LGPMI - Demand-aware drug assignment in manipulator arm automated

Conference paper

Yassine Bouhelassa, Khalid Hachemi
Proceedings of the First National Conference on Innovation in Data Engineering and AI Science (IDEAS 2025), University of Science and Technology of Oran Mohamed Boudiaf, Oran, Algeria, 2025 Jun

Read

Cite

APA Click to copy
Bouhelassa, Y., & Hachemi, K. (2025). Demand-aware drug assignment in manipulator arm automated dispensing systems via graph convolutional network ranking. In Proceedings of the First National Conference on Innovation in Data Engineering and AI Science (IDEAS 2025). Oran, Algeria: University of Science and Technology of Oran Mohamed Boudiaf.

Chicago/Turabian Click to copy
Bouhelassa, Yassine, and Khalid Hachemi. “Demand-Aware Drug Assignment in Manipulator Arm Automated Dispensing Systems via Graph Convolutional Network Ranking.” In Proceedings of the First National Conference on Innovation in Data Engineering and AI Science (IDEAS 2025). Oran, Algeria: University of Science and Technology of Oran Mohamed Boudiaf, 2025.

MLA Click to copy
Bouhelassa, Yassine, and Khalid Hachemi. “Demand-Aware Drug Assignment in Manipulator Arm Automated Dispensing Systems via Graph Convolutional Network Ranking.” Proceedings of the First National Conference on Innovation in Data Engineering and AI Science (IDEAS 2025), University of Science and Technology of Oran Mohamed Boudiaf, 2025.

BibTeX Click to copy

@inproceedings{bouhelassa2025a,
  title = {Demand-aware drug assignment in manipulator arm automated dispensing systems via graph convolutional network ranking},
  year = {2025},
  month = jun,
  address = {Oran, Algeria},
  organization = {University of Science and Technology of Oran Mohamed Boudiaf},
  author = {Bouhelassa, Yassine and Hachemi, Khalid},
  booktitle = {Proceedings of the First National Conference on Innovation in Data Engineering and AI Science (IDEAS 2025)},
  month_numeric = {6}
}

Abstract

This paper introduces a Graph Convolutional Network (GCN)‑based ranking model to optimize drug placements within Automated Drug Dispensing Systems (ADDS). The method represents drugs as nodes with features derived from monthly consumption frequencies and connects them through a co‑use graph. The GCN uses message‑passing operations to learn representations that capture drug demand behavior and co‑prescription relationships, producing drug scoring values. These scores are then used to construct an optimized placement matrix that positions high‑priority drugs in the most accessible compartments of a manipulator‑arm‑based ADDS. The optimized placement configuration achieved a 27 % reduction in average retrieval times compared to random placement strategies. Additionally, t‑SNE analysis of drug embeddings revealed meaningful clusters that reflect drug relevancy. The approach is adaptable to different ADDS configurations and demonstrates potential for significant operational improvements in healthcare logistics.

Keywords:

Artificial intelligence; Graph Neural Networks; Automated Drug Dispensing Systems; Graph Convolutional Networks; Healthcare logistics; Drug assignment optimization.