将主题分布特征集成到预测和库存优化中:来自台湾血液供应系统的证据
Tsung-Hsi Wang1, Tzu-Chien Wang2, Chia-Kai Li3
1Taiwan Blood Services Foundation, Taiwan, ROC; College of Management, National Taiwan University, Taiwan, ROC.
Journal of the Formosan Medical Association = Taiwan yi zhi
|January 15, 2026
概括
这项研究引入了献血的新型预测框架,集成文本数据来预测供应波动. 这种方法显著提高了准确性,提高了血液供应链的弹性.
科学领域:
- 医疗保健 运营 研究 研究 研究
- 数据科学数据科学数据科学
- 公共卫生 公共卫生
背景情况:
- 献血量由于流行病和假日等外部事件而表现出高波动.
- 传统的统计模型很难解释血液供应中的事件驱动的冲击.
- 来自新闻和社交媒体的非结构化文本数据包含有价值的信息,用于预测捐款波动.
研究的目的:
- 为献血开发一个语义增强的预测框架.
- 整合结构化捐赠数据与非结构化文本数据,以改善预测.
- 为了提高血液供应链的弹性,应对不可预测的需求.
主要方法:
- 基于基准的主题建模方法,根据其性能选择BERTopic.
- 汇总文档级主题分布,使用主要组件分析 (PCA) 进行压缩,并转化为滞后预测器.
- 结合语义特征与结构化数据来训练梯度提升,XGBoost,LightGBM,CatBoost,随机森林,GAM和SARIMAX模型.
主要成果:
- 语义特征显著提高了预测准确性,梯度提升实现了最佳性能.
- 观察到根平均平方误差 (RMSE) 减少了28%,R-squared增加了0.33.
- 莎普利添加剂研究 (SHAP) 确定了关键预测因素,包括活动指标,移动捐赠网站,捐赠者人口统计数据以及与流行病/灾害相关的主题.
结论:
- "预测-然后-优化"管道通过结合语义见解来提高血液供应的弹性.
- 该框架可将其应用于其他面临事件驱动波动的医疗保健或易损坏的供应链.
- 准确预测献血量对于有效的库存管理和服务水平优化至关重要.
相关概念视频
Distribution Reliability and Automation
497
Distribution reliability in electrical power systems is critical for ensuring an uninterrupted power supply to consumers at minimal cost. According to IEEE Standard Terms, reliability is the probability that a device will function without failure over a specified time period or amount of usage. For electric power distribution, this translates to maintaining continuous power supply and addressing customer concerns over power outages. Several indices, as defined by IEEE Standard 1366-2012, are...
497
Applications of Integration to Find Blood Flow
17
Blood flow through a cylindrical blood vessel can be mathematically described using the principles of laminar flow, a regime in which fluid moves smoothly in parallel layers. In this model, the velocity of the blood is not uniform across the cross-section of the vessel; rather, it varies with the radial distance from the center. The maximum velocity occurs along the central axis, decreasing progressively toward the vessel walls, where it reaches zero due to viscous drag.Approximating Blood...
17
Drug Distribution: Volume of Distribution
7.2K
The volume of distribution refers to the theoretical volume necessary to contain the entire amount of an administered drug at the same concentration observed in the blood plasma. The body's intracellular fluid compartment, which makes up two-thirds of the total body water, is contrasted with the extracellular fluid compartment—comprising plasma and interstitial fluid—that accounts for one-third. The volume of distribution can vary depending on the characteristics of the drug.
7.2K
Distributions to Estimate Population Parameter
5.0K
The accurate values of population parameters such as population proportion, population mean, and population standard deviation (or variance) are usually unknown. These are fixed values that can only be estimated from the data collected from the samples. The estimates of each of these parameters are sample proportion, the sample mean, and sample standard deviation (or variance). To obtain the values of these sample statistics, data are required that have particular distribution and central...
5.0K
Drug Distribution as One-Compartment Model and Elimination by Nonlinear Pharmacokinetics: Overview
335
Drug administration can occur through various routes, each of which may result in a different process of elimination. This process is often mixed with nonlinear and linear processes. It's important to understand that a single drug can be metabolized into different metabolites through parallel processes.
For instance, consider the metabolism of sodium salicylate. This compound is metabolized into two distinct substances: a glucuronide and a glycine conjugate. The rate of conjugation depends...
For instance, consider the metabolism of sodium salicylate. This compound is metabolized into two distinct substances: a glucuronide and a glycine conjugate. The rate of conjugation depends...
335
Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
242
Pharmacokinetic models are mathematical constructs that represent and predict the time course of drug concentrations in the body, providing meaningful pharmacokinetic parameters. These models are categorized into compartment, physiological, and distributed parameter models.
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
242


