University of Twente Student Theses


Sales Forecasting: A case study in the Retail Business

Groot, Robin D. de (2019) Sales Forecasting: A case study in the Retail Business.

[img] PDF
Abstract:Sales forecasting has experienced much interest over the past few decades. The potential benefits businesses can gain from this are substantial, and this interest is not expected to decrease any time soon. This thesis looks at a case study in retail, specifically at A.S. Watson, the world’s largest health and beauty retailer. This company is interested in applying a sales forecasting algorithm with the implementation being shelf replenishment. For this task, predictions on the level of per product, per store, per day need to be made. Data and a computer to run tests on were provided, which carried their own limitations. Multiple machine learning algorithms were looked at, of which a Neural Network Classifier was tested in depth. A clear step-by-step process was followed for hyperparameter tuning. The results show impressive accuracy values, though other metrics show that accuracy is not always a valid performance metric. In order to arrive at a model that can provide better and more confident results, several limitations of the current thesis and recommendations for future research are provided.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:85 business administration, organizational science
Programme:Technology and Liberal Arts & Sciences BSc (50427)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page