

C4-2

C4-3

C4-4

C1

0.186

0.204

0.140

0.062

0.186

0

0

0

0

0

0

0

0

0

0

0

0

0

C2

0.185

0.207

0.157

0.167

0.502

0

0

0

0

0

0

0

0

0

0

0

0

0

C3

0.129

0.088

0.203

0.122

0.096

0

0

0

0

0

0

0

0

0

0

0

0

0

C4

0

0

0

0.148

0.216

0

0

0

0

0

0

0

0

Goal

0

0

0

0

0

0

0

0

0

0

0

0

0

C1-1

0.276

0

0

0

0

0.314

0

0.152

0.165

0.206

0

0

C1-2

0.142

0

0

0

0

0.117

0.250

0.098

0.085

0.127

0

0

ip t

Table 13: The weighted supermatrix

C1-3

0.082

0

0

0

0

0.070

0

0

0

0

0.250

0.333

C2-1

0

0.202

0

0

0

0

0

0

0.167

0

0

0

C2-2

0

0.101

0

0

0

0

0

0.25

0.083

0.333

0

C2-3

0

0.197

0

0

0

0

0.250

0

0

0

0.250

C3-1

0

0

0.106

0

0

0

0.092

0.250

0

0

0.056

C3-2

0

0

0.341

0

0

0

0.158

0

0.207

C3-3

0

0

0.053

0

0

0.5

0

0

0.043

C4-1

0

0

0

0.064

0

0

0

0.250

0.250

C4-2

0

0

0

0.099

0

0

0

0

C4-3

0

0

0

0.234

0

0

0.250

0

C4-4

0

0

0

0.102

0

0

0

0

0

0

0

0

0

0

0

0

0

0.121

0.155

0.250

0

0

0.251

0.129

0.178

0

0

0.333

0.083

0

0

0

0

0

0

0.25

0

0

0

0

0

0

0

us

cr

0

0

0

0

0

0

0

0.333

0.250

0.333

0

0

0.219

0.052

0.163

0.117

0.113

0.130

0

an

0.333

0.194

0

0.169

0

0

0.220

0.204

0.243

0

0

0.114

0.030

0.171

0.133

0

0

0.090

0

0

0

0.250

0

0.250

0

0

0

M

0.333

0

0.250

0

0

0

0

0.333

0.091

0.177

0

0

0

0

0

0

0

0

0.180

0

0

0

0

0

0

0

0

0

0.062

0.156

Ac ce p

te

d

0

27 Page 27 of 38

Table 14: The limited supermatrix C2

C3

C4

Goal

C1-1

C1-2

C1-3

C2-1

C2-2

C2-3

C3-1

C3-2

C3-3

C4-1

C4-2

C4-3

C4-4

C1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

C2

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

C3

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

C4

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

C1-1

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

0.118

C1-2

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

0.103

C1-3

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

C2-1

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

0.033

C2-2

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

0.079

C2-3

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

0.127

C3-1

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

0.097

C3-2

0.111

0.111

0.111

0.111

0.111

0.111

0.111

0.111

0.111

C3-3

0.102

0.102

0.102

0.102

0.102

0.102

0.102

0.102

0.102

C4-1

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

C4-2

0.052

0.052

0.052

0.052

0.052

0.052

0.052

0.052

0.052

C4-3

0.031

0.031

0.031

0.031

0.031

0.031

0.031

0.031

C4-4

0.002

0.002

0.002

0.002

0.002

0.002

0.002

0.002

us

cr

Goal

an

ip t

C1

0.111

0.111

0.111

0.111

0.111

0.111

0.111

0.111

0.102

0.102

0.102

0.102

0.102

0.102

0.102

0.102

0.102

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.072

0.052

0.052

0.052

0.052

0.052

0.052

0.052

0.052

0.052

0.031

0.031

0.031

0.031

0.031

0.031

0.031

0.031

0.031

0.031

0.002

0.002

0.002

0.002

0.002

0.002

0.002

0.002

0.002

0.002

d

M

0.111

te

As a result and based on table 14, the final weights of all thirteen barriers is obtained from limited supermatrix which are shown in table 15.

Ac ce p

Table 15: final weights and rankings of the barriers Name

Lack of technical infrastructure Security concerns/lack of trust Complexity of EC implementation Lack of financial resources High cost of EC implementation Lack of awareness regarding the nature and benefits of e-commerce Organizational resistance to change Lack of top management support Lack of technical skills/ IT knowledge of personnel Lack of availability of specialists Lack of suitability of EC for products/services Governmental policy, legal issues/standards Lack of trust in supplier of technology

Limited Super matrix weights

0.118 0.103 0.072 0.033 0.079 0.127 0.097 0.111 0.102 0.072 0.052 0.031 0.002

Final rank

2 4 9 11 7 1 6 3 5 8 10 12 13

28 Page 28 of 38

5. Conclusion and recommendation for further research Understanding the barriers to the implementation of e-commerce and the structural relationship between these barriers can lead into better recognition of complexity of system implementation and subsequently, result in the growth of e-commerce in all organizations

ip t

particularly, large industries such as car manufacturing. Despite extensive studies done on the classification and prioritization of the problems facing implementation and adoption of e-

cr

commerce, it should be mentioned that most of these research studies are just limited to the identification of barriers to e-commerce, so the current study attempted to fill this gap in the

us

literature by studying the interactions of barriers to EC implementation and prioritized these barriers using a novel model and useful application of the specific analytic technique.

an

Determination of interactions among elements of barriers along with determination of priorities and ranking of the barriers can be of great value for companies to prioritize their efforts and resources on removing the most important barriers and challenges towards successful

M

implementation of e-commerce. According to the result shown in table 14, “lack of awareness regarding the benefits and nature of e-commerce” with calculated weight of 0.127, is the most

d

important barrier to e-commerce implementation in Iran khodro Industrial group. In other words

te

respondents of this study judged "lack of awareness regarding the benefits and nature of ecommerce" as the most important barrier to e-commerce implementation in their company. This

Ac ce p

finding is consistent with the result of the Yasin et al., (2014) study upon the 50 Iranian manufacturing organizations where they found that lack of e-commerce knowledge among the survey organizations is one of the most important elements of “organizational and technological” factor which had been ranked the top barrier among the various factors of this research. These scholars mention that the determined factors in their study are, for the most part, very similar to those reported by US organizations. For instance in both cases (Iranian and US survey organizations) lacks of management support and lack of technological resources appears to be an important hindering element to effective e-commerce implementation. As can be seen from table 7 these two elements in our study have been ranked as the second and third most important barriers to EC implementation which is in accordance with Yasin el al. (2014) findings. Also, Jahanshahi et al. (2013) by literature mention that “Lack of knowledge and understanding of Ecommerce” is among the most important barriers to EC implementation.

29 Page 29 of 38

To overcome the most important prioritized barrier in this study Iran khodro should hold educational classes for all of its staff in various levels in order to familiarize them with benefits and nature of this advanced trade strategy and make effort to foster an organizational culture which supports the collaboration and build confidence and trust among employees to accept the

ip t

new technologies including IT and e-commerce. All in all, it could be concluded that lack of awareness about the benefits and nature of e-commerce results into the lack of suitability of e-

cr

commerce implementation which could subsequently lead to other severe problems such as lack of suitability of e-commerce to products/services. Ultimately this un- suitability will lead to lack

us

of top management support, organizational resistance, high costs and complexity of e-commerce implementation.

Although the researchers have no predetermined reasons to believe that the findings should not

an

replicate in other industries, it is worth testing whether these findings (priorities) are generalizable, particularly in the context of developing countries. Also, performing this study in

M

non-automotive industries and comparing the obtained results with the present study can clarify the effect of car manufacturing industry, if any, on the barriers to e-commerce implementation.

d

As the levels of development of various countries including cultural, social, technological and even political system vary which affects the level of employment of EC, it is recommended captured in this research.

te

future researchers to consider these factors that are specific to their countries, which have not

Ac ce p

The next alternative can be the repetition of this study by integrating Decision-making Trial and Evaluation Laboratory (DEMATEL) and ANP methods. This integrated method can be helpful in better understanding of functional differences of ISM and DEMATEL methods in terms of defining and recognizing interactions of barriers to e-commerce implementation. Using Total Interpretive Structural Modeling (TISM) method instead of ISM method can be effective at relations interpretation and in clarifying how these relations act. This method is a great step toward higher interpretability of ISM method, i.e. TISM method will highly clarify model interpretation and prevent different interpretations by different commentators (Sushil, 2012). The proposed model has not been statistically tested and validated. Thus, applying other methods such as Structural Equation Modeling (SEM), following the application of ISM method and obtaining conceptual model of interactions can be helpful for understanding the weight of

30 Page 30 of 38

internal impacts among barriers to e-commerce implementation and statistical validation of the

ip t

model (Talib et al., 2011).

Reference

cr

Attri, R., N. Dev, and V. Sharma. (2013) Interpretive Structural Modeling (ISM)

us

approach: An Overview. Research Journal of Management Sciences. 2(2): 3--8. Aljifri, H. A., Pons, A., and Collins, D. (2003). Global e-commerce: a framework for understanding and overcoming the trust barrier. Information Management &

an

Computer Security, 11(3), 130–138.

Al-Somali , S. A., Gholami , R., Clegg , B. (2015) "A stage-oriented model (SOM) for

M

e-commerce adoption: a study of Saudi Arabian organisations", Journal of Manufacturing Technology Management, 26( 1), 2 - 35

d

Chen, J. and Yang, Y. (2011). A fuzzy ANP-based approach to evaluate region

te

agricultural drought risk. Procedia Engineering, 23, 822 – 827.

Ac ce p

Cheng, Y. L., Chiu, A.S.F., Tseng, M.L. and Lin, Y.H. (2007). Evaluation of Worker Productivity Improvement Using ISM and FAHP. IEEE International Conference on Industrial Engineering and Engineering Management, 109–113.

Csutora, R. and Buckley, J. J. (2001). Fuzzy hierarchical analysis: the Lambda-Max method. Fuzzy Sets and Systems, 120(2), 181–195.

Darch, H. and Lucas, T. (2002),“ Training as an e-commerce enabler. Journal of Workplace Learning, 14(4), 148–155.

Eisenhardt, K. M.(1989). Building theories from case study research, Academy of Management Review 14, 532–550.

31 Page 31 of 38

Eswarlal, V., Dey, P.K. and Shankar, R. (2011). Enhanced Renewable Energy Adoption for Sustainable Development in India: Interpretive Structural Modeling Approach. Word Renewable Energy Congress 2011-Sweden, 351-358. Gunasekaran, A., McGaughey, R. E., Ngai, E. W. T., and

Rai, B. K. (2009). E-

ip t

Procurement adoption in the South coast SMEs. International Journal of Production Economics, 122(1), 161–175.

cr

Hajkarimi, A., Azizi, S. and Akhavan, M. (2009). Modeling factors affecting of ecommerce application in SMEs. Journal of Knowledge and Development, 16(27), 51-

us

78.

Hanafizade, P., Rezaii, M. (2012). Electronic commerce, Definitions, barriers, and

an

solutions. Terme Baran publications, 6th edition.

Huang, J.J., Tzeng, G.H. and Ong, C.H. (2005). Multidimensional data in

M

multidimensional scaling using the analytic network process. Pattern Recognition Letters 26, 755–767.

d

Hawking, P., Stein, A., Wyld, D. C. and Foster, S. (2004). E-procurement: is the ugly

te

duckling actually a swan down under? Asia Pacific Journal of Marketing and Logistics, 16(1), 3–26.

Ac ce p

Heung, V. C. S. (2003). Barriers to implementing e-commerce in the travel industry: a practical perspective. International Journal of Hospitality Management, 22(1), 111– 118.

Ihlstrom, C., Magnusson, M., Scupola, A. and Tuunainen, V. K. (2003). SME Barriers to electronic commerce adoption: Nothing Changes - Everything is New. In G. Gingrich (Ed.), Managing IT in Government, Business & Communities, 147-163.

Ihlstrom, C. and

Nilsson, M. (2001). Size Does Matter: SMEs Special barriers in

adopting e-business. Collecter, 171-178. Jahanshahi , A. A., Zhang , S. X. , Brem , A. (2013) "E-commerce for SMEs: empirical insights from three countries", Journal of Small Business and Enterprise Development, 20( 4), 849 – 865.

32 Page 32 of 38

Javalgi, R. and Ramsey, R. (2001). Strategic issues of e-commerce as an alternative global distribution system. International Marketing Review, 18(4), 376–391. Jia, R. and Reich, B. H. (2013), “IT service climate, antecedents and IT service quality

ip t

outcomes: Some initial evidence”, Journal of Strategic Information Systems, 22, 51–69. Jones, L., Tucker, D. and Chan, H (2004). E-Commerce barriers in south china: the

cr

broader perspective, International Journal of Electronic Business Management, 2(2), 77-84.

us

Kaynak, E., Tatoglu, E. and Kula, V. (2005). An analysis of the factors affecting the adoption of electronic commerce by SMEs: Evidence from an emerging market.

an

International Marketing Review, 22(6), 623–640.

Kshetri, N. (2007). Barriers to e-commerce and competitive business models in

M

developing countries: A case study. Electronic Commerce Research and Applications, 6(4), 443–452.

d

Khanzade, A. and Khani, N. (2008). "E-commerce", Babel Computer Science, 2th

te

edition.

Kuzic, J., Fisher, J., and Scollary, A. (2002). Electronic commerce benefits, challenges

Ac ce p

and success factors in the Australian banking and finance industry. In Proceedings of the Tenth European Conference on Information Systems (Wrycza S Ed.), Gdansk, Poland, 1607–1616.

Lombardi, S., Gorgoglione, M. and Panniello, U. (2013) “The effect of context on misclassification costs in e-commerce applications”, Expert Systems with Applications 40, 5219–5227.

Lawrence, J.E. and Tar, U.A. (2010). Barriers to ecommerce in developing countries. Information, Society and Justice, 3(1), 23-35. Lee, J. W. and Kim, S. H. (2000). Using analytic network process and goal programming for interdependent information system project selection. Computers & Operations Research, 27(4), 367–382.

33 Page 33 of 38

Lee, A. H. I., Wang, W. M. and Lin, T. Y. (2010)., “An evaluation framework for technology transfer of new equipment in high technology industry”, Technological Forecasting and Social Change, 77(1), 135–150. Liao, S. H., Cheng, C. H., Liao, W. B. and Chen, I. (2003). “A web-based architecture for

ip t

implementing electronic procurement in military organizations”. Technovation, 23(6), 521–532.

cr

Ma, B. and Wei, Q. (2012). Measuring the coverage and redundancy of information search services on e-commerce platforms. Electronic Commerce Research and

us

Applications 11, 560–56.

MacGregor, R. C. and Vrazalic, L. (2005). A basic model of electronic commerce

an

adoption barriers: A study of regional small businesses in Sweden and Australia. Journal of Small Business and Enterprise Development, 12(4), 510–527.

M

Masrom, M. and Shahbazirad, R. (2011). Enablers and barriers for success of ecommerce in Malaysian small and medium enterprises. Proceeding Business and

d

information, 8(1), 4062- 4072.

te

Moghaddasi, A. (2007). Strategies for implementation of e-commerce and challenge of EC in Iran. Journal of Economics and new Trade, 6, 22-56.

Ac ce p

Mohanna, S., Yaghoubi, N. and Vahidi, S. (2011). Limitations of E-commerce implementation in developing countries: Case study of Iran. American Journal of Scientific and Industrial Research, 2(2), 224–228.

Moodley, S. (2003). The challenge of e-business for the South African apparel sector. Technovation, 23(7), 557–570.

Opricovic, S. and Tzeng, G.-H. (2003). Defuzzification within a multi-criteria decision model. Int. J. Uncertain. Fuzziness Knowl.-Based Syst., 11(5), 635–652. Rao, S. S., Metts, G. and Monge, C. A. (2003). Electronic commerce development in small and medium sized enterprises: A stage model and its implications. Business Process Management Journal, 9(1), 11–32.

34 Page 34 of 38

Saaty, T. (1980). The analytic hierarchy process : planning, priority setting, resource allocation. New York; London: McGraw-Hill International Book Co. Saaty, T. (1996). Decision making with dependence and feedback: the analytic network process: the organization and prioritization of complexity (1st ed.). Pittsburgh PA:

ip t

RWS Publications.

Systems. McGraw-Hill, New York, NY, 91-164.

cr

Sage, A.P. (1977). Interpretive Structural Modelling: Methodology for Large-scale

Iran. Master thesis of Tehran university.

us

Salehnia, A. (2008). Understanding and evaluate barriers to e-commerce development in

an

Salman, A. (2004). Elusive challenges of e-change management in developing countries. Business Process Management Journal, 10(2), 140-157. Saxena, J.P., Sushil, and Vrat, P. (1992). Scenario building: a critical study of energy

M

conservation in the Indian cement industry. Technological Forecasting and Social Change, 41(2), 121-46.

d

Shahbandarzadeh, H. and Ghorbanpour, A. (2011). The applying ISM/FANP approach

te

for appropriate location selection of health centers. Iranian Journal Management Studies, 4(2), 5–28.

Ac ce p

Shamsafar, A. and Sharbafazari, F. (2008), “E-commerce is only solutions to implement Iran's progress, "Fourth Conference on Electronic Commerce, December 2008.

Sheth, J. N. and Sharma, A., (2005). “International e-marketing: Opportunities and issues”. International Marketing Review, 22(6), 611-622.

Singh, M.D. and Kant, R. 2008, "Knowledge management barriers: An interpretive structural modeling approach", International Journal of Management Science and Engineering Management, 3(2), 141-150. Stockdale, R. and Standing, C. (2004). Benefits and barriers of electronic marketplace participation: an SME perspective. Journal of Enterprise Information Management, 17(4), 301–311.

35 Page 35 of 38

Sung, W.C. (2001). application of Delphi method , a qualitative and quantitative analysis into healthcare management. Journal of Healthcare Management. 2(2), 11-19. Sushil. (2012). Interpreting the Interpretive Structural Model”, Global Journal of Flexible

ip t

Systems Management, 13(2), 87-106. Talib, F., Rahman, Z. and Qureshi, M. N. (2011). Analysis of interaction among the barriers to total quality management implementation using interpretive structural

cr

modeling approach. Benchmarking: An International Journal, 18(4), 563–587.

us

Thulani, D., Tofara, C. and Langton, R. (2010). Electronic commerce benefits and adoption barriers in small and medium enterprises in Gweru, Zimbabwe. Journal of

Valmohammadi,

Ch.

an

Internet Banking and Commerce, 15(1), 1–17. Servati,

A.(2011).

Performance

measurement

system

implementation using Balanced Scorecard and statistical methods. International

M

Journal of Productivity and Performance Management 60 (5), 493-511. Valmohammadi, Ch. (2012). Investing the perceptions of Iranian employees on

d

teleworking. Industrial and Commercial Training, 44(4), 236-241.

te

Valmohammadi, Ch. (2010), " Using Analytic Network Process in business Strategy Selection : a Case study", Australian Journal of Basic and Applied Sciences, 4(10),

Ac ce p

5205-5213.

Vijayasarathy, L.R. and Jones, J.M. (2000). Print and Internet Catalog Shopping: Assessing Attitudes and Intentions. Internet Research: Electronic Networking Applications and Policy, 10(3), 191-202.

Wang, G.H., Wang, Y. X. and Zhao, T. (2008). Analysis of interactions among the barriers to energy saving in China. Energy Policy, 36, 1879–1889.

Warfield, J.W. (1974). Developing interconnected matrices in structural modeling. IEEE Transcript on Systems, Men and Cybernetics, 4(1), 51−81. Yasin , M. M., Alavi , J., Czuchry , A., Shafieyoun , R. (2014). An exploratory investigation of factors shaping electronic commerce practices in Iran : Benchmarking

36 Page 36 of 38

the role of technology and culture. Benchmarking: An International Journal, 21(5), 775 – 791.

ip t

Yin, R. (2003). Case Study Research: Design and Methods (3 rd edn.). Thousand Oaks, CA: Sage.

Ac ce p

te

d

M

an

us

cr

Zadeh, L.A. (1965). Fuzzy sets. Information and control, 8(3), 338–353.

37 Page 37 of 38

ip t cr us an M d

Dr.Changiz Valmohammadi is an assistant professor and head of department of information

te

technology Management at Islamic Azad University-South Tehran Branch. His areas of interest lies at quality management, strategic management, Knowledge management, and operations

Ac ce p

management. For over18 years he has taught undergraduate, graduate and industry courses and carried out research in various aspects of industrial engineering and management. He has published research papers in journals such asInternational Journal of Production Economics, The TQM Journal, Journal of Productivity and Performance, Innovation: Management, Policy & Practice, Business Strategy Series, Industrial Engineering International, Journal of Enterprise Information Management, Benchmarking: An International Journal, just to name a few. He is a senior member of American Society for Quality (ASQ) and editorial board of Journals such as Journal of Asia Business Studies and Industrial and Commercial Training. Mrs. Shahrbanoo Dashti holds M.Sc. in Industrial Engineering with major of System and productivity from Islamic Azad University- South Tehran Branch

38 Page 38 of 38