Implementation of the AHP-SMARTER Method in the Decision Support System for Giving Sanctions for Violation of Student Disciplines

−Violations of school rules are often carried out by students, including lack of respect for teachers, students who are not on time, often late for class, skipping classes, jumping fences, smoking and not paying attention to the rules and other regulations in school. This study aims to build a decision support system for sanctions for violations of student discipline that has the ability to analyze each of the criteria and sub-criteria that have been determined by the school. In this case, students who violate school rules will be punished and given sanctions so as to provide an output value of priority intensity which results in a system that provides an assessment of violations against students. The method used in building this decision support system is by combining the Analytical Hierarchy Process (AHP) method and the Simple Multi Attribute Rating Technique Exploiting Rank (SMARTER) method. Weighting criteria using the AHP method and for ranking using the SMARTER method. The system created can be used to assist in processing data on violations of school rules. With this decision support system, it is hoped that policy makers will have no difficulty in determining what types of actions and sanctions will be given to students who violate school rules.

INTRODUCTION Each school has its own policy in determining the level of student discipline. The Integrated Islamic Vocational High School (SMK) of Generasi Muslim Cendikia (GMC) still uses a system of calculating points for violations and determining the sanctions for violations that are still manual, namely by recording all events or student problems into a book. The decision support system suggested by the counseling guidance teacher is a system that makes it easier to evaluate the level of student discipline and sanctions for violations effectively and efficiently. Giving sanctions by teachers in the teaching process is influenced by several factors, namely the seriousness factor in learning, consequences, delinquency at the school level, and family stability factors. [2]. Education in Indonesia not only prioritizes the development of cognitive aspects or knowledge of students, but also pays attention to individual development as a whole person [4].
SMK-IT GMC is a vocational school that has quite a lot of students. Every school must have rules and regulations that must be obeyed and followed by every student but not infrequently these rules and regulations are violated, the violations that often occur are students who are not on time, often late for class, skipping class time, jumping fences, smoking and so on.
According to [5]The system of sanctions for violations of the rules in some schools is still in the form of warning letters and direct reprimands to students. Along with presents development of technology and communication a new challenge that can make guidance and counseling more practical. One of them is a Decision Support System which is an approach to decision making [6]. The method that can support solving this problem is by combining the Analytical Hierarchy Process (AHP) method and the Simple Multi Attribute Rating Technique Exploiting Rank (SMARTER) method.
The system built can be used to assist in processing data on violations of school rules, especially student violations [3]. Although basically there are rules and sanctions that have been implemented in schools, the sanctions are still handled in the usual way without clear diferences between the violations committed and the sanctions given (different violations the sanctions are almost the same).
Therefore, researchers want to design a decision support system for sanctions for violating student rules. Every student who violates the rules will be given sanctions so that it can provide a deterrent effect and increase the values of decency and order in the school environment. This is useful to facilitate decision making related to disciplinary issues.

II. RESEARCH METHODOLOGY A. Decision Support System
Decision Support Systems (DSS) are usually built to support a solution to a problem or to an opportunity. Decision Support System (DSS) applications are used in decision making [7]. Decision Support System (DSS) application uses a flexible, interactive and adaptable CBIS (Computer Based Information System), which was developed to support solutions to unstructured specific management problems [8].

B. AHP (Analitical Hierarchy Process)
This method was first developed by Saaty (Saaty, 1980)[9]. The hierarchical model stated by Saaty is a functional hierarchical model with the main input being human perception. In general, the steps in using the AHP method for solving a problem are as follows [10]: a. Defining the problem and determining the desired solution. b. Determining the priority of elements c. Synthesis The things to do in this step are: 1) Sum the values of each column in the K matrix.
2) Divide each value from the column by the corresponding column total to obtain a normalized matrix. 3) Sum the values of each row and divide by the number of elements to get the priority weight value. d. Measuring Consistency The things that are done in this step are as follows: 1) Each value in the first column is multiplied by the priority weight of the first element, then each value in the second column is multiplied by the priority weight of the second element and so on. 2) Sum each row (∑ row).
3) he result of the sum of the rows is divided by the priority element in question so that it gets Lambda.  [2]states that SMARTER is a multicriteria decision-making technique based on the theory that each alternative consists of a number of criteria that have values and each criterion has a weight that describes its importance when compared to other criteria. This weighting is used to assess each alternative in order to obtain the best alternative. SMARTER uses a linear additive model to predict the value of each alternative. The analysis involved is transparent so this method provides a high level of understanding of the problem and can be accepted by decision makers [1].
The model used in SMART is shown in the equation: Information : Wj = The weighting value of the J-th criterion of the k criteria. U(ai) = The utility value of the I-th criterion for the Ith criterion Where I = 1, 2,…,m The steps of the SMARTER method are as follows [3]: a. Determine the number of criteria for the decision to be taken. b. Giving weight to each criterion by using an interval of 1-100 for each criterion with the most important priority. c. Calculating the normalization of each criterion by comparing the value of the weight of the criteria with the number of weights of the criteria, using the formula: Information : NWj = Normalization of J-th criterion weights Wj =J-th criterion weight k = Number of criteria Wn = The weight of the N-th criterion. d. Provide a criterion value for each alternative e. Calculates final grades and performs rankings using the SMARTER model.

D. Research Stages
To assist in the preparation of this research, it is necessary to have a clear framework for the stages [11]. This framework is the steps that will be taken in solving the problems that will be discussed. Generasi Muslim Cendikia is the current system that is still not standardized in this case different violations (mild and severe) but the handling is the same and the sanctions given are sometimes the same as other violations. In giving sanctions, there is only a warning and a statement letter, so there are several procedures that are not in accordance with the procedures that should have been applied to students. b. In this study, data collection was done by interview, observation and literature study. At this stage, it is done to find out, get data and information that will later support this research [12]. Observation Method Observations were carried out directly at SMK-IT GMC by looking at the daily lives of students and teachers as well as existing problems to find out the types of violations and sanctions that students received if they violated the rules and regulations.

Interview Method
Interviews were conducted by asking directly to the Guidance Counseling teacher who directly handles problematic students at SMK-IT GMC.

Library Study Method
Literature study is done by reading various kinds of information related to the research title. Researchers took reference sources from national scientific journals and books from the internet. c. The problem analysis step is needed to determine recommendations for sanctions for violations of school rules committed by students. With this data analysis, a clear picture of the problems discussed will be obtained [7]. d. Decision Support System Design, this stage is the activity carried out to make the formulation of the model, the selection of what criteria are taken into consideration for decision makers to decide the best alternative, measure and predict the results that occur. [4]. e. In this study, the authors implement the AHP-SMARTER method so that they are able to provide recommendations for sanctions for violations of school regulations committed by students. This phase translates the design results into software. f. The process of testing the application using blackbox.
Testing is done by testing all existing navigation, this test ensures that the processes carried out produce output that is in accordance with the design that has been made [13]. g. Conclusions are drawn after the design, implementation, and testing stages have been completed [14]. This stage discusses the results of the final goal to be achieved, namely the creation of a decision support system application that can later benefit schools related to the provision of appropriate sanctions in accordance with existing standard procedures. [13].

E. Research Material
The research material used to make a decision support system for the awarding of sanctions for student discipline violations is by using the AHP-SMARTER method. With the object of research SMK-IT Generasi Muslim Cendikia.

F. Design Model
Research with the application of the SMARTER Method in determining the sanctions for violations which will be combined with the AHP method, will use linear sequential in the design model. The activities in linear sequential are: Figure 2. Linear Sequential Model [15] a. Requirements analysis is the stage of analyzing the needs needed in making software b. The design stage is the translation stage of the analyzed data into a form that is easily understood by users. c. Coding is the stage of translating data that has been designed using a particular programming language. d. Testing is the stage of testing the software that has been made.

III. RESULTS AND DISCUSSION
The implementation of this system is carried out using two process methods, namely weighting criteria using the AHP method and ranking using the SMARTER method. The dotted line indicates the transfer of the calculation process from the AHP method to the SMARTER method, indicating the separation between the AHP process and the SMARTER process. In the AHP method, after the weights are obtained, it will be continued by testing the consistency. The goal is whether the weights are consistent or not. If the weights are consistent, it will result in weighting, and if not, it will return to the pairwise comparison matrix. After the weight is obtained, it will be continued with ranking using the SMARTER method [1]. The weights obtained in the AHP method will be used as weights on the criteria.

Decision Support System Criteria and Alternatives
The criteria used in this decision support system are as follows: Calculation Using AHP Method The next stage is to determine the priority of the elements by compiling criteria and sub-criteria in the form of a pairwise comparison matrix [8]. To find out the results of the weighting of the criteria used in calculating the priority of criteria and sub-criteria with the AHP method, it is necessary to search for values. How to get a value that can be with a certainty value or by conducting a survey through several respondents using a questionnaire sheet [11].The value of certainty is a value that is directly given for certain criteria, while the value of the questionnaire is the value obtained from the assessment given by the respondent where each respondent gives a different preference value using a scale of 1-9 [8].
Determining the priority of elements by compiling these criteria in the form of a pairwise comparison matrix[9]. Next is to calculate the value of the criteria column elements, where each criterion column element is divided by the number of matrices for each column in table 5, then add up the row matrix of the values of each element. The next stage is to multiply the elements in the pairwise comparison matrix column multiplied by the priority value results in Table 7, the multiplication results are then added up per each row. The next step is to add up the matrix of the sum of each row in Table 6 with the result of the "priority" value in Table 8.  From the results of the calculations in the table above, the value of the preference weights can show that the most important weight order criteria with a weight of 34.1%. Next are the criteria for Politeness in Association with a weighted value of 23.8%, the criteria for leaving school with a value of 18.2%, the criteria for school uniforms 13.0% and the criteria for school attendance with a weighting value of 10.8%.   Use Case Diagram 1. In the Use Case Diagram below, there are 4 actors who play a role in the running of the program. The first actor is the BK teacher, the BK teacher can do the login process, manage data such as student data, violation data, witness data, action data, summons, and change passwords. 2. The second actor is students, in this system students can log in and view their own data. 3. The third actor is the principal, in this system the principal can log in and see all the existing data. The principal also received a report 4. The fourth actor is the student's guardian, in this system the student's guardian can log in and view the data on rules, violations, sanctions and student/children's own data. Guardians of students can also receive summons.   Sistem Implementation This section will discuss the implementation of system analysis and design. The things discussed in the implementation include the opening view, the main menu display, the sub menu display, the input and output design display [16], [17], [18]. [19]. [20]. Display of the Criteria Calculation Form using the AHP Method.      After the violation process will be summed up all types of violations committed then will be shown the type of sanctions that will be given and the actions that will be taken by the Counseling Guidance teacher.

Display of Violation Point Calculation Result Form
The results obtained after all violations are processed are the display of the number of points, the sanctions obtained and the actions to be taken by the Counseling Guidance teacher. After the calculation results appear, the Admin can print the results by clicking the print results button. After the violation committed by the student is processed, the admin can print the violation card. IV. CONCLUSION Based on the research carried out up to the stage of designing, implementing, and testing the software, it can be concluded that from testing the process of calculating student discipline violations with the AHP-SMARTER method, it can be used and is able to provide the right solution in making decisions about giving sanctions to participants. students who violate school rules. From the results of this study, the 5 highest violations committed by students were taken by looking at the first violation point 78.5 sanctions given S6 and actions taken by T6, the two students with 46.5 violation points with S4 sanctions and T4 sanctions, the third students with 31.25 violation points with S3 sanctions and T3 sanctions, the four students with 21.5 violation points with S2 sanctions and T2 actions and the five students with violation points 15.75 with a S2 sanction and T2 action. The decisions taken by the Counseling Guidance Teachers, homeroom teachers and principals can be accounted for with the support of model calculations in the decision support system.