University of Twente Student Theses

Login

Forecasting promotional demand volume

Kerkdijk, J.S. (2019) Forecasting promotional demand volume.

[img] PDF
3MB
Abstract:This research for the master thesis is performed at Unilever Netherlands in Rotterdam. Unilever is a global company selling Fast Moving Consumer Goods (FMCGs). Characteristics of FMCG market are competitive, high volumes, fast turn over, multiple product innovations. In the FMCG market the effect of promotions on sales volume is substantial. Achieving the highest product availability at the lowest costs in combination with high volumes and high inventory turnover is a big challenge. Because of this, Unilever strives to deliver the right amount of products at the right time to the right customer at the lowest cost, which is only possible with an accurate forecast of the demand volume. This emphasises the importance of an accurate forecast. That is why Unilever wants to shift towards automating demand volume forecasts of promotions using predictive modelling. Unilever has co-developed a predictive modelling tool, based on a machine learning algorithm. In the implementation phase the company encountered resistance from the employees responsible for the demand forecast, because the current predictive model does not perform well at their perspective at detailed item level. As consequence, these employees modify data, resulting in an incorrect use of the workflow, leading to lower performance on forecast accuracy in the operation. Low accuracy leads to high stock costs or low fill rate do to underforecasting or overforecasting. This results in the volume forecast for product promotions not being accepted by those employees and therefore not used, which is the main problem in this research. This research is part of continious improvement programm within the company. The goal of this research is to answer the main research question: “How to improve the current forecasting method for product promotions to ensure it is accepted and internalized by the users?”
Item Type:Essay (Master)
Clients:
Unilever, Rotterdam, Netherlands
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:31 mathematics
Programme:Industrial Engineering and Management MSc (60029)
Link to this item:https://purl.utwente.nl/essays/79738
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page