Applying the Multiple-Attribute Decision Making - Simple Additive Weighting to determine the Most Popular Internet Provider among Students

Applying the Multiple-Attribute Decision Making - Simple Additive Weighting to determine the


Introduction
Today, internet access is one of the most important aspects (Mattern & Floerkemeier, 2010), especially during the COVID-19 pandemic, almost all activities are becoming online (Alsoud & Harasis, 2021;Azlan et al., 2020;Dhawan, 2020;Hanaei et al., 2022;Naik et al., 2021). In 2020, the Indonesian people facing the challenges of how to stay connected and doing activities without leaving the house. The pandemic has brought a new culture of digitalization in many sectors, including education (Alsoud & Harasis, 2021;Azlan et al., 2020;Dhawan, 2020;Naik et al., 2021). The COVID-19 pandemic forced students to continue their activities as usual even without meeting face to face (Kulikowski et al., 2022;Simamora, 2020). In order to stay connected to each other, the use of various applications to support learning activities is also increasing rapidly. Along with that, the need for internet services is also increasing. The government through the Ministry of Education and Culture provides quota assistance for students, teachers and lecturers that will be sent regularly to mobile numbers which is registered through schools. However, not infrequently the registered mobile number is no longer active so the assistance provided cannot be used. This causes parents must spend some additional funds for internet quotas so that online learning activities can still be carried out.
The number of internet providers, in this case cellular providers, causes students confused to choose which one of the cellular providers they will use. Expensive internet quota and poor internet connection can be an obstacle to student activities and productivity. To make it easier for students to determine which cellular provider they will use, it is necessary to have a decision support system in choosing one of the various types of cellular providers available. This decision support system can certainly meet the needs of internet quotas and budgets. The type of cellular provider selection system depends on many criteria, so an appropriate decision support method is needed. Various literatures provide many methods of decision support systems. One of them is Multi Attribute Decision Making -Simple Additive Weighting (MADM-SAW) methods. The use of the SAW method is based on its ability to make a more precise assessment because it is based on predetermined criteria values and preference weights. Beside of that, this method is performing a ranking process to select the best alternative from a number of existing alternatives. In this study, the SAW method will be used as a decision support method in determining the type of the most popular cellular provider. The purpose of this study is to determine the best choice of each criteria, sub-criteria and alternatives according to the wishes and needs of the cellular provider using the SAW method. With this decision-making support system, it is hoped that it can assist students in determining the type of cellular provider to use, so that online learning activities are carried out properly, without disturbing by network speed and without spending excessive funds to buy internet quota.

Multiple-Attribute Decision Making (MADM)
Multiple-Attribute Decision Making (MADM) is a branch of science that is generally used in comparing a limited set of alternatives (Alinezhad & Khalili, 2019). In management and planning, MADM has been used to study decision methods and procedures that can accommodate some of usual conflicting criteria (Büyüközkan et al., 2009). MADM model is a decision matrix consisting of ranking alternatives against each criterion. The evaluation rankings were collected by considering the weight of the criteria, and the global evaluation score for each alternative found (Nasab & Milani, 2012). There are several MADM methods including Simple Additive Weighting Method (SAW), Weight Product (WP), ELECTRE, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), and Analytic Hierarchy Process (AHP).

Simple Additive Weighting (SAW)
Simple Additive Weighting (SAW) method is known as the weighted addition method (Haswan, 2019). The basic concept of the SAW method is to find the weighted sum of the performance ratings for each alternative on all attributes (Purba & Sihotang, 2019;Putra & Punggara, 2018). The SAW method requires the process of normalizing the decision matrix (X) to a scale that can be compared with all existing alternative ratings (Kusumadewi et al., 2006). According to Darmastuti (2013), the advantage of the Simple Additive Weighting (SAW) model is in its ability to make a more precise assessment because it is based on predetermined criteria values and preference weights. Beside that, in the SAW method there is a matrix normalization calculation which is suitable to the value of the benefit and cost attributes (Afifah, 2012). According to Cahyapratama & Sarno (2018), the determination of the priority value of the weight vector is carried out according to the manager's policy to provide the weight vector value directly. The steps are as follows: a) Determine the criteria that will be used as a reference in decision making, namely Ci b) Determine the suitability rating of each alternative on each criterion. c) Make a decision matrix based on the criteria (Ci), then normalize the matrix based on the equation that is adjusted to the type of attribute (profit or cost attribute) in order to obtain a normalized matrix R.
The formula for normalization is expressed by: Where: = normalized performance rating = alternative and criteria of matrix = maximum value of each alternative and criteria = minimum value of each alternative and criteria The final result of each ranking process is the sum of the normalized matrix multiplication R with the weight vector (w) so that the largest value is chosen as the best alternative solution (Ai). The preference value for each alternative (Vi) is expressed by: Where: = alternative final result = predetermined weight = matrix normalization

Materials and Methods
This study is done by using the step-by-step for the flow of research, the research flow as seen in Figure 1 below: Step-by-Step Flowchart of Research

Alternative Variables and Criteria Variables
The alternative variables and criteria variables must be determined so that the process of determining decision support can be carried out. The variables of alternative and criteria are as follows : • The alternative variable is the available type of cellular provider at the study location, namely: a) Telkomsel b) 3 (Three Hutchison) c) Smartfren d) Indosat e) XL axiata • The criteria variables that will be used to determine the type of favorite cellular provider are: a) Internet network quality b) Internet quota price c) Service provider The sub-criteria referred in network quality are including upload speed, download speed, latency, coverage area and internet network stability. Internet quota price sub-criteria are including the price of internet packages, internet vouchers, and starter card. Service provider sub-criteria are including daily/weekly/monthly internet package promos, learning platform promos, YouTube/streaming internet quota bonuses, and package validity extensions. In this study, the data used are primary data from respondents to determine the weight of the criteria and assessment of the type of cellular provider by students at the research site.This study was conducted in Griya Martubung area, Medan city.This study used the accidental sampling technique from students in studies location. The total number of respondents is 150 people.

Results and Discussion
The initial step must be taken so that we get a decision matrix is determining the value of each alternative (A i ) from the 150 respondents recorded on the basis of each criteria. (C i ), the result as seen in Table 1 Table 2.  Table 2 shows the result of weighted (w) for three criterions. The result indicates that Criteria 1 (C1) is the provider selected due to internet network quality (w=0.50). Also, Criteria 2 (C2) is the provider selected due to internet quota price (w=0.18) and Criteria 3 (C3) is the provider selected due to services (w=0.32). In addition, this study employs the normalize data on the basis of benefit and cost attributes. It aims to obtain a decision matrix. Internet quota price criteria are categorized as cost attributes, while network quality and provider services are categorized as benefit attributes. The normalized matrix is given in Table 3.  Table 3 captures the result of normalised matrix. In this case, we use alternative providers and criteria by adding the weighted. The result shows that the first alternative is Telkomsel with C1 (w=0.50) is 0.60, C2 (w=0.18) is 1.00 and C3 (w=0.32) is 1.00. 3 (Three Hutchison) with C1 (w=0.50) is 1.00, C2 (w=0.18) is 0.25 and C3 (w=0.32) is 1.00. Indosat with C1 (w=0.50) is 0.30, C2 (w=0.18) is 0.60 and C3 (w=0.32) is 0.63. Smartfren with C1 (w=0.50) is 0.50, C2 (w=0.18) is 1.00 and C3 (w=0.32) is 0.38 and XL Axiata with C1 (w=0.50) is 0.60, C2 (w=0.18) is 0.43 and C3 (w=0.32) is 0.75. Further, the decision matrix is given in Table 4.  Table 4 indicates that decision matrix results with alternative internet provider on the basis of three criteria denoted by C1, C2 and C2. The first alternative is Telkomsel with the criteria values are C1 = 0.30, C2 = 0.18, and C3 = 0.32. 3 (Three Hutchison) with the criteria values are C1 = 0.50, C2 = 0.05, and C3 = 0.32. Indosat with the criteria values are C1 = 0.15, C2 = 0.11, and C3 = 0.20. Smartfren with the criteria values are C1 = 0.25, C2 = 0.18, and C3 = 0.12 and XL Axiata with the criteria values are C1 = 0.30, C2 = 0.08, and C3 = 0.24. In addition, this study uses rank of preferred internet providers. This ranking is done by summing the total value of each alternative for all criteria. The alternative with the highest score is the type of provider that is most popularly used among students, given in Table 5.  Table 5 displays the outcomes of preferred internet providers. The first rank is 3 (Three Hutchison) with the total value is 0.87. Second is Telkomsel with a total value is 0.80. XL Axiata, with the total value is 0.62 at position three. Smartfren is four with the total value is 0.55 and lastly is Indosat with the total value is 0.46.

Conclusions
In conclusion, this study identified that the most popular types of cellular providers among students sequentially are 3 (Three Hutchison), Telkomsel, XL Axiata, Smartfren and the last is Indosat. These results are given as a solution to choosing a cellular provider for students so that online learning activities run well by considering the various criteria and sub-criteria that have been described.n of your study. You may also put your personal reflection after conducting your study. Maximum 100 characters. The selection of the type of cellular provider for students using the SAW method is relatively simple, so it needs to be developed by using other methods so that the alternative selection process becomes better. Further development of this research can be done by considering other criteria, increasing the number of respondents or expanding the scope of the research area.