Welcome to visit Zhongnan Medical Journal Press Series journal website!
Zhongnan Medical Journal Press Series of Journals Center for Evidence-Based and Translational Medicine of Wuhan University

The title
  • The title
  • The keywords
  • The author
Submit
Home Articles Vol 30,2026 No.6 Detail

Research on hierarchical early warning of drug shortages based on probability prediction and interval estimation

Published on Jul. 08, 2026Total Views: 56 times Total Downloads: 17 times Download Mobile

Author: FENG Yuhan 1 JIANG Tingyuan 1 JING Fuhao 1 TANG Pengzhan 1 PANG Zhonghua 1

Affiliation: 1.School of Information Engineering, Zhongyuan Institute of Science and Technology, Xuchang 461000, Henan Province, China

Keywords: Drug shortage Demand forecasting Hybrid model Variational mode decomposition Newton-Raphson-based optimizer Least squares support vector machine DeepAR Long short-term memory network Tiered early warning

DOI: 10.12173/j.issn.2097-4922.202603002

Reference: FENG Yuhan, JIANG Tingyuan, JING Fuhao, TANG Pengzhan, PANG Zhonghua.Research on hierarchical early warning of drug shortages based on probability prediction and interval estimation[J]. Yaoxue QianYan Zazhi, 2026, 30(5): 1001 - 1009.DOI: 10.12173/j.issn.2097-4922.202603002[Article in Chinese]

  • Abstract
  • Full-text
  • References
Abstract

Objective To establish a hybrid forecasting framework for 7- to 30-day drug demand/shortage prediction and tiered early warning during public health emergencies.

Methods Variational mode decomposition (VMD) was used to extract multi-scale features, the Newton-Raphson-based optimizer (NRBO) was used to optimize the key parameters of the least squares support vector machine (LSSVM), and the DeepAR-long short-term memory (LSTM) hybrid model was used to capture nonlinear patterns and uncertainty. Rolling backtesting was conducted with ARIMA and other models as baselines. Tiered warnings were triggered when predicted demand exceeded inventory plus safety stock.

Results The hybrid model achieved lower errors than single-step methods, with greater stability during shock periods. Prediction intervals characterized uncertainty and supported early warning and the generation of procurement/reallocation recommendations.

Conclusion The decomposition-optimization-integration framework improves forecasting accuracy and robustness, offering a deployable prediction-to-early-warning pathway for drug supply assurance at the grassroots and regional levels.

Full-text
Please download the PDF version to read the full text: download
References

1. 郑泽豪, 胡琪骏, 冯达. 国内外药品短缺研究现状与热点前沿的可视化分析[J]. 医药导报, 2022,41(12):1784-1794.ZhengZH, HuQJ, FengD. Visual analysis of research status and hot frontiers of drug shortages at home and abroad[J]. Herald of Medicine, 2022, 41(12): 1784-1794. DOI:10.3870/j.issn.1004-0781.2022.12.011.

2. 周德雨, 蒯丽萍, 徐冬艳, 等. 国家集中带量采购药品的短缺情况分析与对策建议[J]. 中国药房, 2023, 34(7): 769-773.ZhouDY, KuaiLP, XuDY, et al. Analysis and countermeasures of drug shortages under national centralized drug procurement[J]. China Pharmacy, 2023, 34(7): 769-773. DOI:10.6039/j.issn.1001-0408.2023.07.01.

3. World Health Organization. Technical definitions of shortages and stockouts of medicines and vaccines: meeting report[R]. 2017.

4. 张秋玉, 王芸, 胡元霞, 等. 我国药品及医用耗材集中带量采购政策的实施现状及建议[J]. 中国药房, 2022, 33(2): 136-141.ZhangQY, WangY, HuYX, et al. Implementation status and suggestions of centralized volume-based procurement policy for medicines and medical consumables in China[J]. China Pharmacy, 2022, 33(2): 136-141. DOI:10.6039/j.issn.1001-0408.2022.02.02.

5. 谭清云, 梁诗渝. 基于政策工具的短缺药供应保障机制的推进与优化研究[J]. 中国卫生政策研究, 2021, 14(12): 26-33.TanQY, LiangSY. Research on promotion and optimization of shortage drug supply guarantee mechanism based on policy tools[J]. Chinese Journal of Health Policy, 2021, 14(12): 26-33. DOI: 10.3969/j.issn.1674-2982.2021.12.004.

6. 国家卫生健康委员会. 全国公立医疗卫生机构药品使用监测管理标准: WS/T 841-2024[S]. 2024: 1-14.

7. 吴义来, 卢猛猛, 张烨雯, 等. 基于ARIMA模型的医疗机构药品需求量预测研究——以他汀类药物为例[J]. 中国现代应用药学, 2025, 42(9): 1545-1550.WuYL, LuMM, ZhangYW, et al. Research on drug demand forecasting in medical institutions based on ARIMA model: taking statins as an example[J]. Chinese Journal of Modern Applied Pharmacy, 2025, 42(9): 1545-1550. DOI:10.13748/j.cnki.issn1007-7693.20250068.

8. SalinasD, FlunkertV, GasthausJ, et al. DeepAR: probabilistic forecasting with autoregressive recurrent networks[J]. Int J Forecast, 2020, 36(3): 1181-1191. DOI:10.1016/j.ijforecast.2019.07.001.

9. HochreiterS, SchmidhuberJ. Long short-term memory[J]. Neural Comput, 1997, 9(8): 1735-1780. DOI:10.1162/neco.1997.9.8.1735.

10. 肖彩云, 王奇巍, 刘悦, 等. 药品短缺预警量化指标研究综述[J]. 中国医药工业杂志, 2024, 55(4): 584-590.XiaoCY, WangQW, LiuY, et al. Review on quantitative indicators for early warning of drug shortages[J]. Chinese Journal of Pharmaceuticals, 2024, 55(4): 584-590. DOI:10.16522/j.cnki.cjph.2024.04.021.

11. 孙怡园, 刘军教, 倪颖, 等. 基于风险矩阵评价及贝叶斯可信传播神经网络的药品短缺风险预警模型构建[J]. 药学前沿, 2025, 29(12): 2027-2035.SunYY, LiuJJ, NiY, et al. Construction of drug shortage risk early warning model based on risk matrix evaluation and Bayesian belief propagation neural network[J]. Frontiers in Pharmaceutical Sciences, 2025, 29(12): 2027-2035. DOI:10.12173/j.issn.2097-4922.202505060.

12. 黄哲, 于钟怡, 兰毅鹏, 等. 辽宁省短缺药品监测评价模型的构建研究[J]. 沈阳药科大学学报, 2022, 39(8): 1010-1017.HuangZ, YuZY, LanYP, et al. Construction of monitoring and evaluation model for drug shortage in Liaoning Province[J]. Journal of Shenyang Pharmaceutical University, 2022, 39(8): 1010-1017. DOI:10.14066/j.cnki.cn21-1349/r.2020.1090.

13. DragomiretskiyK, ZossoD. Variational Mode Decomposition[J].IEEE Transactions on Signal Processing, 2014, 62(3): 531-544. DOI:10.1109/TSP.2013.2288675.

14. Sowmya R, Premkumar M, Jangir P. Newton-raphson-based optimizer: a new population-based metaheuristic algorithm for continuous optimization problems[J].Eng Appl Artif Intell, 2024, 128(000): 44. DOI:10.1016/j.engappai.2023.107532.

15. SuykensJAK, Van GestelT, De BrabanterJ, et al. Least squares support vector machines[M]. Singapore: World Scientific, 2002: 1-50. doi：10.1142/5089

16. BoxGEP, JenkinsGM, ReinselGC, et al. Time series analysis: forecasting and control, 5th ed[M]. Hoboken: Wiley, 2015: 50-55.

17. HyndmanRJ, AthanasopoulosG. Forecasting: principles and practice, 3rd ed[M]. Melbourne, Australia: OTexts, 2021: 10-12.

Popular papers
Last 6 months