A decision support framework for the selection of appropriate time series forecasting methods in the retail sector

Simple item page
dc.contributor.advisorVan Vuuren, Jan Harm en_ZA
dc.contributor.advisorLindner, Berni G
dc.contributor.advisorDu Toit, Jacques
dc.contributor.authorGanzevoort, Reinard Christiaan
dc.contributor.otherStellenbosch University. Faculty of Engineering. Dept. of Industrial Engineering.en_ZA
dc.date.accessioned2024-02-13T12:58:20Z
dc.date.accessioned2024-04-27T02:16:52Z
dc.date.available2024-02-13T12:58:20Z
dc.date.available2024-04-27T02:16:52Z
dc.date.issued2023-10en_ZA
dc.descriptionThesis (PhD)--Stellenbosch University, 2024.en_ZA
dc.description.abstractENGLISH ABSTRACT:There is a significant trade-off in any high-turnover retail environment between limiting in-store inventory levels and mitigating the risk of stock-outs. This trade-off is caused by the typical aim of retail organisations to minimise the capital tied up in inventory without incurring a significant deterioration of their service levels (i.e. to ensure product availability for customers). In order, therefore, to better manage their inventories, retailers often consider the prediction of customer behaviour as a main priority. In practice, however, sales forecasting processes are usually automated to some extent and practitioners often have limited knowledge pertaining to the selection of appropriate forecasting methods. A generic framework is proposed in this dissertation for assisting retail forecasting practitioners in the selection of appropriate forecasting methods based on available time series data sets pertaining to retail sales. This forecasting framework takes as input a multivariate time series sales data set and facilitates the configuration, transformation and extraction of valuable information from these data in order to partition the data set into clusters of time series exhibiting similar attributes. The working of the framework is based on a generic, two-phased approach. One phase of the framework, called its benchmarking phase, involves establishing a benchmark data set (or updating it if it already exists) which can be leveraged to inform feature-based forecast model identification and ranking for different clusters of time series. The computationally efficient identification of a tailored shortlist of forecast models is thus facilitated during the other framework phase, called its implementation phase, for each sales time series presented to it by a retail organisation, based on the features of the time series presented. The two phases of the framework may be applied repeatedly in alternating fashion, thus enlarging the benchmark data set and improving its representativeness each time after having applied the implementation phase to the sales time series data of a new retail organisation. The framework is verified with reference to well-established retail sales benchmark data. The verified framework is employed to evaluate the difference in forecast quality and computational time, based on the benchmark data, that results from applying the forecasting methods recommended by the framework to newly presented retail timeseries data as opposed to exhaustively applying forecasting methods classified as traditional statistical techniques, machine learning techniques and ensemble techniques. The working of the framework is finally validated by applying computerised instantiations thereof to real-world data sets of time series representing retail sales.en_ZA
dc.description.abstractAFRIKAANSE OPSOMMING: Daar is ’n beduidende afruiling in enige ho¨e-omset kleinhandelomgewing tussen die beperking van voorraadvlakke in die winkel en die vermindering van risiko te wyte aan voorraad tekorte. Hierdie afruiling word veroorsaak deur die tipiese doel van kleinhandelorganisasies om die kapitaal wat in voorraad vasgebind is, te minimeer sonder om ’n beduidende verswakking in hul diensvlakke teweeg te bring (m.a.w. om produkbeskikbaarheid vir kli¨ente te verseker). Ten einde hul voorraad beter te bestuur, beskou kleinhandelaars daarom dikwels die voorspelling van kli¨entegedrag as ’n hoofprioriteit. In die praktyk word verkoopsvoorspellingsprosesse egter gewoonlik tot ’n mate ge-outomatiseer en praktisyns het dikwels beperkte kennis met betrekking tot die keuse van toepaslike voorspellingsmetodes. ’n Generiese raamwerk word in hierdie proefskrif voorgestel om leinhandelvooruitskattingspraktisyns by te staan met die keuse van toepaslike vooruitskattingsmetodes gebaseer op beskikbare tydreeksdatastelle wat met kleinhandelverkope verband hou. Hierdie voorspellingsraamwerk neem ’n meerveranderlike tydreeksverkoopdatastel as inset en fasiliteer die konfigurasie, transformasie en onttrekking van waardevolle inligting uit die data om die datastel in klasse tydreekse te verdeel wat soortgelyke eienskappe vertoon. Die werking van die raamwerk is gebaseer op ’n generiese, twee-fase benadering. Een fase van die raamwerk, bekend as die maatstaf daarstellingsfase, behels die daarstel van ’n maatstafdatastel (of die opdatering daarvan as dit reeds bestaan) wat aangewend kan word om kenmerk-gebaseerde voorspellingsmodel-identifikasie en gepaardgaande rangordes vir verskillende klasse tydreekse te bepaal. Die berekeningsdoeltreffende identifikasie van ’n pasgemaakte kortlys van voorspellingsmodelle word sodoende tydens die ander raamwerkfase, bekend as die implementeringsfase, vir elke verkoopstydreeks gefasiliteer wat deur ’n kleinhandelorganisasie aangebied word, gebaseer op die kenmerke van die tydreekse aangebied. Die twee fases van die raamwerk kan herhaaldelik op afwisselende wyse toegepas word, om sodoende die maatstafdatastel te vergroot en die verteenwoordigendheid daarvan elke keer te verbeter nadat die implementeringsfase op die verkoopstydreeksdata van ’n nuwe kleinhandelorganisasie toegepas is. Die raamwerk word met verwysing na goed-gevestigde kleinhandelverkope-maatstafdata geverifieer. Die geverifieerde raamwerk word gebruik om, gebaseer op die maatstafdata, die verskil in voorspellingskwaliteit en berekeningstyd wat voortspruit uit die toepassing van die voorspellingsmetodes wat deur die raamwerk aanbeveel word, in die konteks van nuwe kleinhandeltydreeksdata te evalueer eerder as om bloot vooruitskattingsmetodes wat as tradisionele statistiese tegnieke, masjienleertegnieke en ensembletegnieke geklassifiseer is op ’n uitputtende wyse volledig toe te pas. Die werking van die raamwerk word uiteindelik gevalideer deur die toepassing van gerekenariseerde instansiasies daarvan op werklike datastelle van tydreekse wat kleinhandelverkope verteenwoordig.en_ZA
dc.description.versionDoctorateen_ZA
dc.format.extentxxiv, 254 pages : illustrations.en_ZA
dc.identifier.urihttps://scholar.sun.ac.za/handle/10019.1/130676
dc.language.isoen_ZAen_ZA
dc.language.isoen_ZAen_ZA
dc.publisherStellenbosch : Stellenbosch Universityen_ZA
dc.rights.holderStellenbosch Universityen_ZA
dc.subject.lcshDecision support systemsen_ZA
dc.subject.lcshInventories, Retailen_ZA
dc.subject.lcshRetail trade -- Forecastingen_ZA
dc.subject.lcshConjoint analysis (Marketing)en_ZA
dc.subject.lcshUCTDen_ZA
dc.titleA decision support framework for the selection of appropriate time series forecasting methods in the retail sectoren_ZA
dc.typeThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
ganzevoort_decision_2024.pdf
Size:
8.44 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Doctoral Degrees (Industrial Engineering)