Drug Inventory Classification Management Based on PCA Algorithm and K-means Clustering Algorithm
10.3870/j.issn.1004-0781.2025.04.029
- VernacularTitle:基于主成分分析算法和K均值聚类算法的药品库存分类管理
- Author:
Lei TANG
1
;
Lei QIU
1
;
Jiahui YU
1
;
Zhaoshuai JI
1
Author Information
1. 清华大学附属北京清华长庚医院/清华大学临床医学院药学部,北京 102218
- Publication Type:Journal Article
- Keywords:
Drug classification;
PCA algorithm;
K-means clustering algorithm;
Drug inventory management
- From:
Herald of Medicine
2025;44(4):682-686
- CountryChina
- Language:Chinese
-
Abstract:
Objective To address the current issues of strong subjectivity in drug classification,vague classification standards,and complex influencing factors,this study aims to explore a scientific method for drug classification to reduce inventory costs and improve inventory effectiveness.Methods A total of 700 drugs were randomly selected from the historical data of a tertiary hospital in Beijing from 2021 to 2022 as the research subjects.The classification was conducted using the Principal Component Analysis(PCA)algorithm and the K-means clustering algorithm(K-means).Results The optimal number of classifications was determined to be 4,with a silhouette coefficient of 0.347 0.The 700 drugs were divided into four categories,with 363 in the first category,186 in the second,94 in the third,and 57 in the fourth.The drug classification method studied in this paper was simulated and applied to the drug inventory management of a certain tertiary hospital in the second quarter of 2023.The simulation results indicated that the classification method studied in this paper could reduce inventory costs and improve inventory effectiveness.Conclusion The drug classification method based on PCA algorithm and K-means clustering algorithm can provide a reliable basis for the management of drug inventory classification.
