TY  - CHAP
U1  - Konferenzveröffentlichung
A1  - Kiefer, Daniel
A1  - Grimm, Florian
A1  - Bauer, Markus
A1  - van Dinther, Clemens
T1  - Demand forecasting intermittent and lumpy time series: comparing statistical, machine learning and deep learning methods
T2  - Proceedings of the 54th Hawaii International Conference on System Sciences (HICSS-54), 4-8 January 2021, online
N2  - Forecasting intermittent and lumpy demand is challenging. Demand occurs only sporadically and, when it does, it can vary considerably. Forecast errors are costly, resulting in obsolescent stock or unmet demand. Methods from statistics, machine learning and deep learning have been used to predict such demand patterns. Traditional accuracy metrics are often employed to evaluate the forecasts, however these come with major drawbacks such as not taking horizontal and vertical shifts over the forecasting horizon into account, or indeed stock-keeping or opportunity costs. This results in a disadvantageous selection of methods in the context of intermittent and lumpy demand forecasts. In our study, we compare methods from statistics, machine learning and deep learning by applying a novel metric called Stock-keeping-oriented Prediction Error Costs (SPEC), which overcomes the drawbacks associated with traditional metrics. Taking the SPEC metric into account, the Croston algorithm achieves the best result, just ahead of a Long Short-Term Memory Neural Network.
KW  - Intelligent Decision Support for Logistics and Supply Chain Management
KW  - demand forecasting
KW  - intermittent
KW  - lumpy
KW  - spec
KW  - deep learning
Y1  - 2021
UN  - https://nbn-resolving.org/urn:nbn:de:bsz:rt2-opus4-34969
SN  - 978-0-9981331-4-0
SB  - 978-0-9981331-4-0
U6  - https://doi.org/10.24251/HICSS.2021.172
DO  - https://doi.org/10.24251/HICSS.2021.172
SP  - 1425
EP  - 1434
S1  - 10
PB  - University of Hawai'i at Manoa
CY  - Honolulu
ER  -