Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
An integrated framework for multi-objective decision support in retail inventory management
Thesis (PhD)--Stellenbosch University, 2023.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2023
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867614029906706432 |
|---|---|
| access_status_str | Open Access |
| author | Zietsman, Hans Jurie |
| author2 | Van Vuuren, Jan Harm
|
| author_browse | Van Vuuren, Jan Harm
Zietsman, Hans Jurie |
| author_facet | Van Vuuren, Jan Harm
Zietsman, Hans Jurie |
| author_sort | Zietsman, Hans Jurie |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (PhD)--Stellenbosch University, 2023. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/126934 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA en_ZA |
| last_indexed | 2026-06-10T12:45:32.686Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2023 |
| publishDateRange | 2023 |
| publishDateSort | 2023 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/126934 An integrated framework for multi-objective decision support in retail inventory management Zietsman, Hans Jurie Van Vuuren, Jan Harm Stellenbosch University. Faculty of Engineering. Dept. of Industrial Engineering. Inventories, Retail Inventory control Decision support systems Supply chain management Thesis (PhD)--Stellenbosch University, 2023. ENGLISH ABSTRACT: Globalisation and the growth of e-commerce have led to retail businesses having to manage an increasing number of stock keeping units. Moreover, the success of any retail company depends on how successfully it can satisfy customer demand for stock keeping units while remaining financially viable. Rule-based inventory management systems are therefore typically employed as the basis for decision making, aimed at achieving a desirable trade-off between balancing the conflicting objectives of providing high-quality customer service and minimising inventory and operating costs. As the number of stock keeping units of a retail business increases, however, so does the complexity of this balancing problem. The main decisions in respect of inventory management in a retail business, which affect this balance, are (1) which stock keeping unit inventories to replenish, (2) when to place replenishment orders, and (3) the appropriate volumes of stock keeping units to include in replenishment orders. High-quality decision making, in this regard, necessitates the utilisation of computerised decision support systems capable of accommodating the complexity presented by a large number of stock keeping units. An abundance of research has been dedicated to various facets of inventory management, such as product segmentation, forecasting of customer demand, and determining the sizes and timings of replenishment orders. Various frameworks have also been proposed for the development of decision support systems aimed at improved inventory management. Most of these existing frameworks are, however, aimed at single facets of the inventory management problem—generic, integrated frameworks are absent from the literature. In this dissertation, a holistic framework design is proposed which integrates all of the major components expected to form part of a generic framework for inventory management. The objective of the framework is ultimately to provide decision support in respect of a number of inventory replenishment operations with a view to achieve a desirable balance between the aforementioned conflicting objectives of inventory management. The framework integrates the processing of historical demand data, the selection of products aimed at preventing product proliferation, the prioritisation of groups of products by means of a multi-criteria clustering approach aimed at product segmentation, and the forecasting of future customer demand distributions based on historical data by means of a generic time series forecasting procedure (which includes time series clustering and model ensembling). The framework also facilitates the recommendation and application of inventory model parameters per stock keeping unit by means of a combinatorial optimisation approach as well as a user-friendly synthesis of recommended replenishment orders, based on pre-specified objectives. An instantiation of the framework is implemented on a personal computer as concept demonstration. The generic applicability of the framework is verified by evaluating the performance of the individual components embedded within the framework in respect of benchmark data sets sourced from the literature. The performances of the framework components are shown to be competitive with published results on these benchmark data. Moreover, the practical applicability of the proposed framework is showcased by applying the framework instantiation to two real-world case studies in the South African retail sector. AFRIKAANS OPSOMMING: Globalisering en die groei van e-handel het daartoe gelei dat kleinhandelondernemings ’n toenemende aantal voorraad-eenhede moet bestuur. Boonop hang die sukses van enige kleinhandelonderneming af van hoe suksesvol dit in die vraag na voorraad-eenhede van kli¨ente kan voorsien en terselfdertyd finansieel lewensvatbaar bly. Re¨el-gebaseerde voorraadbestuurstelsels word dus tipies as basis vir besluitneming gebruik, wat daarop gemik is om ’n wenslike afruiling tussen die teenstrydige doelstellings van die lewering van ho¨e-gehalte kli¨entediens en die minimering van voorraad- en bedryfskoste te behaal. Namate die aantal voorraad-eenhede van ’n kleinhandelonderneming egter toeneem, verhoog die kompleksiteit van hierdie balanseringsprobleem. Die belangrikste besluite ten opsigte van voorraadbestuur in ’n kleinhandelonderneming, wat hierdie balans be¨ınvloed, is (1) watter voorraad-eenhede aangevul behoort te word, (2) wanneer aanvullingsbestellings geplaas behoort te word, en (3) die toepaslike volumes voorraad-eenhede om in aanvullingsbestellings in te sluit. Ho¨e-gehalte besluite in hierdie verband noodsaak die gebruik van gerekenariseerde besluitsteunstelsels wat die kompleksiteit wat deur ’n groot aantal voorraad-eenhede veroorsaak word, kan akkommodeer. ’n Oorvloed van navorsing is al gewy aan verskillende fasette van voorraadbestuur, soos byvoorbeeld produksegmentering, die voorspelling van kli¨ente-vraag en die bepaling van die groottes en tydsberekening van aanvullingsbestellings. Verskeie raamwerke is ook voorgestel vir die ontwikkeling van besluitsteunstelsels gerig op verbeterde voorraadbestuur. Die meeste van hierdie bestaande raamwerke is egter gemik op enkele fasette van die voorraadbestuursprobleem — generiese, ge¨ıntegreerde raamwerke is nie in die literatuur beskikbaar nie. In hierdie proefskrif word ’n holistiese raamwerkontwerp voorgestel wat al die belangrikste komponente integreer wat na verwagting deel behoort uit te maak van ’n generiese raamwerk vir voorraadbestuur. Die doel van die raamwerk is uiteindelik om besluitsteun te bied ten opsigte van ’n aantal voorraadaanvullingstake met die doel om ’n wenslike balans tussen die bogenoemde botsende doelstellings in voorraadbestuur te bereik. Die raamwerk integreer die verwerking van historiese vraagdata, die keuse van produkte met die doel om produk vermenigvuldiging te voorkom, die prioritisering van groepe produkte deur middel van ’n multikriteria-groeperingsbenadering wat gerig is op produksegmentering, en die voorspelling van toekomstige kli¨ente-vraag verspreiding gebaseer op historiese data deur middel van ’n generiese tydreeksvooruitskattingsprosedure (wat tydreeksgroepering en modelversameling insluit). Die raamwerk fasiliteer ook die aanbeveling en toepassing van voorraadmodel-parameters per voorraad-eenheid deur middel van ’n kombinatoriese optimeringsbenadering sowel as ’n gebruikersvriendelike sintese van aanbevole aanvullingsbestellings, gebaseer op vooraf-gespesifiseerde doelstellings. ’n Instansie van die raamwerk word op ’n persoonlike rekenaar as konsepdemonstrasie ge¨ımplementeer. Die generiese toepaslikheid van die raamwerk word geverifieer deur die prestasie van die individuele komponente van die raamwerk ten opsigte van maatstafdatastelle uit die literatuur te evalueer. Daar word getoon dat die prestasie van die raamwerkkomponente mededingend is met gepubliseerde resultate ten opsigte van hierdie maatstafdata. Boonop word die praktiese toepaslikheid van die voorgestelde raamwerk gedemonstreer deur die raamwerk-instansie op twee werklike gevallestudies in die Suid-Afrikaanse kleinhandelsektor toe te pas. Doctoral 2023-02-06T05:08:23Z 2023-05-18T06:56:30Z 2023-02-06T05:08:23Z 2023-05-18T06:56:30Z 2023-02 Thesis http://hdl.handle.net/10019.1/126934 en_ZA en_ZA Stellenbosch University xxxii, 345 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Inventories, Retail Inventory control Decision support systems Supply chain management Zietsman, Hans Jurie An integrated framework for multi-objective decision support in retail inventory management |
| title | An integrated framework for multi-objective decision support in retail inventory management |
| title_full | An integrated framework for multi-objective decision support in retail inventory management |
| title_fullStr | An integrated framework for multi-objective decision support in retail inventory management |
| title_full_unstemmed | An integrated framework for multi-objective decision support in retail inventory management |
| title_short | An integrated framework for multi-objective decision support in retail inventory management |
| title_sort | integrated framework for multi objective decision support in retail inventory management |
| topic | Inventories, Retail Inventory control Decision support systems Supply chain management |
| url | http://hdl.handle.net/10019.1/126934 |
| work_keys_str_mv | AT zietsmanhansjurie anintegratedframeworkformultiobjectivedecisionsupportinretailinventorymanagement AT zietsmanhansjurie integratedframeworkformultiobjectivedecisionsupportinretailinventorymanagement |