呼叫中心到达过程的建模与预测

doi:10.3969/j.issn.0253-2778.2017.09.009

中国科学技术大学学报 ›› 2017, Vol. 47 ›› Issue (9): 770-777.DOI: 10.3969/j.issn.0253-2778.2017.09.009

• 研究论文:管理科学与工程 • 上一篇下一篇

呼叫中心到达过程的建模与预测

霍加冕，谢金贵

1.中国科学技术大学管理学院工商管理系，安徽合肥230026; 2.中国科学技术大学管理学院管理科学系，安徽合肥230026

收稿日期:2016-04-07 修回日期:2016-11-29 接受日期:2016-11-29 出版日期:2023-03-27 发布日期:2016-11-29
通讯作者: 谢金贵
作者简介:霍加冕，男，1992生，硕士.研究方向：计量经济学.E-mail：hjm1992@mail.ustc.edu.cn
基金资助:
国家自然科学基金（71501196）资助.

Modelling and forecasting of call center arrival process

HUO Jiamian, XIE Jingui

1. Department of Business Administration, School of Management, University of Science and Technology of China, Hefei 230026, China; 2. Department of Management Science, School of Management, University of Science and Technology of China, Hefei 230026, China

Received:2016-04-07 Revised:2016-11-29 Accepted:2016-11-29 Online:2023-03-27 Published:2016-11-29

摘要/Abstract

摘要： 呼叫中心到达量的拟合和预测对呼叫中心人力需求计算和排班有着重要作用.根据用户在呼叫中心的不同阶段，可以将到达量分为交互式语音应答 (interactive voice response, IVR)到达量和人工服务到达量.国外学者主要对人工服务到达量进行拟合和预测，非齐次泊松过程作为一种模拟到达过程的方法得到广泛使用.而这里首次提出对IVR到达量进行研究，对比实际数据与非齐次泊松过程的均值和方差，发现该呼叫中心IVR到达量呈现“过离散(overdispersion)”现象，不能使用泊松分布拟合.因此选择时间序列模型对IVR到达量进行拟合和预测，用残差的白噪声检验拟合效果，平均绝对误差 (mean absolute error, MAE)值判断模型预测优劣.最后使用线性回归模型分析IVR到达量与人工服务到达量之间的关系.结果表明：自回归滑动平均ARIMA (autoregressive integrated moving average) (1,0,1)模型能更好地对该呼叫中心正常天数(normal days, ND)的IVR到达量进行短期预测，而Winters指数平滑法能更好地对春节期间(Spring Festival, SF)的IVR到达量进行短期预测；通过回归模型可以预测呼叫中心人工服务到达量.

关键词: 交互式语音应答到达量, 过离散, 时间序列模型, 人工服务到达量

Abstract: Fitting arrival process and forecasting future arrivals are crucial to staffing and scheduling in call centers. According to different stages of a call center, arrivals are classified into IVR arrival and agent arrival. Non-homogeneous Poisson process has been widely used overseas for modeling stochastic agent arrival process. However, this study initially proposed IVR arrival fitting and forecasting. The IVR arrival process of this call center appears to be “overdispersed” when comparing the mean arrival rate and its variance with the corresponding Poisson process. Therefore, time series was used to model and predict the IVR arrival process. White noise test of residuals was applied and the MAE (mean absolute error) was adopted to evaluate the goodness of fit. The results show that ARIMA (1,0,1) is preferable for predicting the IVR arrival in a short period of normal days and Winters is preferable for the Spring Festival period. Finally, the regression method was employed to describe the relationship between IVR arrival and agent arrival, and predict the agent arrival.

Key words: IVR arrival, overdispersion, time series, agent arrival

中图分类号:

TP391

霍加冕, 谢金贵. 呼叫中心到达过程的建模与预测[J]. 中国科学技术大学学报, 2017, 47(9): 770-777.

HUO Jiamian, XIE Jingui. Modelling and forecasting of call center arrival process[J]. Journal of University of Science and Technology of China, 2017, 47(9): 770-777.

参考文献

［1］
GANS N, KOOLE G, MANDELBAUM A. Telephone call centers: Tutorial, review, and research prospects[J]. Manufacturing and Service Operations Management, 2003, 5(2): 79-141.
[2] SHEN H, HUANG J Z. Forecasting time series of inhomogeneous Poisson processes with application to call center work force management[J]. Annals of Applied Statistics, 2008, 2(2): 601-623.
[3] BROWN L, GANS N, MANDELBAUM A,et al. Statistical analysis of a telephone call center: A queueing-science perspective[J]. Journal of the American Statistical Association, 2005, 100(469): 36-50.
[4] KIM S H, WHITT W. Choosing arrival process models for service systems: Tests of a nonhomogeneous Poisson process[J]. Naval Research Logistics, 2014, 61: 66-90.
[5] KIM S H, WHITT W. Are call center and hospital arrivals well modeled by nonhomogeneous Poisson process? [J]. Manufacturing and Service Operations Management, 2014,16(3): 464-480.
[6] AVRAMIDIS A N, DESLAURIERS A, L’ECUYER P. Modeling daily arrivals to a telephone call center[J]. Management Science, 2004, 50(7): 896-908.
[7] SHEN H. Statistical analysis of call-center operational data: Forecasting call arrivals, and analyzing customer patience and agent service[C]// Wiley Encyclopedia of Operation Research and Management Science, 2010.New York: Wiley, 2010.
[8] MAMAN S, MANDELBAUM A, WHITT W, et al. Queues with random arrival rates: Interface, modelling and asymptotics (C-staffing)[R]. Work in Progress, 2015.
[9] CHANNOUF N, L’ECUYER P. A normal copula model for the arrival process in a call center[J]. International Transactions in Operational Research, 2012, 19: 771-781.
[10] SHEN H, HUANG J Z. Interday forecasting and intraday updating of call center arrivals[J]. Manufacturing and Service Operations Management, 2008, 10(3):391-410.
[11] JAOUA A, L’ECUYER P, DELORME L. Call type dependence in multiskill call centers[J]. Journal
Simulation, 2013, 89(6): 722-734.
[12] THOMPSON H E, TIAO G C. Analysis of telephone data: A case study of forecasting seasonal time series[J]. The Bell Journal of Economics and Management Science, 1971, 2(2): 515-541.
[13] MABERT V A. Short interval forecasting of emergency phone call work loads[J]. Journal of Operations Management, 1985, 5(3): 259-271.
[14] TAYLOR J W. A comparison of univariate time series methods for forecasting intraday arrivals at a call center[J]. Management Science, 2008, 54(2): 253-265.
[15] IBRAHIM R, L’ECUYER P. Forecasting call center arrivals: Fixed-effects, mixed-effects, and bivariate models[J]. Manufacturing and Services Operations Management, 2013, 15(1): 72-85.
[16] AKTEKIN T, SOYER R. Call center arrival modeling: A Bayesian state space approach[J]. Naval Research Logistics, 2011, 58(1): 28-42.
[17] SHEN H. Exponentially weighted methods for forecasting intraday time series with multiple seasonal cycles: Comments[J]. International Journal of Forecasting, 2010, 58: 652-654.
[18] TAYLOR J W. Density forecasting of intraday call center arrivals using models based on exponentially smoothing[J]. Management Science, 2012, 58(3): 534-549.
[19] 李诗羽, 张飞, 王正林. 数据分析: R语言实战[M]. 北京：电子工业出版社, 2014: 260-286.
[20] 黄荣坦. ARIMA模型建立与应用[EB/OL]. [2016-04-07]. http://wenku.baidu.com/view/ d88f12f3941ea76e58fa0484.html.
[21] Wikipedia. Overdispersion [EB/OL]. [2016-04-07]. https://en.wikipedia.org/wiki/ Overdispersion.

呼叫中心到达过程的建模与预测

Modelling and forecasting of call center arrival process

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价