Towards Better Demand Forecasting Using Artificial Intelligence

Examensarbete på kandidatnivå

Please use this identifier to cite or link to this item:
Download file(s):
File Description SizeFormat 
TEKX04-20-18.pdf741.12 kBAdobe PDFThumbnail
Bibliographical item details
Type: Examensarbete på kandidatnivå
Title: Towards Better Demand Forecasting Using Artificial Intelligence
Authors: Cederberg, Line
Elliot, Josefin
Engdahl, Alva
Hedenborg, Fredrik
Sikström, Jakob
Wranå, Linnea
Åberg, Johanna
Abstract: Balancing customer demand and supply capacity is crucial for business survival, especially in competitive environments that are subject to short product life cycles and require minimal waste. Most high performing companies rely on demand forecasting to better balance demand and supply, but still embrace significant errors. In this respect, many emerging technologies are finally becoming mature, opening up a new world of potential improvements that forecasting can benefit from. One of these technologies is artificial intelligence (AI), whereby tasks that previously required human cognition now can be solved better and more efficiently. Therefore, the study aims to increase the understanding regarding the potential improvements AI can bring to supply chain performance in the context of demand forecasting in manufacturing companies. Related literature was first reviewed to identify baseline knowledge and develop a theoretical framework. Afterwards, interviews with manufacturing companies in the region of Gothenburg were performed in order to gather qualitative data. The results from seven interviews show that AI within demand forecasting is not used to a wide extent in the region of Gothenburg. Businesses that have implemented AI have seen several improvements as a result, while companies still using traditional forecasting methods are facing challenges to realise potential improvements. Multiple prerequisites needed for implementing AI have been identified. Some prerequisites were identified in both literature and interviews, while others only appeared in one or the other. The conclusion is that AI is expected to improve; better downstream planning, quick adaptation to erratic demand, increased service levels, decreased working capital, and reduced manual work by being implemented in demand forecasting. The general prerequisites are; clean data, sufficient technological infrastructure, available resources, understanding of the improvements AI can bring, and sufficient competence of AI.
Keywords: Artificial Intelligence;Demand Forecasting;Improvements;Prerequisites;Machine Learning;Operations planning
Issue Date: 2020
Publisher: Chalmers tekniska högskola / Institutionen för teknikens ekonomi och organisation
Series/Report no.: TEKX04-20-18
Collection:Examensarbeten för kandidatexamen // Bachelor Theses

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.