文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 工程科技 > 电子/电路 > Density Forecasting of Intraday Call Center Arrivals Using Models Based on Exponential Smoothing
侵权投诉

Density Forecasting of Intraday Call Center Arrivals Using Models Based on Exponential Smoothing

Vol.58,No.3,March2012,pp.534–549

ISSN0025-1909(print) ISSN1526-5501(online)

MANAGEMENT SCIENCE

doc.guandang.net

©2012INFORMS

DensityForecastingofIntradayCallCenterArrivalsUsingModelsBasedonExponentialSmoothing

JamesW.Taylor

SaïdBusinessSchool,UniversityofOxford,OxfordOX11HP,UnitedKingdom,

james.taylor@sbs.ox.ac.uk

A

keyinputtothecallcenterstaf ngprocessisaforecastforthenumberofcallsarriving.Densityforecastsofarrivalratesareneededforanalyticalcallcentermodels,whichassumePoissonarrivalswithastochasticarrivalrate.Densityforecastsofcallvolumescanbeusedinsimulationmodelsandarealsoimportantfortheanalysisofoutsourcingcontracts.Aforecastingmethod,whichhaspreviouslyshownstrongpotential,isHolt–Wintersexponentialsmoothingadaptedformodelingtheintradayandintraweekcyclesinintradaydata.Toenabledensityforecastingofthearrivalvolumeandrate,wedevelopaPoissoncountmodel,withgammadistributedarrivalrate,whichcapturestheessentialfeaturesofthisexponentialsmoothingmethod.Theapparentstationarylevelinourdataleadsustodevelopversionsofthenewmodelforserieswithstationarylevels.Weevaluateforecastaccuracyuptotwoweeksaheadusingdatafromthreeorganizations.We ndthatthestationarylevelmodelsimprovepredictionbeyondapproximatelytwodaysahead,andthatthesemodelsperformwellincomparisonwithsophisticatedbenchmarks.Thisiscon rmedbytheresultsofacallcentersimulationmodel,whichdemonstratestheuseofarrivalratedensityforecastingtosupportstaf ngdecisions.Keywords:callcenters;arrivalrate;densityforecasting;exponentialsmoothing;seasonality

History:ReceivedMarch26,2009;acceptedJuly21,2011,byYossiAviv,operationsmanagement.PublishedonlineinArticlesinAdvanceOctober14,2011.

1.Introduction

Thenumberandsizeofcallcentershasincreasedinrecentyearstotheextentthattheyemployapproxi-mately3%oftheworkingpopulationsintheUnitedStatesandtheUnitedKingdom.Formanyorgani-zations,callcentersprovidethemainchannelofcommunicationwiththeircustomers.Asthehumanresourcecostsforcallcenterstypicallycomprise60%to80%oftheiroveralloperatingbudget,itisofgreatimportancetoestablishanoptimallevelofstaf ng.Overstaf ngobviouslyincursunnecessaryoperatingcosts,whereasunderstaf ngcausesqueuingtimesconsideredbycustomersasunacceptable,whichcanleadtotheabandonmentofcalls.Indeed,staffdeploy-mentisabalancebetweenservicequalityandoper-atingcosts(Ak¸sinetal.2007).Akeyinputtotheschedulingprocessisaforecastforthenumberofcallsarrivingatthecallcenter.Long-termpredictionsareneededforrecruitmentpurposes.Forecastswithleadtimesofseveralweeksarerequiredforstaffschedul-ing,andtheseforecastsandschedulesneedtobereg-ularlyreviseduntilthetargetdayitself(Gansetal.2003).Thiscanbecontinuedwithinthedaywithfore-castsforseveralhoursaheadusedtoenabledynamicupdatingoftheagents’deployment(Mehrotraetal.2010).Theusefulnessofintradayforecastupdating(2007)toextendtheirforecastingmethodsforthispurpose.

Timeseriesofintradaycallcenterarrivalsconsistofvariabilityaroundaseasonalpatternthatismadeupofbothintraweekandintradayseasonalcycles.Antic-ipatingthisvariabilityisimportant,becausehigheruncertaintyrequireshigherstaf nglevelsinordertomeetspeci edservicelevelobjectives.Apredic-tionoftheprobabilitydensityfunction(i.e.,aden-sityforecast)ofthecallvolumecanbeusedforthearrivalprocessinsimulationmodels,whichhavebeengrowinginpopularitydue,atleastinpart,totheincreasedcomplexityofcallcenteroperationsanddevelopmentsincomputationalpower(Gansetal.2003,MehrotraandFama2003).Densityforecastsofcallvolumescanalsobeusedtoanalyzecallcenteroutsourcingcontracts(see,forexample,Ak¸sinetal.2008).Contractsmaybede nedsothatthecallvol-umeaboveaspeci edbaselevelisoutsourced,orsothatonlyabaseloadisoutsourced.

Foranalyticalcallcentermodels,andformanysimulationmodels,thearrivalvolumeismodeledasaPoissonprocess,andsoaforecastisneededforthearrivalrate.Empiricalevidencehasshownthatcallcenterarrivalsshowsigni cantoverdisper-sionrelativetoaPoissonprocess(i.e.,thevarianceisgreaterthanthemean),andthishasledtothearrival

第1页

你可能喜欢

热门文档

相关文档

  • Particle density

    Particle density_英语学习_外语学习_教育专区。Particle density's operation: 1. Archimedes' principle The powder is placed inside of known volume, Particle...

  • Density Column-CH

    LLOY D密度柱 Density Column 1 什么是密度柱 ? 一套密度柱系统包括:– 一组垂直的密封玻璃管 (柱形) – 一个具有恒温器、搅拌器和恒温螺旋管 的有机玻璃罩...

  • density_matrix

    16页 免费 New Trends in Density Ma... 暂无评价 54页 免费 CALCULATION OF THE DE... 暂无评价 10页 免费喜欢此文档的还喜欢...

  • Density

    Density_理学_高等教育_教育专区。什么是种群密度Density 组员:陈颖沛,邓乃中,林越锋 What is density? Density is defined as the number of individuals per ...

  • Copper density_20130304

    Copper density_20130304_电子/电路_工程科技_专业资料。copper density减少阻抗coupon与内容物差异偏移的考量项目 Ground layer & Signal layer 分 扣掉ground 1.残...

  • 如何为一个层设置DotDensity Renderer

    本例要实现的是为一个层设置DotDensity Renderer。一般情况下,用一个Dot Density Fill Symbol和一个Dot Density Renderer来描述层中某一特征的分布密度。本例中描述...

  • Defect Density

    Defect Density_专业资料。"15 years of measuring the effects of defect on scientific and other software: How do we satisfy customers in the long-...

  • Density and Specific Gravity

    Density and Specific Gravity (1) Density (r) of a material is defined as mass per unit volume and is usually -3 -3 represented in the SI system ...

  • USP34-NF bulk density

    页码,1/6 616 BULK DENSITY AND TAPPED DENSITY OF POWDERS BULK DENSITY This general chapter has been harmonized with the corresponding texts of the ...

  • Bulk Density and Tapped Density of Powders

    616? Bulk Density and Tapped Density of Powders 5639 Figure 3. Method I Apparatus—The apparatus (Figure 3) consists of the following: ? A 250-mL ...

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email