Utilization of Predictive Analytics for Metsä Tissue’s Forecasting Process
Ruotsila, Emma (2022)
2022
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2022052712448
https://urn.fi/URN:NBN:fi:amk-2022052712448
Tiivistelmä
In this research-oriented bachelor's thesis, the results of predictive analytics in Metsä Tissue's forecasting process were examined from several perspectives. The aim of this study was to outline whether predictive analytics would be useful in the wider use of Metsä Tissue's forecasting process.
The thesis is divided into two parts, theoretical and empirical. The theoretical part focuses on the different aspects of the forecasting process and its diversity, as well as the components, steps, and insights of predictive analytics. The empirical part is divided into two phases, focusing on detailed process information obtained from internal documents, and expert opinions obtained from interviews.
The internal documents used in this study were covered in Phase 1. Such documents are currently used to improve Metsä Tissue's internal technological capabilities. Phase 2 was covered with expert interviews where four middle management representatives from Metsä Tissue gave insights from different perspectives. There were a total of 22 questions and 27 sub-questions, which were divided into four categories of investigative questions. Data on the interviews was collected both verbally and in writing. Excel was used as a tool to collect the interview data, from which it was then transferred to clear sentences for this study.
The results of the data indicated that predictive analytics can to some extent improve Metsä Tissue's forecasting process. The main areas remain in sales forecasting, where partial implementation has already taken place. The expansion of external use depends on the development of Metsä Group's planning process. Including the group view is important as the raw material flows are integrated and full end-to-end view is needed to create a balanced automated forecast. Restricting to small-scale development, such as automation, would therefore be more beneficial while waiting for a group-wide integrated planning solution.
The key findings and recommendations of this study provide suggestions on how Metsä Tissue should approach the weaknesses of its forecasting process to enable cost optimization and profitability.
The thesis is divided into two parts, theoretical and empirical. The theoretical part focuses on the different aspects of the forecasting process and its diversity, as well as the components, steps, and insights of predictive analytics. The empirical part is divided into two phases, focusing on detailed process information obtained from internal documents, and expert opinions obtained from interviews.
The internal documents used in this study were covered in Phase 1. Such documents are currently used to improve Metsä Tissue's internal technological capabilities. Phase 2 was covered with expert interviews where four middle management representatives from Metsä Tissue gave insights from different perspectives. There were a total of 22 questions and 27 sub-questions, which were divided into four categories of investigative questions. Data on the interviews was collected both verbally and in writing. Excel was used as a tool to collect the interview data, from which it was then transferred to clear sentences for this study.
The results of the data indicated that predictive analytics can to some extent improve Metsä Tissue's forecasting process. The main areas remain in sales forecasting, where partial implementation has already taken place. The expansion of external use depends on the development of Metsä Group's planning process. Including the group view is important as the raw material flows are integrated and full end-to-end view is needed to create a balanced automated forecast. Restricting to small-scale development, such as automation, would therefore be more beneficial while waiting for a group-wide integrated planning solution.
The key findings and recommendations of this study provide suggestions on how Metsä Tissue should approach the weaknesses of its forecasting process to enable cost optimization and profitability.