Automation and Sales Forecasting Patterns. A time series analysis
Keywords
No Thumbnail Available
Authors
Issue Date
2018-07-12
Language
en
Document type
Journal Title
Journal ISSN
Volume Title
Publisher
Title
ISSN
Volume
Issue
Startpage
Endpage
DOI
Abstract
This master thesis analysed the impact of implementing an automated statistical short-medium term sales forecast. The analysis was therefore partially focussed upon how to automate the statistical sales forecasting and partially on the impact it has on decision making.
Statistically, this thesis enhanced the monthly sales forecast accuracy within the case study with 7,8% - 9,8%, by both enabling statistical updating and eliminating the former ‘one-size-fits-all’ forecast methodology.
The main value of this thesis, however, resides in the analytical generalisation. By analysing the conditional factors in terms of integration, approach, systems and performance measurements (and its related fallacies), this thesis identified not solely requirements necessary for statistical improvement, buy also possibilities to enhance utilization, regarding trust and implementation. Within the unit of analysis, final improved decision making included enhanced planning (utilization of production machines) and enhanced procurement (stock and expediting).
Furthermore, one of the main messages this thesis offers regarding decision making is the importance of triangulation. Having multiple sources, systems, and methodologies, potentially offers great value. This indicates that having solely an automated univariate sales forecast system will not improve decision making to its fullest potential. There is value in variety.
Description
Citation
Supervisor
Faculty
Faculteit der Managementwetenschappen