Use of weather data in supply chain management

Decision making in the supply chain can be a difficult process. Forecasts are often used to predict demand since it is important to maintain the balance between keeping inventory to minimal levels and losing consumers due to inventory shortage. Weather inherently affects people and their behavior, and thus, weather influences product demand. With new technologies and increased knowledge, meteorologists have improved their ability to predict the weather, and it is reasonable to try to take advantage of this in decision making in the supply chain. A case study was conducted using data from the Icelandic retail company, N1. The data were tested for correlation between sales and weather. Then a multiple linear regression model was used to make predictions of sale. Estimates of the difference between actual sales and predicted sales were examined, as well as replacing weather variables with weather events. The process was carefully recorded and set forward as a proposed methodology for using multiple linear regression models for predicting sales of products and discovering which weather events affect demand. The results showed reasonably accurate forecasts, so it was realistic to conclude that it is possible to use weather data to predict demand, at least for products with weather driven demand. Being able to identify what kind of weather affects sales is important and provides the decision maker with greater knowledge of his/her products which leads to the ability to make better informed decisions regarding their products. Additional research is needed to test whether or not this method is more viable than the forecasting methods currently in use. Keywords: regression analysis, weather, forecasting, supply chain management, decision making, sales

Author dc:creator

• Elín Anna Gísladóttir 1988-

Contributors dc:contributor

• Háskólinn í Reykjavík

• Rekstrarverkfræði
• Aðhvarfsgreining
• Ákvarðanataka
• Veðurspár
• Tækni- og verkfræðideild
• Meistaraprófsritgerðir
• Engineering management
• Regression analysis
• Decision making
• Weather forecasting
• School of Science and Engineering

Language dc:language.iso

en

Handle dc:identifier.uri

http://hdl.handle.net/1946/22337

OAI identifier oai:identifier

oai:skemman.is:1946/22337