Title: Humans and algorithms : creation and measurement of economic value in demand forecasting
Authors : Peter, Kauf
Ott, Thomas
Conference details: The 3rd Swiss Conference on Data Science (SDS|2016), Winterthur, 16. September 2016
Issue Date: 2016
License (according to publishing contract) : Licence according to publishing contract
Type of review: Not specified
Language : English
Subjects : Data science
Subject (DDC) : 004: Computer science
Abstract: Successful demand planning relies on accurate demand forecasts. Existing demand planning software typically employs (univariate) time series models for forecasting. These methods work well if the demand of a product follows regular patterns. Their power and accuracy are however limited if the patterns are disturbed and the demand is driven by irregular external factors such as promotions, events or specific weather conditions. PrognosiX AG is a start-up company that provides user focused software solutions taking into account external drivers for improved forecasting. The scientific basis has been laid by our research consortium. We developed a novel high-performance forecasting methodology that combines various forecasting approaches with situation-dependent strengths. Yet, to substantiate the impact of this methodology, we were left with the question how to measure and compare the performance or accuracy of different forecasting methods. Standard measures such as root mean square error (RMSE) and mean absolute percentage error (MAPE) may allow for ranking the methods according to their accuracy, but in many cases these measures are difficult to interpret or the rankings are incoherent among different measures. Moreover, the impact of forecasting inaccuracies is usually not reflected by standard measures. In our contribution, we will discuss this issue and define alternative measures that provide intuitive guidance for decision makers and users of demand forecasting. We will describe how such measures, together with forecasting technology, can be applied in realistic cases of big and small data - such that forecasts can be turned into real value.
Departement: Life Sciences und Facility Management
Organisational Unit: Institute of Applied Simulation (IAS)
Publication type: Conference Other
URI: http://www.zhaw.ch/storage/hochschule/institute-zentren/datalab/SDS/2016/Slides/kauf.pdf
Appears in Collections:Publikationen Life Sciences und Facility Management

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.