主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2025, Vol. 33 ›› Issue (10): 350-360.doi: 10.16381/j.cnki.issn1003-207x.2023.0017

• • 上一篇    下一篇

代际更替电子产品的废弃量预测与治理:小数据视角

王方1, 程文鑫1, 余乐安2(), 查锐3   

  1. 1.西安电子科技大学经济与管理学院,陕西 西安 710126
    2.四川大学商学院,四川 成都 610065
    3.哈尔滨工程大学经济管理学院,黑龙江 哈尔滨 150001
  • 收稿日期:2023-01-04 修回日期:2023-03-21 出版日期:2025-10-25 发布日期:2025-10-24
  • 通讯作者: 余乐安 E-mail:yulean@amss.ac.cn
  • 基金资助:
    国家自然科学基金项目(72001165);陕西省创新能力支撑计划项目(2022SR5016);西安市科技计划项目软科学研究重点项目(2023JH-RKXZD-0012)

Predicting and Managing E-waste with Characteristics of Generational Replacement: A Perspective of Small Data

Fang Wang1, Wenxin Cheng1, Lean Yu2(), Rui Zha3   

  1. 1.School of Economics & Management,Xidian University,Xi’an 710126,China
    2.Business School,Sichuan University,Chengdu 610065,China
    3.School of Economics and Management,Harbin Engineering University,Harbin 150001,China
  • Received:2023-01-04 Revised:2023-03-21 Online:2025-10-25 Published:2025-10-24
  • Contact: Lean Yu E-mail:yulean@amss.ac.cn

摘要:

针对小数据代际更替电子产品的废弃量预测问题,提出了基于SIR(susceptible-infective-removed)传染病模型思想的预测方法。松弛SIR模型的约束条件,基于新代电子产品时间序列的小数据特征,构建了时间加权平均误差最小的STa(sales-transfer-adjustment)优化模型;引入PSO(particle swarm optimization)算法对STa模型进行参数估计,按照差分补偿预测的思想,确定转移量的合理滞后期,提高了对新代电子产品量的预测精度;立足电子产品数据的可得性,以电子废弃物生成量估计模型(estimation model of waste quantity of electronic products,EWE)实现了对新代电子产品废弃量的预测。基于我国农村电视机和手机4个数据集的实证分析表明,所构建的STa·PSO-EWE模型的性能总体优于其余基准对比模型。

关键词: 电子废弃物, 传染病模型, 小数据, 代际更替, 预测

Abstract:

To address the issue of predicting the quantity of waste electronic products with small data, a prediction method based on the Susceptible-Infective-Removed (SIR) infectious disease model is proposed. Taking into account the small data characteristics of the time series of new generation electronic products, a Sales-Transfer-Adjustment (STa) optimization model with the minimum time-weighted average error is constructed. To estimate the parameters of the STa model, the Particle Swarm Optimization (PSO) algorithm is introduced. Furthermore, the reasonable lag time of the transfer quantity is determined via the concept of differential compensation prediction, which enhances the prediction accuracy of the new generation electronic product quantity. Based on the availability of electronic product data, the Estimation Model of Waste Quantity of Electronic Products (EWE) is used to predict the waste quantity of new generation electronic products. Through empirical analysis of 4 data sets of rural TV sets and mobile phones in China, it is found that the performance of the STa·PSO-EWE model is generally superior to that of other benchmark comparison models.

Key words: electronic waste, infectious disease model, small data, generational replacement, prediction

中图分类号: