Roxana-Adriana Puiu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Roxana-Adriana Puiu has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Florentina-Geanina Tache
National University of Science and Technology POLITEHNICA Bucharest , Romania
1 shared publication Gabriel Petrea
National University of Science and Technology POLITEHNICA Bucharest Bucharest Romania
1 shared publication- Laura-Florentina Boanță1 shared publication
Marius‐Valentin Drăgoi
Faculty of Industrial Engineering and Robotics National University of Science and Technology Politehnica Bucharest Bucharest Romania
1 shared publicationRomina ‐ Cătălina Pascaru
Faculty of Entrepreneurship, Engineering and Business Management National University of Science and Technology POLITEHNICA Bucharest Bucharest Romania
1 shared publication- Romina-Cătălina Pascaru1 shared publication
Latest Publications
- Open Access
- 3 June 2026
- Published by Wiley in Macromolecular Symposia
Application in Research Projects on the Innovation Found in the Field of Engineering of Polymer and Composites Products
Planning human resources well is very important for the success of research and development (R&D) initiatives, especially in polymer and composite product engineering projects where multidisciplinary expertise and laboratory constraints must be carefully coordinated. This paper presents a prototype system that employs machine learning (ML) to help people decide who should work on a project. A synthetic dataset was designed to emulate realistic HR data, such as domain‐specific expert skills relevant to polymer processing, composite manufacturing, and materials characterization, availability, past performance, cost, and personal preferences. The algorithm suggests the best team members by optimizing many parameters, such as matching talents, balancing workloads, staying under budget, and looking at past performance. The system is set up using a graphical interface that is easy to use. Project managers may use it to create new tasks typical for polymer and composite R&D activities, such as thermal characterization, process validation, or materials data analysis, and receive ranked recommendations for who should do them. Extra features include filtering choices, tools for simulating teams, and basic risk assessments for when there are under‐ or overstaffing. The suggested solution seeks to make HR allocation more open and efficient in academic, industrial, and materials‐focused research organizations involved in polymer and composite product engineering. It will also help the area of Operations Management by automating data‐driven decision‐making in innovation‐oriented materials research projects.
- 3 June 2026
- Published by Wiley in Macromolecular Symposia
Open AccessApplication in Research Projects on the Innovation Found in the Field of Engineering of Polymer and Composites Products
Planning human resources well is very important for the success of research and development (R&D) initiatives, especially in polymer and composite product engineering projects where multidisciplinary expertise and laboratory constraints must be carefully coordinated. This paper presents a prototype system that employs machine learning (ML) to help people decide who should work on a project. A synthetic dataset was designed to emulate realistic HR data, such as domain‐specific expert skills relevant to polymer processing, composite manufacturing, and materials characterization, availability, past performance, cost, and personal preferences. The algorithm suggests the best team members by optimizing many parameters, such as matching talents, balancing workloads, staying under budget, and looking at past performance. The system is set up using a graphical interface that is easy to use. Project managers may use it to create new tasks typical for polymer and composite R&D activities, such as thermal characterization, process validation, or materials data analysis, and receive ranked recommendations for who should do them. Extra features include filtering choices, tools for simulating teams, and basic risk assessments for when there are under‐ or overstaffing. The suggested solution seeks to make HR allocation more open and efficient in academic, industrial, and materials‐focused research organizations involved in polymer and composite product engineering. It will also help the area of Operations Management by automating data‐driven decision‐making in innovation‐oriented materials research projects.
- 1 January 2025
- Published by Editconcept
- p. 468-473
A Systematic Review of Random Forest-Based Decision-Support Systems in Educational Management
An increasing number of educational institutions are using predictive analytics to help with strategic decision-making and programs that help students succeed. This paper looks into the most recent improvements in using supervised learning models, specifically Random Forest classifiers, to keep an eye on behavioral trends in digital learning settings. We focus on studies that use structured data exported from Learning Management Systems (LMS), with a concentration on CSV-based processing procedures because they are easy to use and understand. The research brings together several uses of predictive logic in schools, looking at how well the algorithms work, how easy they are to use, and how they were designed. This review provides a theoretical basis for building smart, user-friendly educational tools that can adapt to students' activity profiles by establishing consistent modeling patterns and ways to make them understandable.
- 1 January 2025
- Published by Editconcept
- p. 468-473
A Systematic Review of Random Forest-Based Decision-Support Systems in Educational Management
An increasing number of educational institutions are using predictive analytics to help with strategic decision-making and programs that help students succeed. This paper looks into the most recent improvements in using supervised learning models, specifically Random Forest classifiers, to keep an eye on behavioral trends in digital learning settings. We focus on studies that use structured data exported from Learning Management Systems (LMS), with a concentration on CSV-based processing procedures because they are easy to use and understand. The research brings together several uses of predictive logic in schools, looking at how well the algorithms work, how easy they are to use, and how they were designed. This review provides a theoretical basis for building smart, user-friendly educational tools that can adapt to students' activity profiles by establishing consistent modeling patterns and ways to make them understandable.