Parallel Class Ranking Model Using Analytic Hierarchy Process With Multi Criteria

Madrasah Aliyah Negeri (MAN) Banyuwangi using a worksheet which can lead to error occurrences and slow decision making. A system for decision support that can improve the ranking process and quality were developed in this paper. The proposed system implemented the codeigniter framework, MySQL database, and PHP programming language. The system provided three user roles which are teacher, student, and administrator role. These four parameters are used as ranking system input, including academic values, non-academic values, violation scores, and student attendance. The ranking process was conducted by applying the analytic hierarchy process (AHP) method. The developed decision support system was tested using two ways: the black box testing method and providing questionnaires. Black box testing result shows that the system has functionally worked, while user’s questionnaire gives 92,29% well accepted by users. The results show that the decision support system can help manage values and determine the parallel ranking list.


I. INTRODUCTION
Class assessments are conducted by teachers to measure the achievement and development of students' learning processes towards the achievement of learning goals and mastery of competencies [1]. The results of the assessment are then used to determine student achievement.
Schools that have many study groups in the same class usually also use a parallel ranking model.
At the Banyuwangi Aliyah State Madrasah (MAN), high school (SMA) level schools under the auspices of the Indonesian Ministry of Religion, value processing and parallel achievement ranking in this school still use conventional systems that use worksheets. With a total of 1,003 students and 31 classes, manual classification results in error vulnerability and a slow ranking process. The primary ranking that has been running so far also only uses academic grades for the ranking of high achieving students. Ranking by including non-academic values will provide a more comprehensive picture of the assessment results [2]. On the other hand, schools also need a system that can manage student grade data properly.
To overcome the problems that have been described, a decision support system is needed to determine computerised academic and non-academic achievements that can later produce quality decisions. The system is built using the analytic hierarchy process (AHP) approach model. AHP method can be used to help the decision-making process of complex problems that have several alternative solutions to obtain the best resolution to resolve the issue. Some users of AHP in decision support systems in education include evaluating teacher performance [3], school selection [4], [5], student career development [6], scholarship determination [7], [8], and commitment of interest and majors [9]. AHP implementation for student ranking has also been carried out [10] but only using one-class data, and Z-Score is used as methods that provide a standardised value based on the deviation value and the average of each criterion Parallel student ranking system is built using the code igniter framework by applying the AHP method. This system is equipped with academic and non-academic value storage features as an essential feature of ranking. With this system, it is expected that the decision-making process of school academic achievement will be faster and more accurate.

II. RESEARCH METHOD
In the research conducted, a web-based application system built that provides an interface to enter grades, set parameters and processing, and to display the results of the system output in the form of student rankings. Implementation of the AHP approach used as a method for INTENSIF, Vol.4 No.1 February 2020 ISSN: 2580-409X (Print) / 2549-6824 (Online) DOI: https://doi.org/10.29407/intensif.v4i1.13769 ranking students. While developing system or software applications as a whole using the waterfall method of software development models. The waterfall method in Figure 1 shows the development of a system that is carried out sequentially step by step from requirements analysis, planning, modelling, development to system implementation [11], [12].

A. Research Data
This research data collection method is done through interviews with the school and direct data collection. Interviews conducted with all parties relating to the system, including students, teachers, the administration department, and elements of school management. While direct data retrieval is done by collecting related documents and direct observation of activities related to the problem [13]. The data obtained are grouped into two parts : 1. Primary data, i.e. data used to derive the criteria used in the process of ranking the AHP method includes: Academic Value is the value taken from the benefits of subjects in MAN Banyuwangi. The subjects can be shown in Table 1. Non-Academic Value is the value of student self-development, which is included in achievements outside of school activities, required how active students are in participating in this activity. The extracurricular activities at MAN Banyuwangi are listed in Table 2.

B. AHP Ranking
Flowchart for the calculation process using the AHP method is shown in Figure 2. First the admin gives the initial value of the criteria that have been determined in the form of academic, non-academic, absenteeism, and violation. After that, the admin enters data from each criterion.  Following is an example of calculation based on Figure 2 using the Analytic Hierarchy Process method according [14], [15] in parallel class ranking: 1. Establish criteria and sub-criteria and alternatives viz: Academic Value, Non-Academic Value, Attendance, and Violation.
2. Determining sub-criteria can be shown in Table 3. 3. Arrange the pairwise Comparison criteria matrix by comparing the ratings of each existing criterion. Comparative assessment of each measure is carried out by the school according to applicable regulations -Pairwise comparate criteria matrix shown in Table 4.   Table 5.  Table 6. Done by adding up the results of the multiplication values in Table 5 with the values in Table 7.  Table 8. 9. Change the matrix sub-criteria pair comparison into decimal numbers and add each column to the sub-criteria, as shown in Table 9. 10. Normalise all values by dividing each element per column by the total number of columns, as shown in Table 10.  Table 11.   Table 12.

A. System Design
The system design compiled in this study uses a use case diagram and entity relationship diagram. Use case diagram illustrates the functionality in a parallel ranking decision support system [16] as shown in Figure 4.

B. Implementasi Sistem
To be able to access the support system of parallel ranking decision, all users are required to log in according to their user name and password. After the admin has logged in, the admin home page will be displayed, as shown in Figure 5.   The Data menu has three submenus, namely class data, student data, and homeroom data.
The class data page is used to manage and display class information in the Banyuwangi MAN school, shown in Figure 7. AHP display is a menu for processing student grades that have been entered by the homeroom teacher using the AHP method as in Figure 10.   Figure 10, there are value criteria used to determine student rankings following the method used, the AHP method. Here the criteria used are determined in advance following the provisions in the AHP. All grades used in the AHP process are entered first by each homeroom teacher in the form of violations, absences, non-academic values, and academic values, as shown in Figure 11. The system then converts these values following Table 3.

Figure 11. PAGE MANAGE VALUE
The system then performs the results of the AHP calculation of all the criteria according to the AHP flowchart in Figure 2, and the results can be seen on the final value page shown in Figure 12. On this page, the value of each AHP process result criteria for all students is displayed. Furthermore, the amount will be multiplied by the criteria weight to get the final value which is then sorted to get the ranking, as shown in Figure 13.  Ranking list display is a student ranking list display from the results of the value processing process, which is shown in Figure 13. Each student can see the ranking of the AHP ranking according to their respective majors. To be able to see it, students enter the username and password that they have on the user's login page.

C. System Testing
System testing is an important thing to do to find out whether the system made runs according to its function or not [18]. The method used by the author is black box testing that is testing carried out by observing the results of system operation. Blackbox test results are in Table 13. Homeroom teachers enter student grades Student grades that have been entered by the homeroom teacher will be entered into the ranks on the system.

Corresp onding
Change student grades Homeroom teacher changes student grades Student grades that have been altered by the homeroom teacher will be entered into the ranks on the system.

Corresp onding
Testing several aspects of the application to users is also done to all system users according to the level of access rights, including admin (administrative staff), homeroom teacher, and students. After the user uses the system, they are asked to rate the system through a questionnaire based on ISO 9126 [19]. The results of the questionnaire gave a final value of 92.29%, which was well received by the user, as in Table 14.

IV. CONCLUSIONS AND RECOMMENDATIONS
In this study, a parallel class ranking model has been developed using the Analytic Hierarchy Process (AHP) method with multiple criteria in Banyuwang State Aliyah Madrasah by using four criteria, namely academic grades, non-academic values, violation scores, and absenteeism. The system is built on a web-based and has been tested using the black box testing method with relevant results, and the results of the user questionnaire obtained 92.29% of the system can be well received. The system can be completed by displaying graphs of the development of student achievement each semester.