Usability evaluation of start-up business platform using fuzzy SWARA–WASPAS: a case of Indonesian e-commerce

The high number of Internet and smartphone users is used by e-commerce actors to create applications to facilitate the trade sector and technology services (Liang & Lai, 2000). Electronic Commerce (e-commerce) is defined as a commercial transaction through a telecommunications network between two or more parties to obtain a mutually beneficial relationship (Beyari, 2021). Savings and convenience are the main reasons for the increasing interest in online shopping for customers (Li & Zhang, 2002).

One of the problems in the use case of e-commerce applications is usability. Usability is defined as “a benchmark for a product that is used by certain users in order to obtain a goal with time flexibility and satisfaction within the scope of usability” (Abran, Al-qutaish, & Cuadrado-gallego, 2006). Fauzi, Az-zahra, and Kharisma (2019) stated that the problems and complaints contained in the Google PlayStore application reviews along with incompatibilities had an impact on usability by new users and old users who tried to optimize all existing features. Solutions that can be used lead to multi-criteria decision-making (MCDM). The MCDM approach is involved in facilitating the process of evaluating and analyzing various alternatives that exist in real-world situations qualitatively and quantitatively, with environments that have various risks and uncertainties to find the appropriate choice among several options (Adepoju, Oyefolahan, Abdullahi, & Mohammed, 2019). Based on previous research, the MCDM variant for evaluating usability in an application, especially e-commerce, is still limited. Recent studies have emphasized that usability-related perceptions significantly influence user engagement and platform effectiveness in digital commerce environments, highlighting usability as a critical determinant beyond purely technical performance (Ünver, Aydemir, & Alkan, 2023).

The e-commerce applications that will be the object of research are platform S and platform P. These two e-commerce applications were selected based on website and social media performance data as well as surveys of customer users as of quarter 2022, namely platform P in first place and platform S in second place. The purpose of this study is to find out whether users choose based on usability or other factors. Fuzzy SWARA and WASPAS integration methods are used to determine weight and ranking on usability criteria. The SWARA method is widely used because it is very close to the human mind and is also easy to understand for beginners and experienced users (Mardani et al., 2017). The WASPAS approach is used because it is more accurate than other approaches because it uses ranking techniques, so it can make optimal decisions (Zavadskas, Antucheviciene, Hajiagha, & Hashemi, 2014). The selection of the Fuzzy SWARA and WASPAS methods is based on their complementary strengths in handling decision-making problems involving subjective human judgment and multiple conflicting criteria. SWARA is particularly advantageous for eliciting expert knowledge and prioritizing criteria because it requires fewer pairwise comparisons and reduces cognitive load, making it more intuitive and efficient than traditional AHP-based methods. It is especially useful when expert consensus is derived linguistically, as in usability evaluation. Several prior studies on e-commerce evaluation still rely on isolated usability indicators or single-method approaches, which may not sufficiently capture the multidimensional nature of user decision-making in online platforms (Alkan, Küçükoglu, & Tutar, 2021).

On the other hand, WASPAS combines the strengths of two well-established decision models – Weighted Sum Model (WSM) and Weighted Product Model (WPM) – to improve ranking accuracy and reliability. The hybridization of these two approaches allows for a more nuanced evaluation of alternatives by balancing additive and multiplicative reasoning. The integration of fuzzy logic with both SWARA and WASPAS further enhances their ability to accommodate uncertainty and vagueness in user perceptions, which is critical when assessing subjective criteria like usability. Compared to existing studies that rely on a single-method framework, this study introduces a more holistic and robust evaluation model that is better aligned with the complexities of real-world user experiences. Recent findings suggest that integrating multiple decision-making techniques provides more reliable insights into user preferences, particularly when subjective judgments and uncertainty are involved in usability assessment (Ünver & Alkan, 2021).

In analyzing and rating the usability of an e-commerce application, this study uses five criteria, namely trust, shopping support, efficiency, design and satisfaction. The usability attributes were selected based on 19 usability factors in the research proposed by Masudin and Saputro (2016), for design criteria obtained from research (Delice & Güngör, 2009) and satisfaction criteria contained in research (Kasali, Adekola, Akinyemi, Ebo, & Balogun, 2020). This study tries to evaluate and rank the usability factors of e-commerce applications with similar characteristics using the Fuzzy SWARA–WASPAS integration method.

This study is motivated by the growing complexity of user experience (UX) in digital commerce ecosystems, where usability is not only a determinant of user satisfaction but also a key differentiator in a saturated market. While traditional usability evaluations focus on surface-level design or task completion metrics, this research aims to go beyond by applying a hybrid MCDM approach that captures the nuanced preferences of experienced users. The integration of Fuzzy SWARA and WASPAS provides a structured yet flexible framework to quantify usability attributes from a user-centered perspective under uncertainty.

The main contribution of this study lies in (1) offering a hybrid usability evaluation framework that combines subjective weighting and objective ranking under fuzzy conditions, (2) validating the robustness of this approach through comparative analysis with established MCDM techniques and (3) providing practical insights for platform developers to prioritize usability improvements based on user-perceived value. This study also fills a gap in the literature where hybrid fuzzy MCDM models remain underutilized in e-commerce usability assessments.

This study introduces a methodological innovation by integrating the Fuzzy SWARA and WASPAS approaches to evaluate usability factors in e-commerce applications – a combination that has not been widely applied in this domain. Unlike prior research that often uses conventional MCDM methods such as Fuzzy AHP or TOPSIS in isolation, this paper contributes a hybrid approach that capitalizes on the intuitive weighting of SWARA and the robust ranking power of WASPAS. Moreover, the study applies this framework to compare two leading Indonesian e-commerce platforms based on real user feedback, offering context-specific insights that are currently underrepresented in the literature. This combination of methodological integration and localized empirical application constitutes a significant advancement in usability evaluation research.

Evaluation regarding application usability has been carried out by several previous researchers. Previous research related to usability was carried out in various types of e-commerce applications, starting from Shopee, Platform P, Bukalapak, Lazada, Amazon, etc. (Ishak et al., 2021). Amazon, Flipkart, Big basket and Paytm are research objects (Jain & Purandare, 2021). In this study, there are four usability testing factors including functional performance, accessibility, connectivity and readability. The study also stated that usability testing should be considered a critical phase of the SDLC (software development life cycle) when developing an application (Kumar & Hasteer, 2017). The research was conducted by distributing questionnaires that targeted all users who had been involved in online shopping, regardless of limitations age, profession, city, etc. The results of this study indicate that customer service and application reliability are the factors that most influence the usability of application selection.

In addition, research related to usability was also carried out by Kasali et al. (2020) to rank the usability of health applications in Nigeria. The MCDM technique is used by adopting the Enhanced Usability Model (EUM) which was designed based on the People at the Center of Mobile Application Development (PACMAD) and the Integrated Measurement Model (IMM). The evaluation results show that effectiveness and efficiency have the highest priority with 40% and 33% (Rekik, Kallel, Casillas, & Alimi, 2016). Whereas satisfaction, user interface aesthetics and universality rank the lowest.

The research raised in this final project also examines the usability evaluation of applications, especially in e-commerce. E-commerce which is used as a research object has a track record as first and second place based on the official website of the ministry of communication and information. Both of these e-commerce sell various types of products that tend to be the same, and researchers want to know and also evaluate, whether the application selection is based on usability factors.

Scholtz (2006) defines usability as a quality attribute so that it can improve the design process of a system. Usability has five quality attributes including learnability, memorability, efficiency, errors and user satisfaction. One attribute does not rule out being more important than other attributes (Scholtz, 2006). In the research of Masudin and Saputro (2016), the usability attributes used are trustworthiness, shopping support, information access efficiency, ease of apprehension and hedonic quality. Whereas in Delice and Güngör (2009), the usability attributes used to analyze websites are design consideration, operation of website and website user according. According to ISO, the usability components used to evaluate a system are effectiveness, efficiency and satisfaction. So this study tries to adopt the usability attributes used in previous studies. In this study, the usability components that will be used to evaluate the platform P and platform S applications are trust, shopping support, efficiency, design and satisfaction. The trust factor is needed to analyze usability in e-commerce applications, because customer trust has a positive impact on customer loyalty and satisfaction (Hoffman, Novak, & Peralta, 1999). Factors of customer trust can also be formed by guaranteeing information security and privacy, product security, security in transactions, also in the delivery and tracking process (Roy Dholakia & Zhao, 2010) and Security guarantees are also an attempt to convince customers to continue using online shopping applications (Paramartha, Fortwonatus, & Setyohadi, 2021).

The shopping support attributes used in this study are in the form of feedback, ease of accessing applications, ease of transactions and assistance features. This refers to previous research, that communication, order fulfillment, speed for transactions, internal activities and quality in service can increase customer buying interest (Shih & Fang, 2006). Efficiency components are also needed when accessing an e-commerce application such as the time it takes to display the main menu and the navigation buttons are easy to find. In the research of Paramartha et al. (2021) effectiveness and efficiency also affect customer satisfaction in using an application.

Design components also need to be considered in terms of usability such as the display on the main screen, the layout of the application and detailed product information. Details for product information are very important because they can be the basis for measuring the quality of a product (Paramartha et al., 2021). Consumers can also make purchasing decisions through the detailed information provided, due to the characteristics of non-physical goods. Information is a prerequisite for building trust in consumers (Casaló, Flavián, & Guinalíu, 2008). Application components such as the main screen appearance, layout and information accuracy have a significant relationship to application user trust (Yoon, 2002). In addition, the last attribute used is satisfaction. Satisfaction (satisfaction) is the level of comfort and pleasure felt by the user through the use of applications or software. This can be illustrated from user behavior toward an application, usually measured subjectively by distributing questionnaires and qualitative techniques (Harrison, Flood, & Duce, 2013). Table 1 shows usability attributes in related studies.

SWARA was first introduced in a study by Keršuliene, Zavadskas, and Turskis (2010). This research focuses on solving rational dispute problems by making the best decision among the existing alternatives. The algorithm used in the SWARA method is simple and very close to the human mind and there is no need to check the consistency of judgments because it is ensured when the criteria are sorted in descending order (Mardani, Zavadskas, Khalifah, Jusoh, & Nor, 2016). This method is an appropriate technique for solving priority problems that have been previously identified, based on conditions. SWARA is able to predict the opinion of experts based on the ratio of interests in each criterion and weight (Turskis, Goranin, Nurusheva, & Boranbayev, 2019).

In research related to usability, the SWARA method has not been widely used. However, in the research of Zolfani, Salimi, Maknoon, and Kildiene (2015) SWARA is used to make decisions to determine the development of shopping areas with a forward-looking perspective. Meanwhile, in Zolfani and Saparauskas (2013) applied SWARA in selecting machine tools. In addition, a research study using the SWARA method was also conducted by Zolfani and Saparauskas (2013) to investigate the success factors of explorer-based online games. This method also provides opportunities for policymakers to make decisions based on different situations and prioritize criteria based on the needs and desired goals (Karabasevic, Paunkovic, & Stanujkic, 2016). This means that in research related to usability, it is also important to use the SWARA method to determine choices in evaluating criteria and weights to determine the use of e-commerce applications based on usability or other factors (Stanujkic, Karabasevic, & Zavadskas, 2015).

Keršuliene et al. (2010) introduced a new method used to assess the weight of an evaluation criterion, namely SWARA. As the level of decision-making develops, an approach is proposed with fuzzy sets and the SWARA method. FSWARA is applied in decision-making to evaluate weights and criteria (Attri, Singh, Dhar, & Powar, 2022). The advantage of using a fuzzy approach is that it gives the relative importance of attributes using fuzzy numbers rather than exact numbers. Fuzzy sets are versatile tools that can be used for linguistic and numerical modeling (Pinto, 2014). Even though fuzzy SWARA is a new technique, various researchers have applied it when solving problems related to MCDM, namely in selecting outsourcing and evaluating construction equipment with sustainability considerations (Perçin, 2019).

Fuzzy WASPAS is a development of WASPAS and the fuzzy method proposed by Turskis, Zavadskas, Antucheviciene and Kosareva (2015) and is one of the MCDM approaches which is considered the most accurate in making decisions on complex problems. This method is a combined combination of the Weighted Product Model (WPM) and the Weighted Sum Model (WSM) (Zavadskas, Turskis, Antucheviciene, & Zakarevicius, 2012). Chakraborty and Zavadskas (2014) proposed seven steps for completing the FWASPAS method. WASPAS enables decision makers to assess and rank alternatives with a high degree.

Research using hybrid fuzzy SWARA–WASPAS is a new method approach that has not been widely used in usability cases. Decision-making techniques that are popularly used in usability cases are fuzzy AHP, fuzzy TOPSIS, fuzzy Chang, etc. Zavadskas et al. (2014) used AHP fuzzy for usability research on learning websites with ease of us, response time, navigable and informative factors. Whereas in the research of Masudin and Saputro (2016) using fuzzy TOPSIS to evaluate B2C websites. In this study, the usability factors used were trustworthiness, shopping support, information access efficiency, ease of apprehension, hedonic quality, effectiveness, safety and flexibility. This research uses Amazon.com and hepsiburada.com as research objects.

The integration of the SWARA fuzzy model is used in this study to estimate the weight of the criteria based on the assessment of the respondents. The results of the respondents' assessments were converted into fuzzy numbers which were originally linguistic variables (Ren, Liao, Al-barakati, & Cavallaro, 2019) and using fuzzy WASPAS to optimize ranking from highest to lowest value based on performance appraisal (Mishra & Rani, 2018).

The researcher proposes the integration of the SWARA fuzzy approach in the case of e-commerce application usability to determine the weight of e-commerce application criteria, and WASPAS to rank application usage based on its usability, whether it is in accordance with the user's choice (Agarwal, Kant, & Shankar, 2020). This study also adopts the criteria in the usability case, namely Trust, Shopping Support, Efficiency and Design.

In summary, while prior studies have explored various usability dimensions across different application domains using models like PACMAD, IMM or ISO-based frameworks, they often rely on conventional MCDM techniques or descriptive approaches that do not fully account for the uncertainty and subjectivity inherent in user evaluations. Furthermore, few studies attempt to combine expert judgment with user-centric ranking in a robust, interpretable way. This study addresses these gaps by integrating Fuzzy SWARA for flexible criteria weighting with Fuzzy WASPAS for stable, dual-mode ranking – applied specifically to leading Indonesian e-commerce platforms. This approach not only advances usability assessment methodology but also responds to the need for localized, data-driven decision support in the evolving digital commerce landscape.

The step in this study starting with conducting a literature study which will be used as material for theoretical studies and sources of information. The second step to determine the object of research. In this study, the objects used were platform P and platform S which are e-commerce applications with the most users based on official website data and social media performance. The third step is to determine the formulation of the problem. The fourth step is setting research objectives, so that the results obtained are in accordance with the objectives to be achieved. The fifth step is selecting usability attributes for evaluation materials. Attributes are selected based on literature studies and have been adjusted to the research object. Usability attributes are in Table 2.

The sixth step is to prepare a questionnaire. The questionnaire was prepared based on the results of selecting usability evaluation criteria. The questionnaire is presented using a linguistic scale. Fuzzy weighting criteria and sub-criteria consist of Very High (VH), High (H), Enough (E), Low (L), Very Low (VL). Then in the seventh step, the distribution of questionnaires was carried out. The respondents selected in this study were users of the platform P and platform S applications. Where respondents have used both applications at least once. Respondents will fill out a questionnaire through the Google Form media. Because in this study the population size was unknown, the sample size was determined using the Bernoulli formula, and the results obtained were 100 respondents.

The eighth stage is to calculate the weight of the criteria using F – SWARA. The following are the steps for calculating the weight of criteria and sub-criteria using Fuzzy SWARA (Agarwal et al., 2020):

1. Rank the evaluation criteria in descending order of expected significance. Ranking is based on the level of importance, namely the most significant (influential) is given the first rank, and criteria that are not significant are determined as the final ranking.

2. Look for the value of comparative importance (Sj). By using a fuzzy comparison scale (fuzzy comparison scale) to determine the relative score criteria. The following is a fuzzy comparison scale to assess the evaluation criteria based on (Table 3).

3. Calculate the comparative coefficient for each criterion, using Equation (7)

1. Determine the recalculated weighting factors (q_j) ̃ using Equation (8)

1. Calculate the relative importance weight using the equation of the evaluation criteria using Equation (9)

where $w~j$ represents the relative weight of the criteria and n represents the number of criteria

1. To find out the normal weight, defuzzification and normalization can be done using Equation (10)

The ninth step is calculating alternative weights using WASPAS. Following are the steps for calculating the weight of criteria and sub-criteria using Fuzzy WASPAS [12]:

1. Making a decision matrix to determine alternative rankings using Equation (11)

1. Normalization of the X matrix. Through the following two equations:

Equation (12) for maximum assessment criteria and Equation (13) for minimum assessment criteria .

• For maximum assessment criteria

• For minimum assessment criteria

1. Calculate the weighting matrix to be normalized by Fuzzy Waspas for the sum. In Equation (14)

1. Calculate the weighting matrix of the decision making is normalized by Fuzzy Waspas for the multiplication department. In Equation (15)

1. Defuzzification from the calculation of fuzzy using the Center of Area (COA) method, which is the most practical method. In Equation (16). Defuzzification serves to normalize the fuzzy value.

1. Calculate the Qi aggregation value of the weeds of addition and multiplication with Equation (17). Alternative selection can be sorted through Qi value, which is the top alternative that has the maximum QI value, and is a parameter of Fuzzy WASPAS.

Then the tenth step is to determine the ranking of alternative usability in the e -commerce application. Alternative ranking is the final result of data processing where the best ranking is obtained between the platform P or platform S application based on its usability. This alternative rating was obtained after calculating using Fuzzy SWARA–WASPAS.

After data processing using Fuzzy SWARA–WASPAS, it can be seen which e-commerce application has the best usability with the highest weight value and top rank. And can be given recommendations at the end of writing the final project. Then in the last stage a conclusion is obtained.

From the results of the recapitulation of assessment part 1 of the platform P application, an assessment with the highest number of respondents from each sub-criteria was taken. The assessment results are used to calculate the weight of each sub-criteria that has been ranked based on the results of the questionnaire assessment (see Table 4).

After getting the weighting results from the calculation of Fuzzy Swara, then the value will be calculated to determine the ranking of each criterion. Table 5 shows the value of the decision matrix of each criterion and sub-criteria is obtained from the results of the assessment of part 2 of the platform P application.

Normalization of the decision matrix is done by adding the 0 number behind the comma, this is to clarify the results of the assessment that will appear at the end. Table 6 shows normalization of the decision matrix results.

After normalization, then calculate the normalized matrix for the summarizing part. Table 7 shows the results of the normalized matrix.

After doing the calculations for the summarizing part, then calculate the normalized matrix for the multiplication part. Table 8 shows results of Normalization of Decision Matrix.

The last stage that must be calculated is the result of the calculation of Fuzzy WASPAS and the Ranking Criteria of Usability from the platform P application. The following are the results of the calculation. Table 9 shows the results of platform P usability ranking.

Table 10 shows the result of a consistency test between defuzzified relative weight $wj$ with relative fuzzy weight $wjˆ$ on platform P Application:

From Table 10, it can be seen that the results obtained from the calculation of Fuzzy WASPAS show that the Shopping Support (C2) criteria are the highest ranking criteria, and most importantly to support the usability of the platform P application. Shopping support attributes used in this study in the form of ease for checkouts, assistance features and ease of payment. This shows that the platform P application is in great demand in terms of shopping support. Whereas in the research Masudin and Saputro (2016) using the weighting of Fuzzy AHP showed the results of the Shopping Support criteria as the last rank in successive order, namely the criteria for trust, loading time, ease of transaction and shopping support. Trust criteria (C1) shows the second highest rank in the usability attribute. The trust attributes used are security privacy, confirmation, payment, insurance and tracking. The trust factor greatly affects customer loyalty and satisfaction in using platform P (Hoffman et al., 1999). In research Masudin and Saputro (2016) using the weighting of Fuzzy AHP showed the results of the confidence criteria as the first rank in the sequence of criteria namely trust, loading time, ease of transaction and shopping support.

Then in the third rank there is a satisfaction criterion (C5), which consists of feedback and comfort (comfort). From ISO research the satisfaction criteria occupy the last position in the sequence of effectiveness, efficiency and satisfaction. Satisfaction becomes a benchmark for the level of comfort and pleasure of the user through a software. This can be measured subjectively and varies between users (Harrison et al., 2013). In the fourth place, there are efficiency criteria consisting of time efficiency and easy to navigate. From ISO research, the efficiency criteria rank second in the sequence of effectiveness, efficiency and satisfaction. Efficiency reflects the productivity of an application in completing its tasks accurately and quickly (Harrison et al., 2013) and in fifth place there are design criteria consisting of sub-criteria home page, application layout and product information. In research Delice and Güngör (2009) design criteria become the most important, in the sequence of criteria namely design consideration, operation of website and user accordance website. This refers to Yoon (2002), namely application components such as the main screen display and layout design have a significant relationship on the usability of an application. The dominance of trust- and support-related criteria observed in this study aligns with recent empirical findings, which indicate that user trust and system support functions remain central in determining perceived usability across digital platforms (Alkan, Güney, & Kilinç, 2023).

Based on the ranking that has been obtained, the efficiency criteria and design criteria are ranked two lowest, which means an evaluation of the usability attributes is needed. Because the efficiency of an application is a very influential thing for users, the faster the application operates the higher the efficiency (Bandyopadhyay & Sen, 2011). And also on the design criteria need to be considered such as display on the main screen, application layouts and product information details must be displayed clearly. And based on the results of the test, it can be stated that the ranking sequence between Defuzzified relative weight $wj$ and relative fuzzy weight $wjˆ$ is consistent, because it has the same ranking sequence.

From the results of the evaluation recapitulation of part 1 of the platform S application, the assessment with the highest number of respondents was taken from each sub-criteria. The results of this assessment are used to calculate the weight of each sub-criteria that has been ranked based on the results of the questionnaire assessment. So that the results are shown in Table 11:

After getting the weighting results from the SWARA Fuzzy calculation, then the value will be calculated to determine the ranking of each criterion. Following are the WASPAS fuzzy data processing steps:

The last step that must be calculated is the result of the Fuzzy WASPAS calculation and the usability criteria ranking of the platform P application. Table 12 shows the results of the calculations.

Table 13 shows the result of a consistency test between defuzzified relative weight $wj$ with relative fuzzy weight $wjˆ$ on platform S application:

From Table 13, it can be seen that the results obtained from the WASPAS fuzzy calculations show that the trust criterion (C1) is the criterion with the highest rating and is the most important for supporting the usability of the platform S application. The trust attributes used in this study are security privacy, confirmation, payment, insurance and tracking. This is in accordance with research Masudin and Saputro (2016) using AHP fuzzy weighting which shows the results of the trust criteria as the first rank in a row, namely the criteria of trust, loading time, ease of transaction and shopping support. Customer trust factors can be formed through guarantees of information security and privacy, product security, transaction security as well as shipping and tracking processes provided by the platform S application (Wolfinbarger & Gilly, 2003). The second rank is occupied by efficiency criteria, which consist of time efficiency and easy to navigate sub-criteria. From ISO research the efficiency criterion occupies the second position in the order of effectiveness, efficiency and satisfaction. Which means that the efficiency criteria have the same ranking order as the research conducted by ISO. Based on reviews on the Google Playstore, this usability attribute can affect customer loyalty in using the application, so that it will affect the rating (rating) given and will have an impact on other customers who will try to start using an application (Bandyopadhyay & Sen, 2011).

Then in the third place the platform S application is occupied by the Shopping Support criteria which consists of the sub-criteria easy to checkout, help and support, and payment tools. In research Masudin and Saputro (2016) using AHP fuzzy weighting, the results of the shopping support criteria were shown as the last rank in a row, namely the criteria of trust, loading time, ease of transaction and shopping support. This criterion is ranked third because based on reviews and complaints submitted by users in the Google Playstore, it is very difficult to find a help feature when there are problems related to the application. In the fourth rank, there are design criteria consisting of sub-criteria for the home page, application layout and product information. There is a difference between this study and research design criteria being the most important, in order of criteria namely design consideration, operation of website and website user according (Delice & Güngör, 2009). This corresponds to the user experience when using the application and the interactions that occur within it. When the design displayed is very user friendly, the design can be used as a reference to measure the usability of an application (Masudin, Aprilia, Nugraha, & Restuputri, 2021a). Then in the last ranking, there is a satisfaction criterion (C5), which consists of feedback and comfort. From ISO research, the satisfaction criterion occupies the last position in the order of effectiveness, efficiency and satisfaction. Satisfaction is a benchmark for the level of comfort and enjoyment of users through a software. This can be measured subjectively and varies between users (Harrison et al., 2013).

The design and satisfaction criteria are ranked the bottom two, which means that an evaluation is needed regarding the application, especially on the design and satisfaction criteria. For design criteria, it is necessary to consider the re-layout of the application and detailed product information must be more detailed. For satisfaction criteria, it is necessary to consider the level of comfort and pleasure felt by the user through using the application. So that it can provide feedback in accordance with the usability that has been given. And based on the results of the test it can be stated that the ranking sequence between Defuzzified relative weight $wj$ and relative fuzzy weight $wjˆ$ is consistent, because it has the same ranking sequence.

Based on the results of the consistency test that has been done before, a ranking result has been obtained for each application. These results will be compared to prove that users use the application because of usability factors. Here are the comparisons:

Based on Table 14, it can be seen that the ranking comparison of the usability criteria in the platform P and platform S applications. The results show that the trust criterion (C1) in the platform P application is ranked 2, while the platform S application is ranked 1. Then the shopping support criterion (C2) is ranked 1 in the platform P application while the platform S application is ranked 3. This also refers in research Masudin and Saputro (2016) where the ranking criteria used were trust, shopping support, information access efficiency, ease of apprehension and hedonic quality. This difference refers to the method used. This study uses the SWARA–WASPAS fuzzy approach to give weights and ratings, while research Masudin and Saputro (2016) uses the AHP – TOPSIS fuzzy. For efficiency criteria, the platform P application is ranked 4, while the platform S application is ranked 2. In research Jain and Purandare (2021), efficiency is ranked third, where there are differences from this study. For design criteria (C4) it gets a rating of 5 on the platform P application and gets a rating of 4 on the platform S application, where in research Delice and Güngör (2009) places design criteria in first place. Then finally the satisfaction criterion (C5), the platform P application gets a rating of 3, while the platform S application gets a rating of 5. This is consistent with ISO research which places the satisfaction criteria at the last place in the two criteria used, namely effectiveness and efficiency. Prior studies have shown that usability dimensions such as efficiency, trust and interface clarity play a mediating role in shaping user satisfaction and continued usage intention, reinforcing the behavioral relevance of the present findings (Tutar, Küçükoğlu, Özdemir, Alkan, & Ipekten, 2024).

Then seen from the Qi value of each – each criterion which has a significant difference. This can also be a differentiating factor from the ranking resulting from each criterion. Based on these results, it can be analyzed that each application has its own level of usability, based on the criteria used. And users can use applications according to their needs and desires. Methodologically, this study extends recent work that advocates for advanced decision-support models by demonstrating that hybrid fuzzy MCDM approaches can effectively address complexity and uncertainty in usability evaluation contexts (Kabakuş, Ünver, Çelik, & Alkan, 2025).

Managerial implications are intended to develop managerial policies that are expected to be able to provide a theoretical contribution to management practice (Masudin et al., 2021a). This study proposes a SWARA–WASPAS fuzzy to determine the weight and ranking of the criteria and sub-criteria used in usability. The proposed framework is very important for users, practitioners and researchers in making effective use of e-commerce applications. The ranking of the criteria carried out is useful for knowing the components used to support the usability of an e-commerce application in making even better components. The results of this study become information for users and help in choosing applications that are suitable for use for their interests and needs more effectively. Researchers provide suggestions for application management to be able to improve usability criteria that influence respondents in choosing e-commerce applications:

1. Customer trust can be established by providing guarantees of information and privacy security, product security, transaction security and providing security in delivery and tracking (Roy Dholakia & Zhao, 2010; Mukherjee & Nath, 2007). Therefore, researchers suggest that management can organize and improve network security in existing e-commerce applications. According to Hoffman et al. (1999) customer trust is a very influential thing on customer loyalty and satisfaction.

2. Detailed information of a product or service can be accessed quickly and easily. As explained in the previous section, the speed and ease of access to information is very important because each user has their own level of activity, where the time spent accessing an application is very valuable. Radio Frequency Identification Technology can help provide information more effectively and efficiently because it uses a large memory capacity and can perform automatic scanning simultaneously. RFID makes it possible to help the availability of easily accessible information systems (Masudin, Ramadhani, Restuputri, & Amallynda, 2021b).

3. The importance for e-commerce applications to provide satisfaction for users when using the application. The level of comfort and enjoyment that users feel can be reflected in the user's attitude towards the application (Harrison et al., 2013). Researchers suggest management to improve the service quality of e-commerce applications, among others, by providing a quick response in resolving problems and complaints given by customer (Sharma & Lijuan, 2015; Sheu & Chang, 2022)

Theoretically, the use of the SWARA–WASPAS method in evaluating usability is used to optimize the use of e-commerce applications, so that there will be significant improvements in the criteria that are in the lowest order. The usability attribute developed using the integration of the SWARA–WASPAS fuzzy approach is considered as one of the most accurate and efficient approaches. This is in line with research Agarwal et al. (2020) which seeks to streamline the provision of solutions with ranking results generated from the SWARA–WASPAS method. By considering this study, users and researchers can consider that existing e-commerce applications have their own level of usability. Ratings from Google Playstore can also be used as reference material for users before deciding to use certain e-commerce applications. Because high assessment attributes can provide indicators of the ability and quality of an application (Harrison et al., 2013).

The managerial implications of this study are closely tied to the prioritization of usability criteria identified using the Fuzzy SWARA–WASPAS approach. The finding that Shopping Support and Trust are consistently ranked highest indicates that managers should invest in improving transaction ease, support responsiveness and security features. These elements are not only key drivers of user satisfaction but also crucial for differentiating platforms in a competitive e-commerce market. Managers can use the relative weights generated in this study to allocate resources more efficiently toward the usability dimensions that users value most. For instance, platforms scoring low on Design or Efficiency should prioritize UI/UX redesign and system optimization to improve performance on lower-ranked but still important criteria.

From a strategic perspective, this study provides a structured decision-support model that platform managers can replicate when evaluating usability changes over time or when benchmarking against competitors. The hybrid method also supports managerial decision-making in contexts where user preferences are uncertain or difficult to quantify, offering actionable insights based on rigorous data aggregation.

To validate the rationality and effectiveness of the proposed Fuzzy SWARA–WASPAS approach, a comparative experiment was conducted using an alternative method: Fuzzy AHP combined with TOPSIS. This alternative framework was selected due to its widespread application in usability evaluation and its ability to handle subjective judgments through pairwise comparisons. Using the same dataset and usability criteria (trust, shopping support, efficiency, design, and satisfaction), both methods were applied to assess the ranking of usability factors for platforms P and S.

The results revealed several similarities and differences. For platform S, both methods consistently identified Trust as the most influential usability factor, reinforcing its critical role in user adoption. However, minor variations were observed in the ranking of mid-tier criteria, such as Efficiency and Shopping Support. While SWARA–WASPAS emphasized Shopping Support more prominently for platform P, the AHP–TOPSIS approach gave slightly higher importance to Efficiency. These differences illustrate the sensitivity of ranking outcomes to the choice of weighting method.

Nevertheless, the overall consistency in top-ranked and bottom-ranked factors across both methods confirms the robustness of the findings. The comparative experiment supports the argument that the hybrid Fuzzy SWARA–WASPAS approach is both effective and interpretable for prioritizing usability criteria, particularly when rapid decision-making and minimal consistency checking are desired.

This study has several limitations that should be acknowledged. First, the analysis was restricted to two e-commerce platforms within a specific regional and temporal context. As such, the results may not be generalizable to other platforms or user populations with different digital literacy levels, economic backgrounds or cultural preferences. Second, while the Fuzzy SWARA–WASPAS approach offers simplicity and robustness, the study did not consider other MCDM techniques in the main analysis until the comparative experiment, which may have provided more comprehensive insights.

Moreover, the research relies on cross-sectional data gathered through user questionnaires, which capture perceptions at a single point in time. Longitudinal studies could uncover how usability perceptions evolve with system updates or user experience maturity. Future research could expand the usability criteria to include dynamic aspects such as personalization, gamification or adaptive interfaces, which are increasingly relevant in modern e-commerce. Additionally, the integration of machine learning or AI-based decision models may enhance the predictive power and automation of usability evaluations.

Based on the reviews and ratings given by users on the Google Playstore for the platform P and platform S applications, it shows usability problems, then the existing criteria and sub-criteria include criteria of trust, shopping support, efficiency, design and satisfaction as materials for usability evaluation. This study aims to identify usability problems based on e-commerce application criteria and sub-criteria based on journals with similar research. Then determine the weight of the usability criteria and sub-criteria using the Fuzzy SWARA method. Then, the usability criteria were ranked using the WASPAS method. So that it can be seen which e-commerce application has the best usability. The difference in the results of the ranking order on the WASPAS fuzzy shows that each e-commerce application has its own level of usability.