First Author\[1\]\*, Second Author2 (Double Blind Review: Please do not type or edit anything here until final-camera ready submissions) 1First affiliation, City and Country (Please do not type or edit anything here, our editors will do the work for you) 2Second affiliation, City and Country (Please do not type or edit anything here, our editors will do the work for you)
ARTICLE INFO ABSTRACT

Article history:

Received XX Month 2024

Revised XX Month 2024

Accepted XX Month 2024

Online first

Published 1 September 2024

 This study addresses the problem of understanding users' preferences for courier services in Malaysia. With the increasing reliance on online shopping and package delivery, it is crucial to identify the key factors that shape users' decision-making processes when selecting a courier service. To tackle this problem, the Fuzzy Analytical Hierarchy Process (FAHP) method is employed. The study aims to determine the relative importance of criteria such as responsiveness, reliability, empathy, assurance, and tangibility, along with their corresponding sub-criteria. A group of experts defined these criteria and sub-criteria, and a questionnaire was distributed to gather data on customer preferences. Using the FAHP technique, the study constructed a hierarchy tree, created a fuzzy pairwise comparison matrix, and calculated criterion and sub-criteria weights. The findings highlight the importance of responsiveness, particularly easy contact with courier companies, as the primary criterion. Enhancing services based on these insights can help users select suitable courier services while improving competitiveness in the industry.

Keywords:

Courier Services

Fuzzy AHP

Online Shopping

Customer Preferences

DOI:

10.24191/jcrinn.v9i2
# INTRODUCTION Courier services are crucial in today’s globalized and fast-paced economy, acting as the backbone for efficient business operations and seamless customer communication. These services ensure the timely and reliable delivery of goods, which is essential for maintaining customer satisfaction and operational efficiency. Prominent international courier companies such as DHL, JNT, GDEX, POS LAJU, and NINJA VAN have significantly contributed to the courier industry by providing extensive local and international express services. Their ability to deliver products swiftly and reliably has made them indispensable to both individuals and businesses worldwide (Gruenwald, 2020). Despite their importance, courier services face numerous challenges, particularly in aligning with customer preferences. The COVID-19 pandemic has further complicated this landscape, with lockdowns in Malaysia leading to a surge in online shopping and an unprecedented demand for delivery services. In 2021, Malaysian courier service providers delivered approximately 737.36 million domestic items, driven by an e-commerce market valued at 28.5 billion Malaysian ringgit (Statista Research Department, 2022). However, this rapid growth has also highlighted issues such as service quality, reliability, and the handling of increased volumes of packages. Reports of mishandling and delays have surfaced, emphasizing the need for improved service management to maintain customer satisfaction and competitive edge. Understanding and addressing these challenges related to customer preferences is the main intention of this research. To address the complex decision-making involved in selecting courier services, several Multi-Criteria Decision-Making (MCDM) techniques have been developed. These techniques aid in evaluating and prioritizing multiple criteria to make informed decisions. Common MCDM methods include the Analytical Hierarchy Process (AHP), which uses a structured hierarchy to model a decision problem and performs pairwise comparisons to determine the relative importance of criteria (Siekelova et al., 2021). Another method is the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), which identifies solutions from a finite set of alternatives based on their distance to an ideal solution (Dutta et al., 2019). Additionally, the Elimination and Choice Expressing Reality (ELECTRE) method is based on outranking and is used to handle complex decision-making scenarios with conflicting criteria (Zahid et al., 2022). The Fuzzy Analytical Hierarchy Process (FAHP) extends the traditional AHP by incorporating fuzzy logic to handle the uncertainty and vagueness in human judgment. FAHP is particularly advantageous in situations where decision-makers face ambiguity in their evaluations. The method's ability to process imprecise data makes it highly suitable for assessing complex, subjective preferences in courier service selection. Previous applications of FAHP have demonstrated its effectiveness in various fields, including supply chain management, where it has been used to prioritize factors affecting supply chain performance (Mistarihi et al., 2023), healthcare, where it has helped in evaluating healthcare service quality by integrating fuzzy logic to manage subjective assessments (Azam et al., 2017), and transportation, where it has been applied to assess transportation modes considering multiple criteria and their fuzzy interactions (Zaid et al., 2024). Given its ability to handle complex, subjective, and imprecise data, FAHP is chosen for this research to assess customer preferences for courier services in Malaysia effectively. The structure of this paper is as follows: Section 2: Preliminaries provides an overview of the key concepts and background information relevant to the study, including an introduction to the FAHP method and a review of the literature on courier service preferences. Section 3: Methodology describes the research design, including the selection and definition of criteria and sub-criteria, data collection methods, and the implementation of the FAHP technique. Section 4: Results and Discussion presents the findings from the FAHP analysis, detailing the calculated weights of criteria and sub-criteria, and discusses the implications of these findings in the context of existing literature and industry practices. Finally, Section 5: Conclusion summarizes the key findings of the study, discusses its contributions to the field, outlines its limitations, and provides suggestions for future research directions. Through this FAHP-based assessment, the study aims to contribute to the academic understanding of customer preferences in the courier service industry and offer practical recommendations for service improvement. # Preliminaries ## Background of Courier Services in Malaysia The courier service industry in Malaysia has seen significant growth over the past few decades, becoming a vital part of the country's economy. This growth is largely due to the rapid rise of e-commerce, which has increased the demand for reliable and efficient delivery services. Key players in the Malaysian market include international companies such as DHL and FedEx, as well as local providers like POS LAJU, JNT, GDEX, and NINJA VAN. These companies offer a variety of services, including local and international deliveries and comprehensive logistics solutions for both individual customers and businesses. However, despite their importance, courier services face several challenges. Issues of service quality and reliability are ongoing, particularly during peak shopping seasons and promotional periods when the volume of packages increases. The COVID-19 pandemic further intensified these challenges, as lockdowns led to a surge in online shopping and increased pressure on courier services. This surge exposed weaknesses in the system, with delays, mishandling of packages, and customer dissatisfaction becoming more common. Addressing these issues is crucial for maintaining customer trust and competitive advantage. Several studies have examined these challenges. For example, Yee & Daud (2011) investigated the impact of service quality dimensions on customer satisfaction in package delivery. They found that tangibility, reliability, and assurance significantly affected customer satisfaction, while empathy and responsiveness did not. Similarly, Tabassum & Badiuddin (2014) used a modified SERVQUAL approach to measure the gap between customer expectations and perceptions of service quality, identifying dependability and responsiveness as areas needing improvement. SERVQUAL, though widely used, has been criticized for its generality and lack of differentiation between service quality and customer satisfaction. This has led researchers to explore other methods for evaluating courier services. Liu & Liu (2014) used SERVQUAL and Logistic Service Quality (LSQ) models to study express logistics in Changdao County, China. They found significant gaps between customer expectations and perceived service quality, particularly in the protection dimension. Another study by Valaei et al. (2016) developed a specialized scale called CouQual, which includes dimensions such as promptness, convenience, accuracy, safety, and tangibles. They found that promptness, safety, and convenience were the most critical factors influencing customer perception of service quality. In summary, while traditional methods like SERVQUAL have provided insights into service quality, they have limitations. Researchers have therefore developed and tested alternative models to better capture the specific characteristics and challenges of the courier service industry. ## Understanding Customer Preferences Understanding customer preferences is crucial for courier service providers to improve their services and enhance customer satisfaction. Key factors influencing these preferences include responsiveness, reliability, empathy, assurance, and tangibility. These factors significantly shape customer decisions and perceptions, impacting their overall experience and loyalty to the service provider. **2.2.1 Responsiveness** Responsiveness refers to how quickly and effectively a company responds to customer inquiries and issues. It includes administrative efficiency, attention to customer needs, flexible operating hours, and personal attention from staff (El Saghier & Nathan, 2013). A study found that responsiveness positively correlates with customer happiness (Al-Weshah et al., 2013). Ensuring staff can communicate directly with customers and assist promptly improves responsiveness. Delays and distance can affect response times, especially in supply chains (Naik et al., 2010). **2.2.2 Reliability** Reliability involves consistently providing services correctly, on time, and maintaining accurate records (El Saghier & Nathan, 2013). This includes fulfilling customer demands accurately, managing records correctly, and ensuring timely deliveries. A study indicated that reliability significantly impacts customer satisfaction (Janahi & Al Mubarak, 2017). Customers expect high reliability from service providers, and when met, it increases their satisfaction and trust (Shahin & Chan, 2006; Alnsour et al., 2014). **2.2.3 Empathy** Empathy is about providing personalized attention and understanding customer needs (El Saghier & Nathan, 2013). It includes convenient office hours and hiring agents who care about customers' demands. Studies have shown that empathy positively affects customer satisfaction (Sureshchandar et al., 2002; Alnsour et al., 2014). Customers appreciate when service providers consider their needs, leading to higher satisfaction. **2.2.4 Assurance** Assurance involves the confidence customers have in the service, influenced by employees' expertise, friendliness, and professionalism (El Saghier & Nathan, 2013). It includes aspects like capability, kindness, validity, and security. Studies show that assurance positively impacts customer satisfaction and trust (Shahin & Chan, 2006; Naik et al., 2010). Ensuring that customers feel confident in the services provided is crucial for maintaining trust and satisfaction. **2.2.5 Tangibility** Tangibility refers to the physical aspects of the service, such as the condition of delivered packages and the appearance of facilities and equipment (El Saghier & Nathan, 2013). While tangibility has a less significant link to customer satisfaction, it still plays a role in the overall perception of service quality (Parasuraman et al., 1998). Contradictory findings suggest that in some sectors, like telecommunications in China, tangibility significantly impacts service quality and customer happiness (Noble et al., 2002). ## Fuzzy Analytical Hierarchy Process (FAHP). The Fuzzy Analytic Hierarchy Process (FAHP) is a robust methodology that integrates the traditional Analytic Hierarchy Process (AHP) with fuzzy set theory, employing triangular fuzzy numbers to enhance decision-making and problem-solving (Chang, 1996). While AHP is widely recognized and utilized, it faces criticism for not adequately addressing the inherent volatility and imprecision when translating decision-makers' subjective perceptions into exact values (Deng, 1999). FAHP was developed to mitigate these limitations by incorporating fuzziness into the decision-making process (Mikhailov & Tsvetinov, 2004). Through the use of fuzzy numbers, FAHP allows decision-makers to express preferences and approximations that more accurately reflect the inherent uncertainty in their judgments (Erensal et al., 2006; Wang et al., 2008; Feng, 1995). Fuzzy set theory, developed by Zadeh (1965), underpins FAHP by providing a mathematical framework to simulate the fuzziness of human cognitive processes. Unlike traditional set theory, fuzzy set theory does not have sharp boundaries between classes (Van Laarhoven & Pedrycz, 1983). The membership function of a fuzzy set spans the range of real numbers, typically normalized to the interval \[0, 1\], allowing for a spectrum of membership levels. FAHP utilizes this flexibility by allowing decision-makers to choose from a range of values that reflect their confidence levels and to describe their outlook in broad terms, such as optimistic, pessimistic, or moderate (Jeganathan, 2003; Lee et al., 2008). The versatility and effectiveness of FAHP have been demonstrated across various fields. For instance, Bhatt et al. (2021) applied FAHP to select an appropriate ERP system for SMEs, identifying deployment cost as the most critical factor. Büyüközkan et al. (2011) used FAHP to evaluate healthcare service quality in Turkey, highlighting the importance of compassion, competence, and reliability in delivering satisfactory services. Al-Shammari & Mili (2021) implemented FAHP for customer selection in commercial banks, showcasing the method's efficacy through a detailed numerical example. Recent research further underscores the broad applicability of FAHP. Idris (2020) used FAHP to determine and rank the factors contributing to divorce in Perlis, Malaysia. FAHP was also employed to identify effective measures to prevent the spread of COVID-19 (Idris et al., 2023), select the best student award (Abd Aziz et al., 2023), determine the criteria for choosing tour packages in Langkawi Island (Zahrin et al., 2022), analyze the factors influencing career choices of graduate students at UiTM Perlis (Abd Aziz et al., 2024), and rank factors in selecting online shopping platforms in Malaysia (Abd Aziz et al., 2024). From the extensive review of literature, it is evident that FAHP is a practical and straightforward approach to tackling complex selection problems, particularly in scenarios where uncertainty and imprecision are significant. The following section will outline the characteristics and applications of FAHP, providing a deeper understanding of its utility and effectiveness. # METHODOLOGY This section outlines the methodology employed to assess courier service preferences among Malaysian users using the Fuzzy Analytic Hierarchy Process (FAHP). The FAHP approach is utilized through a structured set of procedural steps to ensure a comprehensive and systematic analysis. Step 1: Establish the hierarchical structure. The methodology begins with the construction of a comprehensive three-tiered hierarchical framework to assess courier service preferences among Malaysian users, as depicted in Fig. 1. This hierarchical structure is designed to systematically break down and evaluate the complex factors influencing user preferences. At the top level, the primary objective is to rank the factors that influence users in selecting the most preferable courier services in Malaysia. The second level, known as the criteria layer, categorizes the factors influencing user preferences into five main groups: Responsiveness, Reliability, Empathy, Assurance, and Tangibility. Responsiveness (C1) refers to the ability of the courier service to respond promptly and effectively to customer needs, while Reliability (C2) denotes the consistency and dependability of the courier service in fulfilling its promises. Empathy (C3) involves the courier service's ability to understand and address customer needs and concerns. Assurance (C4) encompasses the confidence and trust that customers have in the courier service, including the knowledge and courtesy of employees. Tangibility (C5) pertains to the physical aspects of the courier service, such as the condition of delivered packages and the appearance of personnel. The third level, the sub-criteria layer, further breaks down these main criteria into more specific factors, resulting in a detailed and nuanced assessment framework. Under Responsiveness, the factors include easy contact with the courier company (SC1), efficient communication with the courier company (SC2), and efficient handling of returns (SC3). Reliability is broken down into successful delivery attempt (SC4), lack of damage to packages (SC5), and timeliness of delivery (SC6). Empathy includes readiness to react quickly to problem interference (SC7), availability of services (SC8), and understanding the demand of customers (SC9). Assurance is divided into knowledge and competency of employees (SC10), polite and courteous employees (SC11), and trust in the courier company (SC12). Finally, Tangibility includes the aesthetic and neat appearance of the courier (SC13), characteristic trademark and uniform color (SC14), and the goodness of the package condition (SC15). By structuring the decision-making problem into these hierarchical levels, we can methodically evaluate each factor's relative importance. This framework provides a clear and organized approach to understanding and prioritizing the various elements that influence courier service preferences among Malaysian users. ![](667a357f9e1d5_media/media/image1.png) Fig. 1: A hierarchical model for preferable courier service. Step 2: Create a Fuzzy Pairwise Comparison Matrix In this step, the decision-maker utilizes linguistic concepts, as provided in Table 1, to perform pairwise comparisons of criteria or sub-criteria. These linguistic concepts are associated with triangular fuzzy numbers (TFNs) to incorporate the inherent fuzziness and uncertainty in human judgment. The process for creating a fuzzy pairwise comparison matrix using TFNs is detailed below. The decision-maker starts by comparing each pair of criteria (or sub-criteria) using the linguistic terms that best describe their relative importance. These linguistic terms are then converted into TFNs, which are represented by three values: the minimum possible value 𝑎, the most possible value 𝑏, and the maximum possible value 𝑐. The membership function of a TFN, which defines how each value within the range contributes to the degree of membership, is mathematically expressed as follows: | | | | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --- | | \[\mu_{\text{TFN}}\left( x \right) = \left\{ \begin{matrix} \frac{x - a}{b - a} & ; & a \leq x \leq b \\ \frac{c - x}{c - b} & ; & b \leq x \leq c \\ 0 & ; & \text{otherwise} \\ \end{matrix} \right.\ \] | (1) | These TFNs are typically expressed using a 1-9 scale, where each scale value corresponds to a linguistic term. This approach acknowledges that experts often find it challenging to quantify their judgments precisely using numerical values, hence the use of linguistic variables (Zadeh, 1965). For instance, if the decision-maker states that "Criterion 1 (C1) is Slightly Important compared to Criterion 2 (C2)," they would use the fuzzy triangular scale (2, 3, 4). Conversely, when comparing C2 to C1, the inverse fuzzy triangular scale (1/4, 1/3, 1/2) would be employed, reflecting the reciprocal relationship between the two criteria (Ayhan, 2013). This process is repeated for all pairs of criteria and sub-criteria, resulting in a complete fuzzy pairwise comparison matrix. \[CHART\]\[CHART\] Fig. 2: (a) Membership function of triangular fuzzy numbers, (b) Linguistic variables. Fig. 2 illustrates the degree of possibility for TFNs, while Table 1 provides the non-fuzzy numbers representing the 1-9 scale, linguistic variables, and the TFNs with their reciprocal values used in the comparisons. By systematically applying this method, the decision-maker can construct a matrix that accurately captures the relative importance of each criterion and sub-criterion, taking into account the fuzziness and uncertainty inherent in human judgments. Table . Linguistic variables with its associated non-fuzzy number and TFN for pairwise comparison.
Linguistic Variables

AHP Scale

(Non-Fuzzy Numbers)

Fuzzy AHP Scale

(TFNs)

Fuzzy AHP Scale

(Reciprocal TFNs)
Equally Important 1 (1,1,1) (1,1,1)
Slightly Important 3 (2,3,4) (1/4,1/3,1/2)
Moderately Important 5 (4,5,6) (1/6,1/5,1/6)
Strongly Important 7 (6,7,8) (1/8,1/7,1/6)
Extremely Important 9 (9,9,9) (1/9,1/9,1/9)
Intermediate Value 2 (1,2,3) (1/4,1/2,1)
4 (3,4,5) (1/5,1/4,1/3)
6 (5,6,7) (1/7,1/6,1/5)
8 (7,8,9) (1/9,1/8,1,7)
The approach used to obtain input from expert decision-makers involves conducting interviews or using questionnaires. At this stage, the scale used is the non-fuzzy number scale, specifically the 1-9 scale. The collected input is then converted into a pairwise comparison matrix (PCM) to be checked for consistency, which will be demonstrated in the next step. Once the consistency value has been verified, the original PCM will be transformed into a PCM with TFN values, as shown in equation (3). From this updated PCM, the subsequent steps in the FAHP process will be carried out. The following equation (2) is the general form of the PCM using the 1-9 scale values. | | | | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --- | | \[A^{k} = \begin{bmatrix} \begin{matrix} \ \ \ 1 & d_{12}^{k} & \begin{matrix} \cdots & d_{1n}^{k} \\ \end{matrix} \\ \end{matrix} \\ \begin{matrix} d_{21}^{k} & 1 & \begin{matrix} \cdots & d_{2n}^{k} \\ \end{matrix} \\ \end{matrix} \\ \begin{matrix} \begin{matrix} \vdots & \vdots & \begin{matrix} \ddots & \vdots \\ \end{matrix} \\ \end{matrix} \\ \begin{matrix} d_{n1}^{k} & d_{n2}^{k} & \begin{matrix} \cdots & 1 \\ \end{matrix} \\ \end{matrix} \\ \end{matrix} \\ \end{bmatrix}\] | (2) | where \(k\) denotes the \(k\)-th expert, and \(d_{11}^{k},d_{12}^{k},\ldots,d_{\text{nn}}^{k}\) represent the input values for each factor given by the experts on a scale of 1 to 9. It is important to note that \(d_{\text{ji}}^{k} = \frac{1}{d_{\text{ij}}^{k}}\) and \(d_{\text{ii}}^{k} = 1\) for every \(i,j = 1,2,\ldots,n\). In other words, if the essential preference \(d_{\text{ij}}^{k}\)​ is located in the upper triangle of the matrix, then the reciprocal value \(d_{\text{ji}}^{k} = \frac{1}{d_{\text{ij}}^{k}}\)​ must be positioned in the lower triangle, and vice versa. This ensures that the matrix accurately reflects the relative importance of each factor as assessed by the experts. Next, we present the general form of the pairwise comparison matrix (PCM) with triangular fuzzy number (TFN) values. | | | | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --- | | \[{\widetilde{A}}^{k} = \begin{bmatrix} \begin{matrix} (1,1,1) & {\widetilde{d}}_{12}^{k} & \begin{matrix} \cdots & {\widetilde{d}}_{1n}^{k} \\ \end{matrix} \\ \end{matrix} \\ \begin{matrix} {\widetilde{d}}_{21}^{k} & (1,1,1) & \begin{matrix} \cdots & {\widetilde{d}}_{2n}^{k} \\ \end{matrix} \\ \end{matrix} \\ \begin{matrix} \begin{matrix} \vdots & \vdots & \begin{matrix} \ddots & \vdots \\ \end{matrix} \\ \end{matrix} \\ \begin{matrix} {\widetilde{d}}_{n1}^{k} & {\widetilde{d}}_{n2}^{k} & \begin{matrix} \cdots & (1,1,1) \\ \end{matrix} \\ \end{matrix} \\ \end{matrix} \\ \end{bmatrix}\] | (3) | Here, \({\widetilde{d}}_{11}^{k},{\widetilde{d}}_{12}^{k},\ldots,{\widetilde{d}}_{\text{nn}}^{k}\)​ represent the input values that have been transformed from non-fuzzy numbers to TFNs, as illustrated in Table 1. This transformation allows for a more nuanced and accurate representation of the expert assessments, capturing the inherent uncertainty and fuzziness in their judgments. Step 3: Verifying the Consistency of the Judgement Matrix. Verifying the consistency of the judgment matrix is an indispensable task due to the complexity of the objective and the subjectivity inherent in decision-makers' understanding. To ensure the reliability of the judgments, the consistency ratio index (CR) is calculated using the following formula: | | | | ------------------------------------ | --- | | \[CR = \frac{\text{CI}}{\text{RI}}\] | (4) | where | | | | ------------------------------------------ | --- | | \[CI = \frac{\text{λ}_{\max} - n}{n - 1}\] | (5) | and \(\text{λ}_{\max}\) represents the largest eigenvalue of the comparison matrix, and *n* denotes the order or size of the matrix. *RI* stands for the average random consistency index of the judgment matrix, as shown in Table 2. Table 2. Random consistency index. | | | | | | | | | | | | | --------------------- | ---- | ---- | ---- | ---- | ---- | ---- | ---- | ---- | ---- | ---- | | **Matrix Order, *n*** | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | | **Ratio Index, *RI*** | 0.00 | 0.00 | 0.58 | 0.90 | 1.12 | 1.24 | 1.32 | 1.41 | 1.45 | 1.49 | If the *CR* is less than 10% or 0.1, the matrix consistency is considered reliable and acceptable. If it exceeds 10%, the assessor must re-evaluate and adjust the pairwise comparison matrix. The matrix used in calculating this consistency index is in the form of a matrix with non-fuzzy number elements, as shown in Equation (2). Step 4: Calculate the Fuzzy Geometric Mean. Once the consistency of the judgment matrices has been confirmed, these matrices are transformed into pairwise comparison matrices (PCM) with triangular fuzzy numbers (TFN), as shown in Equation (3). If multiple experts are involved, the next step is to calculate the average value for each PCM obtained. The formula used to obtain the updated matrix is as follows: | | | | -------------------------------------------------------------------------------------------------------------------------------------- | --- | | \[\widetilde{A} = \left\lbrack {\widetilde{d}}_{\text{ij}} \right\rbrack = \frac{\sum_{k = 1}^{K}{\widetilde{d}}_{\text{ij}}^{k}}{K}\] | (6) | where K is the number of experts. Here \({\widetilde{d}}_{\text{ij}}^{k} = \left( l_{\text{ij}}^{k},m_{\text{ij}}^{k},u_{\text{ij}}^{k} \right)\), where *l,m,u* represent the values in the TFNs separately. According to Buckley & Eslami (2002), the fuzzy geometric mean, \({\widetilde{r}}_{i}\) for the fuzzy comparison values of each criterion *i*, is calculated using Equation (7). | | | | -------------------------------------------------------------------------------------------------------------------------------------------------------- | --- | | \[{\widetilde{r}}_{i} = \left( {\widetilde{d}}_{i1} \times {\widetilde{d}}_{i2} \times \cdots \times {\widetilde{d}}_{\text{iN}} \right)^{\frac{1}{N}}\] | (7) | where *N* is the number of factors. Following this, determine the vector summation of the geometric mean, \(\sum_{i = 1}^{N}{\widetilde{r}}_{i}\) and its reciprocal, \(\left( \sum_{i = 1}^{N}{\widetilde{r}}_{i} \right)^{- 1}\) by applying Equations (8) and (9), respectively. | | | | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --- | | \[\sum_{i = 1}^{N}{\widetilde{r}}_{i} = \left( \sum_{}^{}{l_{{\widetilde{r}}_{i}},\sum_{}^{}{m_{{\widetilde{r}}_{i}},\sum_{}^{}u_{{\widetilde{r}}_{i}}}} \right)\] | (8) | | \[\left( \sum_{i = 1}^{N}{\widetilde{r}}_{i} \right)^{- 1} = \left( \frac{1}{\sum_{}^{}u_{{\widetilde{r}}_{i}}},\frac{1}{\sum_{}^{}m_{{\widetilde{r}}_{i}}},\frac{1}{\sum_{}^{}l_{{\widetilde{r}}_{i}}} \right)\] | (9) | Step 5: Calculate the Fuzzy Weight. To obtain the fuzzy weight of each criterion \({\widetilde{w}}_{i}\), multiply each \({\widetilde{r}}_{i}\) by the reciprocal of the vector summation of the geometric mean, as specified in Equation (10). | | | | ------------------------------------------------------------------------------------------------------------- | ---- | | \[{\widetilde{w}}_{i} = {\widetilde{r}}_{i} \times \left( \sum_{i = 1}^{N}{\widetilde{r}}_{i} \right)^{- 1}\] | (10) | Step 6: Defuzzify and Normalize the Weight of Criteria It must be noted that \({\widetilde{w}}_{i}\) are still expressed as triangular numbers, \(\left( {\widetilde{l}}_{i},{\widetilde{m}}_{i},{\widetilde{u}}_{i} \right)\). To convert these into crisp values, the Center of Area method proposed by Chou & Chang (2008) can be employed in the defuzzification process, as shown in Equation (11): | | | | ------------------------------------------------------------------------------------- | ---- | | \[M_{i} = \frac{{\widetilde{l}}_{i} + {\widetilde{m}}_{i} + {\widetilde{u}}_{i}}{3}\] | (11) | Next, to obtain the final weight values, Equation (12) is used to normalize \(M_{i}\) and find the normalized weight, \(Z_{i}\). | | | | ----------------------------------------------- | ---- | | \[Z_{i} = \frac{M_{i}}{\sum_{i = 1}^{N}M_{i}}\] | (12) | By following these steps, the fuzzy weights are converted into crisp, normalized values suitable for further analysis. These six processes determine the normalized weights of the criteria and sub-criteria. The global weight for each sub-criterion is then calculated by multiplying each sub-criteria's weight by the corresponding criterion's weight. These results highlight to the decision-maker which sub-criteria have the greatest global weight, providing clear insights into the most influential factors. # RESULT AND DISCUSSION This section presents the results and discussion based on the FAHP process to identify the factors influencing users in choosing courier services. The study gathers insights from two experts with extensive experience in the courier service industry. Separate discussions and interviews were conducted with each expert to obtain input, which was then used to create the Pairwise Comparison Matrix (PCM) for criteria and sub-criteria from Fig. 1- an essential first step in the FAHP process. These matrices are shown in Tables 3 and 4, along with the consistency ratio (CR) values for each. Table 3. Pairwise comparison matrix for criteria, including the consistency ratio for experts 1 and 2. | | | | | | | | | ------------ | ------ | ------ | ------ | ------ | ------ | ------ | | **EXPERT 1** | | | | | | | | **Criteria** | **C1** | **C2** | **C3** | **C4** | **C5** | **CR** | | **C1** | 1 | 5 | 6 | 8 | 9 | | | **C2** | 1/5 | 1 | 1/2 | 3 | 7 | | | **C3** | 1/6 | 2 | 1 | 2 | 6 | 0.0849 | | **C4** | 1/8 | 1/3 | 1/2 | 1 | 4 | | | **C5** | 1/9 | 1/7 | 1/6 | 1/4 | 1 | | | **EXPERT 2** | | | | | | | | **Criteria** | **C1** | **C2** | **C3** | **C4** | **C5** | **CR** | | **C1** | 1 | 8 | 3 | 4 | 1 | | | **C2** | 1/8 | 1 | 1/5 | 1/2 | 1/4 | | | **C3** | 1/3 | 5 | 1 | 2 | 1 | 0.0448 | | **C4** | 1/4 | 2 | 1/2 | 1 | 1/6 | | | **C5** | 1 | 4 | 1 | 6 | 1 | | Table 4. Pairwise comparison matrix for sub-criteria including the consistency ratio for experts 1 and 2. | | | | | | | ---------------- | -------- | -------- | -------- | ------ | | **EXPERT 1** | | | | | | **Sub-Criteria** | **SC1** | **SC2** | **SC3** | **CR** | | **SC1** | 1 | 5 | 9 | | | **SC2** | 1/5 | 1 | 4 | 0.062 | | **SC3** | 1/9 | 1/4 | 1 | | | **Sub-Criteria** | **SC4** | **SC5** | **SC6** | **CR** | | **SC4** | 1 | 5 | 1/3 | | | **SC5** | 1/5 | 1 | 1/9 | 0.025 | | **SC6** | 3 | 9 | 1 | | | **Sub-Criteria** | **SC7** | **SC8** | **SC9** | **CR** | | **SC7** | 1 | 1/2 | 1/9 | | | **SC8** | 2 | 1 | 1/5 | 0.001 | | **SC9** | 9 | 5 | 1 | | | **Sub-Criteria** | **SC10** | **SC11** | **SC12** | **CR** | | **SC10** | 1 | 1/5 | 1/8 | | | **SC11** | 5 | 1 | 1/3 | 0.038 | | **SC12** | 8 | 3 | 1 | | | **EXPERT 2** | | | | | | **Sub-Criteria** | **SC1** | **SC2** | **SC3** | **CR** | | **SC1** | 1 | 3 | 9 | | | **SC2** | 1/3 | 1 | 2 | 0.016 | | **SC3** | 1/9 | 1/2 | 1 | | | **Sub-Criteria** | **SC4** | **SC5** | **SC6** | **CR** | | **SC4** | 1 | 1/6 | 1/4 | | | **SC5** | 6 | 1 | 1 | 0.016 | | **SC6** | 4 | 1 | 1 | | | **Sub-Criteria** | **SC7** | **SC8** | **SC9** | **CR** | | **SC7** | 1 | 1/9 | 1/5 | | | **SC8** | 9 | 1 | 4 | 0.062 | | **SC9** | 5 | 1/4 | 1 | | | **Sub-Criteria** | **SC10** | **SC11** | **SC12** | **CR** | | **SC10** | 1 | 1/8 | 1/6 | | | **SC11** | 8 | 1 | 3 | 0.064 | | **SC12** | 6 | 1/3 | 1 | | Based on the consistency ratio values obtained for each criterion and sub-criterion as shown in Tables 3 and 4, all values are within the acceptable range of less than 10% or 0.1. This indicates that the evaluations made by all experts are reliable and acceptable. The next step is to calculate the Fuzzy Geometric Mean. However, before proceeding, all PCM values in Tables 3 and 4 need to be converted to TFNs as indicated in Table 1. After converting to TFNs, the average value is calculated for each PCM of criteria and sub-criteria, as more than one expert is involved, as shown in Equation (6). These values are presented in Tables 5 and 6. Table 5. Average of pairwise comparison matrix for criteria | | | | | | | | ------------ | ------------------- | ------------------- | ----------------- | ----------------- | ------------------- | | **Criteria** | **C1** | **C2** | **C3** | **C4** | **C5** | | **C1** | (1,1,1) | (5.5,6.5,7.5) | (3.5,4.5,5.5) | (5,6,7) | (5,5,5) | | **C2** | (0.139,0.163,0.196) | (1,1,1) | (0.25,0.35,0.625) | (1.167,1.75,2.5) | (3.1,3.625,4.167) | | **C3** | (0.196,0.25,0.35) | (2.5,3.5,4.5) | (1,1,1) | (1,2,3) | (3,3.5,4) | | **C4** | (0.156,0.188,0.238) | (0.625,1.167,1.75) | (0.333,0.5,1) | (1,1,1) | (0.171,0.208,0.267) | | **C5** | (0.556,0.556,0.556) | (1.563,2.071,2.583) | (0.571,0.583,0.6) | (2.6,3.125,3.667) | (1,1,1) | Table 6. Average of pairwise comparison matrix for sub-criteria. | | | | | | ---------------- | ------------------- | ------------------- | ------------------- | | **Sub-Criteria** | **SC1** | **SC2** | **SC3** | | **SC1** | (1,1,1) | (3,4,5) | (9,9,9) | | **SC2** | (0.208,0.267,0.375) | (1,1,1) | (2,3,4) | | **SC3** | (0.111,0.111,0.111) | (0.267,0.375,0.667) | (1,1,1) | | **Sub-Criteria** | **SC4** | **SC5** | **SC6** | | **SC4** | (1,1,1) | (2.071,2.583,3.1) | (0.225,0.292,0.417) | | **SC5** | (2.583,3.1,3.625) | (1,1,1) | (0.556,0.556,0.556) | | **SC6** | (2.5,3.5,4.5) | (5,5,5) | (1,1,1) | | **Sub-Criteria** | **SC7** | **SC8** | **SC9** | | **SC7** | (1,1,1) | (0.222,0.306,0.556) | (0.139,0.156,0.181) | | **SC8** | (5,5.5,6) | (1,1,1) | (1.583,2.1,2.625) | | **SC9** | (7.5,7,6.5) | (2.1,2.625,3.167) | (1,1,1) | | **Sub-Criteria** | **SC10** | **SC11** | **SC12** | | **SC10** | (1,1,1) | (0.139,0.163,0.196) | (0.127,0.146,0.171) | | **SC11** | (5.5,6.5,7.5) | (1,1,1) | (1.125,1.667,2.25) | | **SC12** | (6,7,8) | (1.125,1.667,2.25) | (1,1,1) | Once the average PCM values for each criterion and sub-criterion have been obtained, the process continues with the calculation of the Fuzzy Geometric Mean as shown in Equations (7), (8), and (9). The results of these calculations are presented in Tables 7 and 8. Table 7. Fuzzy Geometric Mean for all criteria. | | | | | | ------------------------------------------------------------ | --------- | --------- | --------- | | **Criteria,** \({\widetilde{r}}_{i}\) | ***l*** | ***m*** | ***u*** | | **C1,** \({\widetilde{r}}_{1}\) | 3.439 | 3.878 | 4.284 | | **C2,** \({\widetilde{r}}_{2}\) | 0.660 | 0.816 | 1.050 | | **C3,** \({\widetilde{r}}_{3}\) | 1.081 | 1.437 | 1.800 | | **C4,** \({\widetilde{r}}_{4}\) | 0.354 | 0.469 | 0.644 | | **C5,** \({\widetilde{r}}_{5}\) | 1.052 | 1.160 | 1.259 | | \[\sum_{i = 1}^{5}{\widetilde{r}}_{i}\] | **6.586** | **7.760** | **9.038** | | \[\left( \sum_{i = 1}^{5}{\widetilde{r}}_{i} \right)^{- 1}\] | **0.111** | **0.129** | **0.152** | Table 8. Fuzzy Geometric Mean for all sub-criteria. | | | | | | ------------------------------------------------------------ | --------- | --------- | --------- | | **Sub-Criteria,** \({\widetilde{r}}_{i}\) | ***l*** | ***m*** | ***u*** | | **SC1,** \({\widetilde{r}}_{1}\) | 3.000 | 3.302 | 3.557 | | **SC2,** \({\widetilde{r}}_{2}\) | 0.747 | 0.928 | 1.145 | | **SC3,** \({\widetilde{r}}_{3}\) | 0.309 | 0.347 | 0.420 | | \[\sum_{i = 1}^{3}{\widetilde{r}}_{i}\] | **4.056** | **4.577** | **5.122** | | \[\left( \sum_{i = 1}^{3}{\widetilde{r}}_{i} \right)^{- 1}\] | **0.195** | **0.218** | **0.247** | | **Sub-Criteria,** \({\widetilde{r}}_{i}\) | ***l*** | ***m*** | ***u*** | | **SC4,** \({\widetilde{r}}_{1}\) | 0.775 | 0.910 | 1.089 | | **SC5,** \({\widetilde{r}}_{2}\) | 1.128 | 1.199 | 1.263 | | **SC6,** \({\widetilde{r}}_{3}\) | 2.321 | 2.596 | 2.823 | | \[\sum_{i = 1}^{3}{\widetilde{r}}_{i}\] | **4.224** | **4.705** | **5.175** | | \[\left( \sum_{i = 1}^{3}{\widetilde{r}}_{i} \right)^{- 1}\] | **0.193** | **0.213** | **0.237** | | **Sub-Criteria,** \({\widetilde{r}}_{i}\) | ***l*** | ***m*** | ***u*** | | **SC7,** \({\widetilde{r}}_{1}\) | 0.314 | 0.362 | 0.465 | | **SC8,** \({\widetilde{r}}_{2}\) | 1.993 | 2.260 | 2.507 | | **SC9,** \({\widetilde{r}}_{3}\) | 2.507 | 2.639 | 2.741 | | \[\sum_{i = 1}^{3}{\widetilde{r}}_{i}\] | **4.813** | **5.261** | **5.712** | | \[\left( \sum_{i = 1}^{3}{\widetilde{r}}_{i} \right)^{- 1}\] | **0.175** | **0.190** | **0.208** | | **Sub-Criteria,** \({\widetilde{r}}_{i}\) | ***l*** | ***m*** | ***u*** | | **SC10,** \({\widetilde{r}}_{1}\) | 0.260 | 0.287 | 0.323 | | **SC11,** \({\widetilde{r}}_{2}\) | 1.836 | 2.213 | 2.565 | | **SC12,** \({\widetilde{r}}_{3}\) | 1.890 | 2.268 | 2.621 | | \[\sum_{i = 1}^{3}{\widetilde{r}}_{i}\] | **3.986** | **4.768** | **5.509** | | \[\left( \sum_{i = 1}^{3}{\widetilde{r}}_{i} \right)^{- 1}\] | **0.182** | **0.210** | **0.251** | | **Sub-Criteria,** \({\widetilde{r}}_{i}\) | ***l*** | ***m*** | ***u*** | | **SC13,** \({\widetilde{r}}_{1}\) | 0.439 | 0.471 | 0.538 | | **SC14,** \({\widetilde{r}}_{2}\) | 0.512 | 0.615 | 0.716 | | **SC15,** \({\widetilde{r}}_{3}\) | 3.377 | 3.653 | 3.915 | | \[\sum_{i = 1}^{3}{\widetilde{r}}_{i}\] | **4.327** | **4.738** | **5.168** | | \[\left( \sum_{i = 1}^{3}{\widetilde{r}}_{i} \right)^{- 1}\] | **0.193** | **0.211** | **0.231** | The process continues by obtaining the Fuzzy Weight for each criterion using Equation (10). Following this, the fuzzy weight values are defuzzified, and then normalized using Equations (11) and (12) respectively. These results are presented in Table 9. From the final values, we can establish a ranking and proceed with the decision analysis related to the criteria. Table 9. Fuzzy Weight, defuzzified and normalized value for all criteria. | | | | | | | ------------ | ----------------------------------------- | --------------------------- | --------------------------- | ---------- | | **Criteria** | \[{\widetilde{\mathbf{w}}}_{\mathbf{i}}\] | \[\mathbf{M}_{\mathbf{i}}\] | \[\mathbf{Z}_{\mathbf{i}}\] | ***Rank*** | | **C1** | (0.381,0.500,0.651) | 0.510 | 0.494 | 1 | | **C2** | (0.073,0.105,0.159) | 0.113 | 0.109 | 4 | | **C3** | (0.120,0.185,0.273) | 0.193 | 0.186 | 2 | | **C4** | (0.039,0.060,0.098) | 0.066 | 0.064 | 5 | | **C5** | (0.116,0.149,0.191) | 0.152 | 0.147 | 3 | | | **TOTAL** | **1.034** | **1.000** | | The same procedure will be applied to the sub-criteria using Equations (10), (11), and (12) to obtain the Fuzzy Weight, defuzzified, and normalized values for the sub-criteria. To determine the overall ranking of the sub-criteria, the normalized value of each criterion is multiplied by the normalized value of each sub-criterion, resulting in the global weight. This global weight is then analyzed to establish the comprehensive ranking of the sub-criteria. All these results are presented in Table 10. Table 10. Fuzzy Weight, defuzzified and normalized value for all sub-criteria. | | | | | | | | | | | ------------ | --------------------------------------- | ------------------ | ----------------------------------------- | --------------------------- | ---------------------------------------- | ------------------------------------------------------------------------------------------------------ | ------------------- | ---------- | | **Criteria** | \[\mathbf{Z}_{\mathbf{i}}\mathbf{(C)}\] | ***Sub-Criteria*** | \[{\widetilde{\mathbf{w}}}_{\mathbf{i}}\] | \[\mathbf{M}_{\mathbf{i}}\] | \[\mathbf{Z}_{\mathbf{i}}\mathbf{(SC)}\] | \[\mathbf{Z}_{\mathbf{i}}\mathbf{(C) \times}\mathbf{Z}_{\mathbf{i}}\left( \mathbf{\text{SC}} \right)\] | ***Global Weight*** | ***Rank*** | | | | SC1 | (0.586,0.721,0.877) | 0.728 | 0.715 | 0.3532 | 0.353 | 1 | | **C1** | 0.494 | SC2 | (0.146,0.203,0.282) | 0.210 | 0.207 | 0.1023 | 0.102 | 3 | | | | SC3 | 0.060,0.076,0.104) | 0.080 | 0.078 | 0.0385 | 0.039 | 7 | | | | SC4 | (0.150,0.193,0.258) | 0.200 | 0.198 | 0.0216 | 0.022 | 11 | | **C2** | 0.109 | SC5 | (0.218,0.255,0.299) | 0.257 | 0.254 | 0.0277 | 0.028 | 10 | | | | SC6 | (0.448,0.552,0.668) | 0.556 | 0.549 | 0.0598 | 0.060 | 6 | | | | SC7 | (0.055,0.069,0.097) | 0.073 | 0.073 | 0.0136 | 0.014 | 14 | | **C3** | 0.186 | SC8 | (0.349,0.430,0.521) | 0.433 | 0.429 | 0.0798 | 0.080 | 5 | | | | SC9 | (0.439,0.502,0.569) | 0.503 | 0.498 | 0.0926 | 0.093 | 4 | | | | SC10 | (0.047,0.060,0.081) | 0.063 | 0.061 | 0.0039 | 0.004 | 15 | | **C4** | 0.064 | SC11 | (0.333,0.464,0.643) | 0.480 | 0.464 | 0.0297 | 0.030 | 9 | | | | SC12 | (0.343,0.476,0.657) | 0.492 | 0.475 | 0.0304 | 0.030 | 8 | | | | SC13 | (0.085,0.099,0.124) | 0.103 | 0.102 | 0.0150 | 0.015 | 13 | | **C5** | 0.147 | SC14 | (0.099,0.130,0.165) | 0.131 | 0.103 | 0.0191 | 0.019 | 12 | | | | SC15 | (0.653,0.771,0.905) | 0.776 | 0.768 | 0.1129 | 0.113 | 2 | | | | | **TOTAL** | | | **1.0001** | **1.000** | | The analysis of Table 9 reveals that Responsiveness (C1) is the most critical factor for users when choosing a courier service in Malaysia, with the highest normalized value \(z_{i}\) of 0.494. This is followed by Empathy (C3) and Tangibility (C5), with normalized values of 0.186 and 0.147, respectively. Reliability (C2) and Assurance (C4) are considered less critical, with values of 0.109 and 0.064. These findings suggest that courier services should prioritize improving responsiveness and empathy to meet user expectations, while also maintaining a focus on reliability and assurance. The consistency in weight distribution using the FAHP method highlights the reliability of these results. The analysis of Table 10 identifies the most and least influential sub-criteria impacting user preferences for courier services in Malaysia. The sub-criteria with the highest global weight is SC1 (Easy contact with the courier company), which has a normalized value of 0.353, indicating its paramount importance. This is followed by SC15 (The goodness of the condition), with a global weight of 0.113, highlighting the significance of the condition in which items are delivered. Other noteworthy sub-criteria include SC9 (Understanding the demand of customers) and SC2 (Efficient communication with the courier company), which rank third and fourth, respectively. This suggests that clear communication and understanding customer needs are also critical factors for users. In contrast, the sub-criteria with the least influence include SC10 (Knowledge and competency of employees), SC7 (Readiness to react quickly to problem interference), and SC13 (Aesthetic and neat appearance of the courier), with the lowest global weights. While these factors are still important, they are less critical compared to responsiveness and the physical condition of deliveries. These insights can guide courier service providers in prioritizing improvements in areas that matter most to users, such as ensuring ease of contact, maintaining the condition of deliveries, and understanding customer needs. However, it is also important to recognize that aspects like employee knowledge and aesthetic appearance, though less critical, should not be neglected entirely. # CONCLUSION The study highlights that responsiveness, empathy, and tangibility are the most critical factors influencing user preferences for courier services in Malaysia, with responsiveness being the highest priority. Conversely, reliability and assurance are less critical but still important. The most influential sub-criteria include ease of contact with the courier company and the condition of delivered items. These findings suggest that courier services should prioritize improving responsiveness and empathy, while maintaining reliability and assurance, to meet user expectations. For future studies, it is recommended to include a larger and more diverse sample size, explore the impact of technological advancements, and conduct comparative and longitudinal analyses. Additionally, integrating customer satisfaction metrics, qualitative insights, and examining the effects of e-commerce growth and environmental factors can provide a deeper understanding of user preferences. Investigating policy implications and conducting global comparisons can also offer valuable insights and help tailor services to evolving demands. # Acknowledgements/Funding (DOUBLE-BLIND reviewing. Leave the section as is. Only include Acknowledgements text in the final submission paper) This section is compulsory. The following is an example of an acknowledgement statement: # Conflict of interest statement (DOUBLE-BLIND reviewing. Leave the section as is. Only include CONFLICT OF INTERESTS text in the final submission paper) # This section is compulsory. The following is an example of a conflict-of-interest statement: # Authors’ contributions Each author contribution must be stated clearly reflecting each contribution to the body of the work and manuscript. Authors can refer to [CRediT](http://credit.niso.org/) (Contribution Roles Taxonomy) for the detailed information about individual contributions to the work. For example ***(Double Blind Review: Leave this section blank until final camera-ready submission)*:** # References Abd Aziz, K. A., Daud, W. S. W., Idris, M. F. I. M., & Saimuddi, S. (2024). Fuzzy Analytical Hierarchy Process for Analyzing the Factors that Influence the Career Choices of Graduate Students at UiTM Perlis. *Journal of Computing Research and Innovation*, *9*(1), 180-196. Abd Aziz, K. A., Daud, W. S. W., Idris, M. F. I. M., & Zakaria, S. N. (2024). The Ranking of Factors in Selecting the Online Shopping Platform in Malaysia based on Fuzzy Analytic Hierarchy Process. *Journal of Computing Research and Innovation*, *9*(1), 31-42. Abd Aziz, K. A., Idris, M. F. I. M., Daud, W. S. W., & Fauzi, M. M. (2023). Application of fuzzy analytic hierarchy process (FAHP) for the selection of best student award. *Journal of Computing Research and Innovation*, *8*(2), 80-90. Al-Shammari, M., & Mili, M. (2021). A fuzzy analytic hierarchy process model for customers’ bank selection decision in the Kingdom of Bahrain. *Operational Research*, *21*(3), 1429-1446. Al-Weshah, G. A., Alnsour, M. S., Al-Hyari, K., Alhammad, F., & Algharabat, R. (2013). Electronic Networks and Relationship Marketing: Qualitative Evidence from Jordanian Travel Agencies. *Journal of Relationship Marketing*, *12*(4), 261–279. https://doi.org/10.1080/15332667.2013.846245 Ayhan, M. B. (2013). A Fuzzy AHP Approach for Supplier Selection Problem: A Case Study In A Gearmotor Company. *International Journal of Managing Value and Supply Chains, 4*(3), 11. Azam, M., Qureshi, M. R. N. M., & Talib, F. (2017). Quality evaluation of health care establishment utilizing fuzzy AHP. *International Journal of Service Science, Management, Engineering, and Technology (IJSSMET), 8*(4), 83-120. Bhatt, N., Guru, S., Thanki, S. et al. (2021). Analyzing the factors affecting the selection of ERP package: a fuzzy AHP approach. *Inf Syst E-Bus Manage 19*, 641–682. https://doi.org/10.1007/s10257-021-00521-8 Buckley, J. J., & Eslami, E. (2002). An introduction to fuzzy logic and fuzzy sets (Vol. 13). *Springer Science & Business Media*. Büyüközkan, G., Çifçi, G., & Güleryüz, S. (2011). Strategic analysis of healthcare service quality using fuzzy AHP methodology. *Expert systems with applications, 38*(8), 9407-9424. Chang, D.Y. (1996) Applications of the Extent Analysis Method on Fuzzy AHP. *European Journal of Operational Research, 95*, 649-655. http://dx.doi.org/10.1016/0377-2217(95)00300-2 Chou, S. W., & Chang, Y. C. (2008). The implementation factors that influence the ERP (enterprise resource planning) benefits. *Decision support systems, 46*(1), 149-157. Deng, H. (1999). Multicriteria analysis with fuzzy pairwise comparison. *International journal of approximate reasoning, 21*(3), 215-231. Dutta, B., Singha, T., Goh, M., Lamata, M. T., & Verdegay, J. L. (2019). Post factum analysis in TOPSIS based decision making method. *Expert Systems with Applications, 138*, 112806. El Saghier, N., & Nathan, D. (2013). Service Quality Dimensions and Customers' Satisfaction of Banks in Egypt. *In Proceeding of 20th International Business Research Conference*. Erensal, Y. C., Öncan, T., & Demircan, M. L. (2006). Determining key capabilities in technology management using fuzzy analytic hierarchy process: A case study of Turkey. *Information Sciences, 176*(18), 2755-2770. Feng, C. (1995). Fuzzy multicriteria decision-making in distribution of factories: an application of approximate reasoning. *Fuzzy Sets and Systems, 71*(2), 197-205. Gruenwald, H. (2020). Covid-19 and Logistics Articles Revisited. *Information Management and Business Review*. Idris, M. F. I. M. (2020). Determining the causes of divorce in Perlis using fuzzy analytic hierarchy process. *Jurnal Intelek, 14*(1), 56-65. Idris, M. F. I. M., Abd Aziz, K. A., & Abd Aziz, N. S. F. (2023). Selecting the effective ways to prevent covid-19 from spreading using fuzzy AHP method. *Journal of Computing Research and Innovation, 8*(2), 112-123. Janahi, M.A. and Al Mubarak, M.M.S. (2017). The impact of customer service quality on customer satisfaction in Islamic banking. *Journal of Islamic Marketing, 8*(4), 595-604. https://doi.org/10.1108/JIMA-07-2015-0049 Jeganathan, C. (2003, March). Development of fuzzy logic architecture to assess sustainability of the forest management. ITC. Lee, A.H., Chen, W., & Chang, C. (2008). A fuzzy AHP and BSC approach for evaluating performance of IT department in the manufacturing industry in Taiwan. *Expert Syst. Appl., 34*, 96-107. Liu, L.R., & Liu, C. (2014). Empirical Study of Express Logistics Service Quality –A Survey of Changdao County Express Sector. Mikhailov, L., & Tsvetinov, P.E. (2004). Evaluation of services using a fuzzy analytic hierarchy process. *Appl. Soft Comput., 5*, 23-33. Mistarihi, M. Z., & Magableh, G. M. (2023). Prioritization of Supply Chain Capabilities Using the FAHP Technique. *Sustainability 2023, 15*, 6308. Naik, C. K., Gantasala, S. B., & Prabhakar, G. V. (2010). SERVQUAL, customer satisfaction and behavioural intentions in retailing. *European Journal of Social Sciences, 17*(2), 200-213. Noble, C.H., Sinha, R.K. and Kumar, A. (2002). Market Orientation and Alternative Strategic Orientations: A Longitudinal Assessment of Performance Implications. *Journal of Marketing, 66*, 25-39. https://doi.org/10.1509/jmkg.66.4.25.18513 Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1994). Alternative scales for measuring service quality: a comparative assessment based on psychometric and diagnostic criteria. *Journal of retailing, 70*(3), 201-230. Shahin, A., & Chan, J. F. L. (2006). Customer requirements segmentation (CRS): A prerequisite technique for quality function deployment (QFD). *Total Quality Management & Business Excellence, 17*(5), 567–587. https://doi.org/10.1080/14783360600588117 Siekelova, A., Podhorska, I., & Imppola, J. J. (2021). Analytic hierarchy process in multiple–criteria decision–making: a model example. In SHS web of conferences (Vol. 90, p. 01019). *EDP Sciences*. Statista Research Department (2022). Number of domestic courier items delivered by courier service providers in Malaysia from 2016 to 2021. *Statista.com*. Sureshchandar, G. S., Rajendran, C., & Anantharaman, R. N. (2002). The relationship between service quality and customer satisfaction–a factor specific approach. *Journal of services marketing, 16*(4), 363-379. Tabassum, R. and Badiuddin, A. (2014). Measuring the Service Quality Gap in Courier Industry. *IRJA-Indian Research Journal. 1*(5), 1-11. Valaei, N., Rezaei, S., & Shahijan, M. K. (2016). CouQual: assessing overall service quality in courier service industry and the moderating impact of age, gender and ethnicity. *International Journal of Management Concepts and Philosophy, 9*(2), 144-169. Van Laarhoven, P.J.M. and Pedrycz, W. (1983) A fuzzy extension of Saaty’s priority theory. *Fuzzy Sets and Systems, 11*, 229-241. Wang, Y. M., Luo, Y., & Hua, Z. (2008). On the extent analysis method for fuzzy AHP and its applications. *European journal of operational research, 186*(2), 735-747. Yee, H. L., & Daud, D. (2011). Measuring customer satisfaction in the parcel service delivery: a pilot study in Malaysia. *Business and Economic Research, 1*(1), 1-10. Zadeh, L.A. (1965). Fuzzy Sets. *Information and Control, 8*, 338-353. http://dx.doi.org/10.1016/S0019-9958(65)90241-X Zaid, F., Gazder, U., & Barbieri, D. M. (2024). Multi-criteria analysis for freight transport decision-making with fuzzy analytic hierarchy process: A top management’s perspective for Bahrain. *Transportation Research Interdisciplinary Perspectives, 23*, 101017. Zahid, K., Akram, M., & Kahraman, C. (2022). A new ELECTRE-based method for group decision-making with complex spherical fuzzy information. *Knowledge-Based Systems, 243*, 108525

A picture containing text, clipart Description automatically generated

 © 2024 by the authors. Submitted for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
1. \* Corresponding author. *E-mail address*: (Add the e-mail in the final camera-ready submission)