This study aims to design and develop a Decision Support System for Faculty Selection, Promotion, and

Reclassification Using Predictive Analytics to replace the inefficiencies of manual processes in higher education

institutions. Using logistic regression, the system evaluates faculty performance, tenure, and credentials to ensure

fair, data-driven decisions. Guided by Agile Scrum, it was iteratively refined through stakeholder feedback.

System testing, based on ISO 25010 standards, showed high ratings in functionality, performance, usability,

reliability, security, and maintainability, with an overall weighted mean of 4.56, described as Highly Acceptable.

User evaluation via the Technology Acceptance Model (TAM) also indicated strong acceptance, with an overall

processes for their selection, promotion, and reclassification [1]. Traditional decision-making often relies on

subjective evaluations, which can introduce biases and inconsistencies, thereby hindering institutional

effectiveness and fairness [2]. This paper proposes the integration of predictive analytics into a comprehensive

decision support system to enhance the objectivity, transparency, and efficiency of these critical human resource

functions. This advanced system will leverage data-driven insights to forecast future faculty performance,

identify optimal candidates, and streamline career progression pathways, fostering a more meritocratic academic

environment [3] [4].

Despite the growing use of data-driven tools in higher education, several research gaps remain in the

development of a Decision Support System for Faculty Selection, Promotion, and Reclassification using

predictive analytics. Existing studies often focus on faculty selection and promotion but give limited attention

traditional techniques such as regression, while the use of advanced or hybrid predictive models remains

underexplored [5][8]. Moreover, there is limited literature on incorporating systematic stakeholder

involvement—particularly from faculty, administrators, and HR—in defining and validating promotion and

reclassification criteria, even though human-centered approaches are essential for alignment with institutional

policies [9]. Finally, while some systems report usability, comprehensive evaluations using recognized

frameworks such as ISO 25010 software quality standards and acceptance models like TAM are still lacking,

leaving gaps in understanding user perceptions and system quality [6].

In the Philippines, many higher education institutions are considering implementing HRIS solutions to improve

To address these operational challenges, the study proposes the development of a Decision Support System

(DSS) for Faculty Selection, Promotion, and Reclassification using Predictive Analytics. Unlike a generic HRIS,

this system focuses on streamlining processes critical to academic institutions by automating data collection,

faculty evaluation, and promotion workflows. Tailored modules such as the Faculty Information Module,

Performance and Appraisal Module, Promotion and Reclassification Module, and Decision Support Dashboard

are designed to meet institutional requirements. Advanced features, including predictive analytics for faculty

ranking based on tenure, credentials, and performance indicators, as well as reporting tools for generating

insights on faculty demographics, status, and career progression, will further enhance transparency and evidence-

based decision-making. By implementing this system, higher education institution can overcome inefficiencies

in manual evaluation, ensure fairness in faculty advancement, and strengthen data-driven HR practices, fostering

Predictive Analytics employed the Agile Scrum methodology. This approach was chosen because of its

flexibility, iterative process, and emphasis on stakeholder collaboration, which are essential for projects with

evolving requirements such as faculty promotion and reclassification.

Development was carried out in sprints, where system features were planned, implemented, and tested in short

cycles. Regular stakeholder feedback ensured that modules, such as data integration, predictive analytics, and

decision dashboards, were refined continuously to meet institutional policies and user needs. Unlike traditional

linear models, Scrum allowed adjustments at any stage, reducing risks and improving system quality.

Agile Scrum was selected because it supports dynamic requirements, encourages active stakeholder

participation, ensures incremental delivery of functional components, and maintains quality through iterative

testing. This methodology ensured that the system was user-centered, adaptable, and aligned with institutional

Conceptual Framework

jQuery, Apache, MySQL, and Bootstrap, which are essential for developing a dynamic, web-based system.

Third, the *hardware requirements* include servers, client computers, tablets, and mobile devices that ensure

accessibility, deployment, and scalability of the system across platforms.

Figure 2. Conceptual Framework of the Study

The Process represents the development methodology, guided by the Agile Scrum framework Activities include

preparing the project backlog, sprint planning, and sprint backlog creation, followed by iterative sprint cycles.

Each sprint involves daily scrums for coordination, sprint reviews for stakeholder feedback, and retrospectives

for continuous improvement. This iterative process ensures that system functionalities, such as faculty profiling,

performance evaluation, and predictive ranking, are incrementally refined until the complete product is

security, and maintainability, and the Technology Acceptance Model (TAM), which assesses user perceptions

of usefulness and ease of use.

Logistic Regression Algorithm

The researcher employed a linear regression algorithm to predict which faculty members are potential candidates

for promotion or reclassification. Linear regression was selected because it is one of the most effective and

widely used predictive modeling techniques for analyzing relationships between variables. In this study, faculty

performance indicators such as teaching effectiveness, years of service, educational attainment, research output,

extension services, and professional development serves as independent variables, while promotion or

reclassification status is treated as the dependent variable.

Logistic Regression Model

Figure 4. Logistic Regression Modelling Using Orange Visual

The dataset for this study was obtained from the Human Resource Department of Colegio de Sta. Teresa de

Historical data served as the foundation for developing the predictive model, enabling the system to uncover

patterns and relationships among variables. By analyzing these trends, the model generates insights into faculty

performance and potential, supporting objective, data-driven decisions in faculty selection, reclassification, and

promotion. To ensure practical applicability, the researchers used Orange, a user-friendly yet powerful data

mining tool, which provided both efficiency in model development and interpretability of results for institutional

decision-making.

Before model construction, the dataset underwent data cleansing. An imputation technique was applied to handle

missing values and outliers by replacing them with the mean of each variable. Afterward, the dataset was divided

into two subsets: 80% for training and 20% for testing, ensuring a reliable evaluation of the model’s performance.

user-friendly. Overall, the findings demonstrate that the system is effective, easy to use, and has strong potential

System Evaluations Test Result