Ranking the strengths and weaknesses of software engineering students in software development life cycle (SDLC) process level is a challenging task owing to (1) data variation, (2) multievaluation criteria, (3) criterion importance and (4) alternative member importance. According to the existing literature, no specified procedure can rank the ability of software engineering students based on SDLC process levels to figure out the strengths and weaknesses of each student. This study aims to present a novel triplex procedure for ranking the ability of software engineering students to address the literature gap. The methodology of the proposed work is presented on the basis of three phases. In the identification phase, four steps are implemented, namely, processing dataset, identifying the criteria, distributing the courses to the software engineering body of knowledge and proposing the pre-decision matrix (DM). The data comprise the GPA and soft skills from 60 software engineering students who graduated from Universiti Pendidikan Sultan Idris in 2016. In the pre-processing phase, three steps are involved as follows. Analytic hierarchy process (AHP) is first used to assign weights to the courses and then multiply the assigned weight by courses, which is the first procedure in the proposed work. In this phase, the construction of DM is presented based on multimeasurement criteria (GPA and soft skills), with SDLC process levels as alternatives. In the development phase, AHP is used again to weight the multimeasurement criteria, and this is the second procedure. In such case, the coordinator and head of the software engineering department are consulted to obtain subjective judgments for each criterion. Technique for order performance by similarity to ideal solution (TOPSIS) is then used to rank the students, which is the third procedure. In the validation, statistical analysis is performed to validate the results by checking the accuracy of the systematic ranking. Results show that (1) integrating AHP and group TOPSIS is suitable for ranking the ability of students. (2) The 60 students are categorized into five ranking groups based on their strength level: 14 collector requirements, 13 designers, 5 programmers, 13 testers and 15 maintenances. (3) Significant differences are observed between the groups’ scores for each level of SDLC, indicating that the ranking results are identical for all levels.

中文翻译：

---

基于两级层次分析法和群TOPSIS技术的软件工程专业学生能力排名新三重程序

由于（1）数据变化，（2）多重评估标准，（3）标准重要性和（4）替代成员重要性，对软件工程专业学生在软件开发生命周期（SDLC）过程级别的优缺点进行排名是一项具有挑战性的任务。根据现有文献，没有特定的程序可以根据 SDLC 过程级别对软件工程专业学生的能力进行排名，以找出每个学生的优势和劣势。本研究旨在提出一种新颖的三重程序，用于对软件工程专业学生解决文献空白的能力进行排名。拟议工作的方法是在三个阶段的基础上提出的。在识别阶段，执行四个步骤，即处理数据集，识别标准，将课程分发到软件工程知识体系并提出预决策矩阵（DM）。数据包括 2016 年毕业于 Pendidikan Sultan Idris 大学的 60 名软件工程专业学生的 GPA 和软技能。在预处理阶段，包括以下三个步骤。层次分析法 (AHP) 首先用于为课程分配权重，然后将分配的权重乘以课程，这是拟议工作中的第一个程序。在此阶段，DM 的构建基于多重测量标准（GPA 和软技能），SDLC 过程级别作为替代方案。在开发阶段，再次使用层次分析法对多重测量标准进行加权，这是第二个程序。在这种情况下，咨询协调员和软件工程部门负责人，以获得对每个标准的主观判断。然后使用与理想解相似度排序的技术（TOPSIS）对学生进行排名，这是第三个过程。在验证中，进行统计分析，通过检查系统排序的准确性来验证结果。结果表明：（1）层次分析法与小组TOPSIS相结合适合学生能力排名。(2) 60 名学生按实力等级分为 5 个等级组：14 名采集员要求、13 名设计师、5 名程序员、13 名测试人员和 15 名维护人员。(3) SDLC各级别的组得分存在显着差异，表明各级别的排名结果相同。