Evaluating ASEAN E-Commerce Laws Using Fuzzy Multi-Criteria Decision Making

An evaluation of the ASEAN Electronic Commerce Acts (ECAs) is a specific primary step toward their harmonization. This evaluation is complex due to multiple criteria and uncertain information, especially in the case of electronic contracts. This study aims to evaluate and rank the ECAs of the ASEAN countries using a fuzzy multi-criteria decision making method. Seven criteria are determined for the evaluation of ASEAN ECAs. The Fuzzy Analytic Hierarchy Process (FAHP) is applied for weighting of the criteria as well as the evaluation of alternatives. We applied the judgment of 10 experts to fill out the FAHP comparison matrices. The experts’ opinions are fuzziefied and aggregated using the fuzzy set theory. We found the weight of criteria, rank of ECAs in each criterion and the final rank on overall comprehensiveness of the ASEAN ECAs. The results show that the FAHP address the multi criteria evaluation of the laws and regulations. The recognition of the Electronic Message with a weight of 0.216684 is the most important criterion in the evaluation of ECAs, and the Place of Dispatch with a weight of 0.0276018 is the least important. This study has the potential to harmonize the ECAs among ASEAN countries which in turn will increase the number of online transactions and reduce the number of legal disputes.


Introduction
The association of Southeast Asian Nations (ASEAN) includes ten countries with nine Electronic The legal analysis and evaluation of ASEAN ECAs are a strategic approach to identify the strengths and weaknesses of these Acts. As a practical reference, it will lead the member countries in future amendments of their ECA to be more harmonized. In this study, we purpose to evaluate and rank the ECAs of eight ASEAN countries with respect to electronic contracts. In the case of Lao People's Democratic Republic (LPDR), there is no official translation of its ECA and Cambodia lacks any ECA.
Currently, lawyers use the theoretical and general analytic method for evaluation and comparison of laws [1]. However, the evaluation of law is complex due to the consideration of a set of qualitative criteria. Moreover, the factors have considerably different importance in terms of decision-making. Multi Criteria Decision-making (MCDM) methods have frequently been applied in decision-making with multiple criteria [2].
The Analytic Hierarchy Process (AHP) is the mathematical based MCDM method. Fuzzy AHP is a more systematic method than other MCDM methods, and, significantly, it captures the human assessment of qualitative options. This is true because pairwise comparisons provide a flexible and realistic way to accommodate real-life data [3]. Moreover, it is the most applicable MCDM method for decision-making with a limited number (less than ten) of criteria and alternatives [3].
We determined a set of seven criteria by reviewing and analyzing the ASEAN ECAs. The pairwise comparison matrices have been constructed to compare the criteria and countries with each other.
A group of ten experts in the area of electronic commerce law were requested for data collection.
An easy reference based on the ASEAN ECAs, including the related provisions in each criterion along with the related provisions of the international instruments, were prepared for experts. The in depth interviews with experts, which is a qualitative strategy [4], was used in the determination of the evaluation criteria. We requested the experts to complete the questionnaire regarding the pairwise comparison of the criteria and alternatives.
A nine-point linguistic variable scale was used for the experts' responses in the questionnaire. The experts expressed their opinion regarding the qualitative criteria through uncertain linguistic variables. As the conventional AHP method cannot deal with uncertainty [5], the fuzzy set theory was applied to handle the uncertainty of the decision-makers [6]. Various studies have used the fuzzy technique to deal with the uncertainty in the AHP [7][8][9][10]. We employed the FAHP for the weighting of each criterion, the ranking of countries in each criterion and the final ranking concerning the overall comprehensiveness of ASEAN countries in respect of their ECA.
The remainder of this paper is organized as follows. The background of this research is provided in section 2; the general process and methodology is explained in section 3; section 4 provides the implementation of the FAHP for the weighting of criteria and evaluation of countries. In this section, the achieved results are discussed. Finally, section 5 concludes the paper.

Research Background
In August 1967, five founding countries -Indonesia, Thailand, Singapore, Malaysia and the Philippinesestablished ASEAN in Bangkok. Between 1984 and 1999, Brunei, Vietnam, Lao PDR, MrECA and Cambodia joined the association, and now ASEAN has ten members [11].
ASEAN with an area of 4,435,617 million km 2 has a population of approximately 625 million people and US$2.511 trillion of total international merchandise trade [12].
Accelerating the economic growth, social progress and cultural development in the region is the first clause of the aims and purposes of ASEAN [13].  [14].
The Blueprint asks member states to provide harmonized laws on electronic contracting, mutual recognition of digital signature and dispute resolution. Thus, the harmonization of the electronic commerce laws is in line with these initiatives to facilitate regional and international trade, economic integration and to minimize conflicts.

ASEAN e-Commerce Acts and Harmonization
The development in the areas of information and communication technology (ICT) have changed the methods of traditional paper based trade into electronic commerce (e-commerce), and, such advancements have affected the international and domestic socio-economic and legal schemes.
From the legal point of view, regional initiatives to harmonize the e-commerce laws are admirable, such as the Central America and Caribbean [15], or East African Community [16].
The extra-border feature of e-commerce and facilitating international trade, have led the universal and regional trends toward a universal law governing the online transactions. The harmonization of electronic commerce laws will highly affect the ASEAN economic integration mission and may increase the trust factor and reduce future legal disputes.
ASEAN is one of the pioneers to have initiated the e-commerce legal harmonization for its member states through international cooperation. The United Nations Conference on Trade and Development (UNCTAD) collaborated with the ASEAN through two projects since 2003 [17].
The electronic commerce legal harmonization project with the Australian Agency for International Development (AusAID) 2004-2009 resulted in considerable legal ratifications in the region [17].
The enactment of ECAs with an emphasis on harmonization was based on a Reference Framework for Electronic Commerce Legal Infrastructure provided by ASEAN in 2001.
This Framework itself was based on some partial electronic commerce laws of five ASEAN member states at that time. These partial laws were also prepared based on the United Nations  [18]. The member states have enacted their domestic ECA during the last decade in line with the ASEAN harmonization action plan and the said ASEAN collaborative projects. However, the legal analysis of these ECAs indicates that they are still far from an ideal harmonized legal regime. This legal disharmony will affect the ASEAN economic integration and create legal barriers for regional e-transactions.

Analytic Hierarchy Process
The AHP method is a mathematical based MCDM method that was proposed by Saaty [19]. It is based on the additive weighting method, in which a number of criteria are represented by their relative importance. The AHP can deal with decision-making that has multiple levels of criteria [20]. This method is suitable for the evaluation and ranking in MCDMs with less than ten criteria and alternatives [3]. Generally, the AHP is based on composing a hierarchy decision tree [21]. A hierarchy has three levels as with the overall goal of the problem at the top, multiple criteria that define alternatives in the middle, and decision alternatives at the bottom (Fig 1).
The AHP method can be employed for the weighting of criteria as well as for the evaluation of alternatives. It uses the same strategy to calculate criteria weights and alternative ranks. The considered strategy is to obtain an eigenvector as a vector of ranks and weights. The evaluation of criteria is based on pairwise comparisons of criteria. The experts express the relative importance of one criterion over another criterion. Finally, the calculation of the eigenvector draws out the weighting of the criteria and the ranking of alternatives from the pairwise comparison matrix.

Method
This study is an applied research because it undertakes to solve practical problems in e-commerce law. It comprises both qualitative and quantitative research. We use the qualitative approach in the determination of criteria for the evaluation ECAs. The comparative analysis of the ECAs is also a part of the qualitative method. The in-depth interviews and questionnaire is utilized for data collection while using related documents.
In the interviews, we discussed with the experts regarding the current e-commerce Acts in the ASEAN countries and the determination of criteria. Then we requested them to complete the questionnaire. The questionnaire has two parts, the first and second parts are related to the pairwise comparison of criteria and ASEAN ECAs, respectively. The quantitative strategies, such as the AHP method and Fuzzy techniques are used to prioritize the criteria and rank of the ASEAN ECAs.
We proposed a three-stage method for the evaluation of ASEAN ECAs based on multiple criteria ( Fig 2). The first stage is the determination of the critical factors, which is done through the literature review and analysis of the ASEAN ECAs, as well as consulting with experts. The literature review is used for a survey of all the existing secondary sources, which are interpreted as the characteristics and general principles of the electronic contracts. The second stage is weighting of the determined criteria. The weighting of criteria is conducted using the Fuzzy AHP method. In this stage, the required data for the pairwise comparison of criteria is collected from ten experts. The experts with expertise in the electronic commerce law and the contract law, were selected from the faculties of law in various Malaysian universities.

Alternative m
Due to the inconsistency of linguistic variables expressed by the experts, the Fuzzy set theory is used in fuzzification, aggregation of linguistic variables and defuzzification of fuzzy numbers.
Finally, in the third stage, we provide a structured evaluation of ASEAN ECAs through Fuzzy AHP. The pairwise comparison of ECAs in all criteria delivers a comprehensive assessment.
Calculation of the inconsistency index is used to check the constancy of collected data. The Fuzzy AHP produces the important measures of criteria in stage 2 and the preference measure of ASEAN ECAs in stage 3. We rank the countries in each criterion according to their related rates, which are obtained by Fuzzy AHP. The summation of the rates in all criteria and descending sort is used to obtain the final ranking of ASEAN ECAs.

Determination of criteria
We applied three approaches for the determination of criteria, i) the literature review of electronic commerce laws, ii) the legal analysis of the Acts, and iii) consulting with experts. The analysis of the legal instruments and review of secondary sources concerning e-commerce law, especially the e-contracts, indicates that the general principles of the law of contracts are applicable in e-contracts but may differ in application [22]. However, there are some procedural differences in the case of the online contracting arising from the means of formation. Therefore, the scope of e-commerce law is usually delimited to the consideration of the new features of online transactions [23][24][25][26][27].
These features constitute the general discussion in the area of electronic commerce laws. We

Determination of criteria
The review of general features of electronic commerce law, especially e-contracts and the legal analysis of the ECAs led to the provision of a primer set of ten important criteria (Fig 3). We presented the primer set of criteria to a group of ten experts who were also asked for comparisons.
The experts' opinions were used to select the correct criteria. Finally, seven criteria were

Time of Dispatch of an Electronic Message (TD)
The time of dispatch of an electronic message is the time when the originator sends the offer of a contract to the addressee. According to Article 15 (1) of the MLEC, unless otherwise agreed, the time of dispatch is when a data message enters an information system outside the control of the originator or his representative. Article 10 (1) of the UNCUEC considers the time of dispatch as the time when the electronic communication leaves the originators' information system. If it did not leave the information system, then the time when it was received.
The drafters of the UNCUEC tried to consider new technologies that emerged after the adoption of the MLEC. Hence, they adopted new rules concerning the time of dispatch to be more adaptive with new technologies and the Internet compared with the MLEC, which was mostly based on Electronic Data Interchange (EDI) [28]. It has to be mentioned that the time and place of dispatch and receipt play a significant role in determining the commencement of a contract and the governing law in the case of disputes.

Place of Dispatch of an Electronic Message (PD)
Under Article 10 (3) of the UNCUEC, the place of business is adopted to determine the place of dispatch of an electronic message. The convention provided detailed rules in Article 6. Article 15 (4) of the MLEC also clarifies that unless otherwise agreed, the place of business of the originator is considered as the place of dispatch of a data message.

Time of Receipt of an Electronic Message (TR)
Article 15 (2) of the MLEC clarifies that unless otherwise agreed, the time of receipt is when the data message enters the designated information system of the addressee. If sent to a non-designated information system, the time is when it is retrieved by the addressee. In case there is no determination by the addressee, then it is the time when it enters any information system of the addressee. However, the UNCITRAL changed the receipt rule in the later convention. Under Article 10 (2) of the UNCUEC, the time of receipt is when the electronic communication is capable of being retrieved at the addressee's designated electronic address. If sent to another electronic address, then it is when it becomes retrievable at that address by the addressee and they become aware of it.

Place of Receipt of an Electronic Message (PR)
Article 15 (4) of the MLEC, adopted the place of business as a criterion to determine the place of receipt unless otherwise agreed by the contracting parties. The place of legal domicile (habitual residence) has also been referred to as another criterion. As the location of communication systems may change, the UNCITRAL selected the place of business as an objective criteria [29]. The UNCUEC also adopted the place of business as a criterion for PR in Article 10 (3) and Article 6.

Electronic Agent (EA)
Article 12 of the UNCUEC on the use of automated message systems for contract formation provides that: "A contract formed by the interaction of an automated message system and a natural person, or by the interaction of automated message systems, shall not be denied validity or enforceability on the sole ground that no natural person reviewed or intervened in each of the individual actions carried out by the automated message systems or the resulting contract".

Electronic Message (EM)
The legal recognition of an electronic communication or data message is an important provisions embedded in almost all ECAs and the said international instruments. This means that a data message is regarded valid and enforceable like a written document. The said recognition will automatically indicate the legal effect and validity of the electronic contracts, e-documents and esignatures. The said international instruments have addressed this principle.

Electronic Signature (ES)
Article 2 (a) of the MLES defines electronic signature as: "Electronic signature" means data in electronic form in, affixed to or logically associated with, a data message, which may be used to identify the signatory in relation to the data message and to indicate the signatory's approval of the information contained in the data message".

Fuzzy AHP
The Fuzzy AHP is the fuzzy extension of AHP method that addresses the uncertainty of experts' opinions using the fuzzy set theory [5]. In pairwise comparisons, the experts express the relative importance of one criterion over another criterion by linguistic variables. The fuzzified pairwise comparison matrix is obtained from the fuzzification of linguistic variables. Let = [C i ] n i = 1, 2, … , n be the set of criteria. The results of the pairwise comparison can be summarized in an evaluation matrix as: (1) where = [cw ij ] n×n and cw ij is the importance of the criterion C i over criterion C j .
Through the following steps the weighting of criteria and ranking of countries using Fuzzy AHP are conduct: 1-Squaring pairwise comparison matrix and construct S as: 2-Summarization row elements of matrix S and construct ⃗⃗⃗⃗ = [cs i ] n where: 3-Normalization of vector ⃗⃗⃗⃗ to reach eigenvector ⃗⃗⃗⃗⃗ = [cn i ] n where: 4-Iteration of steps 1-3 and compare eigenvector in each iteration with previous one until differences between eigenvectors become much lower. Therefore, the last eigenvector is the priority vector.
The probability of high inconsistency is a problem of the AHP method. Therefore, to rely on an answer, the rate of inconsistency should be calculated. The principal eigenvalue λ max is used to calculate the ratio of inconsistency. This value is calculated from the summation of product between each element of the priority vector ⃗⃗⃗⃗⃗ and the sum of columns of the pairwise comparison matrix: The relation between CW entries determines the inconsistency index (II) as: When λ max = n , it means that the answers are completely consistent. The ratio of inconsistency is achieved by dividing II by the random index (RdI) as: The reliability of the result of AHP is strictly related to the ratio of inconsistency. If IR is more than 0.1 then the results cannot be reliable. This ratio of inconsistency can also evaluate the inconsistency of decision-makers' opinions in pairwise comparison [30].

Basic definitions of Fuzzy Sets
The Fuzzy set theory is an extension of the classical set theory proposed by Prof. Zadeh, which is utilized to defuzzify and computerize Linguistic or fuzzy variables [31]. The Triangular Fuzzy Number (TFN) is a simple and popular fuzzy number, which can be utilized to manage the ambiguity of the decision-makers' judgments in Fuzzy AHP. Here, some of the basic major definitions of fuzzy sets are provided [32][33][34][35].

Definition 1.
A fuzzy set Ã in a universe of discourse X is characterized by a MF µ Ã (x), which is associated with every element x in X a real number in the interval [0, 1]. The function value µ Ã (x) is termed the grade of membership of x in Ã.
Definition 3. Linguistic variables are variables with linguistic term values. The concept of a linguistic variable is very useful in dealing with the situations that are too complex or too illdefined to be reasonably described in conventional quantitative expressions [31,36].

Fuzzification and aggregation of linguistic variables
The experts' opinions are the main source for the evaluation of criteria and ECAs. In this study, we asked experts to express their ideas through linguistic terms. In the fuzzy set theory, the linguistic terms are considered as variables. The linguistic variables related to comparison of the importance of criteria are defined as: • ''Perfect", ''very important", "fairly important" ''important" to express the higher importance.
In this paper, the computational technique for the fuzzification of linguistic terms is according to the triangular Membership Functions (MFs).  Membership function of linguistics variables for comparing two criteria Table 1 shows the linguistic variables with their corresponding TFNs. If criterion i has one TFN as (l, m, u) when compared to factor j, then j has the reciprocal value as (1/l, 1/m, 1/u) when compared to i. In this case, criterion i is more important than j and criterion j is less important than i. The reciprocal TFN is considered to cover the inverse importance relation between the criteria.

Results and Discussion
This study evaluated and ranked the ASEAN ECAs except the Lao PDR and Cambodia. The seven

Weighting of criteria
The experts were requested to complete the part of the questionnaire regarding pairwise comparison of criteria through the linguistic terms that are shown in table 1. Table 3 Table 5).
The pairwise comparison matrix of criteria is constructed with the aggregated fuzzy values obtained. (Table 6).  We had one comparison for each two criteria to avoid the inconsistency of data; hence, half of the comparison matrix is empty. The reciprocal values are used to fill in the pairwise comparison matrix (see table 8). The completed matrix (table 8) is used in equations (1-7) for calculation of the eigenvalue and obtaining the weight of the criteria. The obtained weights and ranks are shown in Table 9. Table 9 Criteria weights and ranks The inconsistency of FAHP is checked using equation 6. The maximum eigenvalue = 8.14415 and the inconsistency ratio is IR=0.090692 when it is less than 0.1; therefore, the answer is significant.
The results indicate that the EM with a weight of 0.216684 is the most important criterion in the electronic commerce law and that TR, with a difference of 0.009675 to the EM, is in second place.
The PD, with a weight of 0.0276018, was allocated the least importance and stood in last place.

Ranking of ASEAN ECAs
The experts filled out the second part of the questionnaire regarding pairwise comparison of ASEAN ECAs for all of the criteria. Table 10 is obtained from the second part of the questionnaire that is related to a comparison of countries in terms of ECA. This table shows the answers of ten experts (E1, E2, …, E10) for a comparison of the countries with TD as a sample.  The process of defuzification and aggregation of data is similar to what we did for the criteria weighting. The results of the ranking of countries in each criterion are stated separately in table 11.
The value of IR in all calculations is less than 0.1, which shows the significance of data and the consistency of the experts' answers.
The overall ranking of the eight ASEAN ECAs with respect to the TD rule is shown in Figure 6.
While    VLET is the last one. SETA followed the UNCUEC in drafting the EA rule and stood in sixth place. Figure 10 demonstrates the rank of ASEAN ECAs with respect to EA. The legal recognition of EM, which results in recognition of the e-contracts, is one of the important principles in all ECAs. Seven states were placed in the second rank, as they have a similar weight for the EM. They drafted their EM rules following Article 5 and Article 5 bis of the MLEC.
Although the MLEC and the UNCUEC have almost similar rules concerning the legal recognition of the EM, they have a different appearance. However, METL was ranked as first using the pairwise comparison of the ECAs by experts. METL has drafted a more comprehensive provision on legal recognition of the EM, although the general rule stated is the same as that of the other countries. Figure 11 shows the overall ranking of the ASEAN ECAs regarding the EM criterion.

Figure 11
Rank of ASEAN ECAs with regard to EM Act even before enacting the ECA may be considered its priority with respect to the ES criterion.
The rest have recognized the validity and legal effects of ES in their ECAs. The PECA and BETA were both placed in fourth, and VLET, with a low difference in weight, obtained fifth position. Figure 12 illustrates the evaluation of ASEAN ECAs concerning the ES rules.  This new Act was mostly based on the UNCUEC, which may be the main reason for SETA obtaining the maximum rate in the TR and TD rules (Fig 13). As stated earlier, the UNCUEC is the second general document adopted by the UNCITRAL on the law of electronic commerce and is binding upon its signatories. The drafters of this convention, tried to consider new technologies in drafting the provisions and somehow amend the MLEC rules. The other seven states followed the MLEC as a basis for their ECAs.

Conclusions and Future Work
There are various considerations, such as time and place for the evaluation of ECAs, with respect to the electronic contracts. In investigating these considerations, we established the procedures for identifying the most important attributes of the e-commerce Acts. Since the characteristics of electronic contracts are the main concern in the ECAs, we focused on it for our evaluation. We captured the experts' assessment on these ECAs based on the said attributes.
The evaluation procedure consists of the following steps: i) determination of the evaluation criteria for ECA; ii) obtaining the importance of each criterion by Fuzzy AHP according to all of the experts' opinion; iii) representing the evaluation of ASEAN ECAs for each criterion by Fuzzy AHP. Fuzzification of assessors' responses has captured their real preference accurately.
The individual assessments were aggregated as an overall assessment for each country under each criterion. We employed the fuzzy set theory for the fuzzification of AHP to deal with uncertain and subjective data, and the environment.
In traditional legal research, the evaluation of Acts is not structured. To the best of our knowledge, quantitative strategies have not been used for the evaluation and comparison of the Acts. In this paper, we used the Fuzzy AHP and the concept of hierarchical structure to make the pairwise comparison among elements.
In the AHP method, where there are many elements, the long investigation time causes the interviewees to feel impatient and inconsistency increases. However, in this research the number of determined criteria and the ASEAN ECAs is less than ten; therefore, applying Fuzzy AHP is not only permissible with a lower inconsistency index, but it is significant due to constructing pairwise comparison and structured evaluation.
In this study, using the membership function to measure the linguistic variables can exactly reflect the experts' opinions. Therefore, the fuzzy logic, thinking and results of the fuzzy approach are better than the traditional MCDM approach. The significant findings of this study cover several perspectives. The ECAs that have followed the UNCUEC in drafting TD and TR rules, rated in the higher level. However, as the majority followed the MLEC, those that followed the MLEC provisions accurately or improved its principles, obtained a significant rate. The findings suggest that the integration of MLEC and UNCUEC resulting in a unified electronic commerce regulation would be more comprehensive if taking into account the recent technological developments.
The achieved results could provide a reference guide in the said legal harmonization process or for future amendments of the present ECAs. This would be in line with the ASEAN regional economic integration mission and the ASEAN Economic Community (AEC) scheme.
The harmonization of ECAs among ASEAN countries will increase the number of online transactions affecting the traders' trust. Subsequently, the number of related legal disputes will reduce. Furthermore, it would be a reference to harmonize related commercial laws in the region to support economic integration. The evaluation of ASEAN ECAs would play a significant role to help legislators to provide a unified e-commerce law at the regional level.
This study has a few limitations. Firstly, our survey respondents were chosen from academic members of the Malaysian universities due to the difficult process of approaching legislators of ASEAN member states. Therefore, the results represent the preference of academic lawyers.
Secondly, two countries are not involved because Laos lacks any English translation of its ECA and Cambodia has no law in place. Future research should prepare a unified academic draft on ECA for the ASEAN region considering other related criteria that is compatible with new technological advancements and new commercial needs.