Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Creating a Back Stock to Increase Order Delivery and Pickup Availability
KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Health Informatics and Logistics.
KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Health Informatics and Logistics.
2019 (English)Independent thesis Basic level (university diploma), 10 credits / 15 HE creditsStudent thesisAlternative title
Framtagning av ett baklager för att öka tillgängligheten av leverans och upphämtning av ordrar (Swedish)
Abstract [en]

Apotek Hjärtat wants to keep developing their e-commerce website and improve retrieval and delivery of orders to customers. Click and Collect and Click and Express are two options for retrieving e-commerce orders that are available if all products in the order are present in the store. By implementing a back stock in the stores with popular e-commercial items, all products of an order will more often be present in the store. The back stock will in such a way increase the availability of Click and Collect and Click and Express. The goals for the study are to conduct a pilot study, compare methods and possible solutions to implement a model to reach the goals. The pilot study was made by studying previous works in mathematical statistics methods and machine learning methods. The statistical method was accomplished through the analytical tool Statistical Package for the Social Sciences (SPSS) and Java. The machine learning method was accomplished through Python and the Scikit-learn library. The machine learning method was performed by a regression algorithm that was used to find relations between category sales and pollen forecasts. The statistical and machine learning methods were compared to each other. Both gave identical results, but the machine learning method was more functional and easier to further develop and consequently was chosen. Several models were created for a few selected product categories. The categories that did not work for the models had an unrealistic amount of sold products. These amounts could be negative or extremely high when unknown inputs were introduced. A simulation was made of the back stock to estimate how it would increase the availability of Click and Collect/Click and Express. The machine learning models could need more data for more accurate predictions. A conclusion could be made though that is possible to predict the amount of sold products of certain categories such as Allergy and Child Medicine with pollen halt taken into account.

Abstract [sv]

Apotek Hjärtat vill fortsätta utveckla sin e-handelssida och förbättra upphämtning och leverans av ordrar till kund. Click and Collect och Click and Express är två val för att hämta upp e-handelsordrar som finns tillgängliga om alla produkter i ordern finns i butik. Genom att implementera ett baklager i butiker med populära unika ehandelsprodukter kommer alla produkter i en order oftare att finnas i butik. Baklagret kommer på så vis öka tillgängligheten av Click and Collect och Click and Express. Målen är att utföra en förstudie, samt att jämföra och hitta en bra lösning att implementera en modell för att uppnå målen. Förstudien gick ut på att analysera tidigare arbeten inom matematiska statistikmetoder och maskininlärningsmetoder. Den statistiska metoden utfördes genom det analytiska verktyget Statistical Package for the Social Sciences (SPSS) och Java. Maskininlärningsmetoden utvecklades med hjälp av Python och Scikit-learn biblioteket. Maskinlärningsmetoden utfördes genom en regressionsalgoritm som användes för att ta fram flera modeller för relationer mellan försäljning av kategorier och pollenprognoser. Statistiska metoden och maskininlärningsmetoden jämfördes med varandra. Båda gav identiska resultat men maskininlärning var mer funktionellt och enklare att vidareutveckla och därför valdes den metoden. Flera olika modeller lyckades tas fram för en del produktkategorier. De kategorier som inte fungerade för modellerna hade orealistiska mängder sålda varor. Dessa mängder kunde vara negativa eller extremt höga när okända inputs introducerades. Med hjälp av simulationen var det möjligt att uppskatta hur baklagret skulle öka tillgängligheten av Click and Collect/Express. Maskininlärningsmodellerna skulle behöva mer data, som kommer i framtiden, för att ge en mer precis prediktering mellan pollenvärden. Som slutsats är det möjligt att använda dem i framtiden för vissa kategorier som allergi och barnmedicin.

Place, publisher, year, edition, pages
2019. , p. 39
Series
TRITA-CBH-GRU ; 2019:030
Keywords [en]
e-commerce, back stock, statistics, supervised machine learning, linear regression, Scikit-learn
Keywords [sv]
e-handel, baklager, statistik, övervakad maskininlärning, linjär regression, Scikit-learn
National Category
Computer Engineering
Identifiers
URN: urn:nbn:se:kth:diva-252798OAI: oai:DiVA.org:kth-252798DiVA, id: diva2:1321924
Subject / course
Mathematical Statistics
Educational program
Bachelor of Science in Engineering - Computer Engineering
Supervisors
Examiners
Available from: 2019-06-10 Created: 2019-06-10 Last updated: 2019-06-10Bibliographically approved

Open Access in DiVA

Creating a Back Stock - John Nguyen and Kasper Lindén(1419 kB)15 downloads
File information
File name FULLTEXT01.pdfFile size 1419 kBChecksum SHA-512
c75d561edf83e9d3a5138d8e6651a741577424084aafbd0d0a7f0dcbe3d83c10f6b22e3709a273705e5138106acc6bb8c686c7276c1fc5581d45c5b8ef8a1015
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Nguyen, JohnLindén, Kasper
By organisation
Health Informatics and Logistics
Computer Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 15 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 107 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf