Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model

Hadid Wahedi; Kacper Wrona; Mads Heltoft; Sarkaft Saleh; Thomas Roum Knudsen; Ulrik Bendixen; Izabela Nielsen; Subrata Saha; Gregers Sandager Borup

doi:10.1007/978-3-031-16407-1_1

Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model

Hadid Wahedi
, Kacper Wrona
, Mads Heltoft
, Sarkaft Saleh^*
, Thomas Roum Knudsen
, Ulrik Bendixen
, Izabela Nielsen
, Subrata Saha
, Gregers Sandager Borup

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

Abstract

Accurate demand forecasting is critical for any small and medium-sized manufacturer. Limited structured data sources commonly prevent small and medium-sized manufacturers from improving forecasting accuracy, affecting overall performance. We classified products, then implemented a hybrid forecasting method and compared the outcome with Exponential smoothing, ARIMA, LSTM, and ANN forecasting techniques. Numerical results demonstrate that a selection of forecasting methods is not independent of product categorization. For slow-moving products, careful consideration is required. The hybrid ARIMA-ANN method can outperform some existing techniques and lead to higher prediction accuracy, by capturing both linear and nonlinear variations.

Original language	English
Title of host publication	Advances in Production Management Systems. Smart Manufacturing and Logistics Systems : Turning Ideas into Action - IFIP WG 5.7 International Conference, APMS 2022, Proceedings Part I
Editors	Duck Young Kim, Gregor von Cieminski, David Romero
Number of pages	8
Publisher	Springer
Publication date	2022
Pages	3-10
ISBN (Print)	978-3-031-16406-4, 978-3-031-16409-5
ISBN (Electronic)	978-3-031-16407-1
DOIs	https://doi.org/10.1007/978-3-031-16407-1_1
Publication status	Published - 2022
Event	IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2022 - Gyeongju, Korea, Republic of Duration: 25 Sept 2022 → 29 Sept 2022

Conference

Conference	IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2022
Country/Territory	Korea, Republic of
City	Gyeongju
Period	25/09/2022 → 29/09/2022

Series	I F I P Advances in Information and Communication Technology
Volume	663 IFIP
ISSN	1868-4238

Bibliographical note

Publisher Copyright:
© 2022, IFIP International Federation for Information Processing.

Keywords

ARIMA-ANN
Forecasting
Machine learning

Access to Document

10.1007/978-3-031-16407-1_1

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Wahedi, H., Wrona, K., Heltoft, M., Saleh, S., Knudsen, T. R., Bendixen, U., Nielsen, I., Saha, S., & Borup, G. S. (2022). Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model. In D. Y. Kim, G. von Cieminski, & D. Romero (Eds.), Advances in Production Management Systems. Smart Manufacturing and Logistics Systems: Turning Ideas into Action - IFIP WG 5.7 International Conference, APMS 2022, Proceedings Part I (pp. 3-10). Springer. https://doi.org/10.1007/978-3-031-16407-1_1

Wahedi, Hadid ; Wrona, Kacper ; Heltoft, Mads et al. / Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model. Advances in Production Management Systems. Smart Manufacturing and Logistics Systems: Turning Ideas into Action - IFIP WG 5.7 International Conference, APMS 2022, Proceedings Part I. editor / Duck Young Kim ; Gregor von Cieminski ; David Romero. Springer, 2022. pp. 3-10 (I F I P Advances in Information and Communication Technology, Vol. 663 IFIP).

@inproceedings{e5b8dafb252c445b8c9684a0869ca93d,

title = "Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model",

abstract = "Accurate demand forecasting is critical for any small and medium-sized manufacturer. Limited structured data sources commonly prevent small and medium-sized manufacturers from improving forecasting accuracy, affecting overall performance. We classified products, then implemented a hybrid forecasting method and compared the outcome with Exponential smoothing, ARIMA, LSTM, and ANN forecasting techniques. Numerical results demonstrate that a selection of forecasting methods is not independent of product categorization. For slow-moving products, careful consideration is required. The hybrid ARIMA-ANN method can outperform some existing techniques and lead to higher prediction accuracy, by capturing both linear and nonlinear variations.",

keywords = "ARIMA-ANN, Forecasting, Machine learning",

author = "Hadid Wahedi and Kacper Wrona and Mads Heltoft and Sarkaft Saleh and Knudsen, \{Thomas Roum\} and Ulrik Bendixen and Izabela Nielsen and Subrata Saha and Borup, \{Gregers Sandager\}",

note = "Publisher Copyright: {\textcopyright} 2022, IFIP International Federation for Information Processing.; IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2022 ; Conference date: 25-09-2022 Through 29-09-2022",

year = "2022",

doi = "10.1007/978-3-031-16407-1\_1",

language = "English",

isbn = "978-3-031-16406-4",

series = "I F I P Advances in Information and Communication Technology",

publisher = "Springer",

pages = "3--10",

editor = "Kim, \{Duck Young\} and \{von Cieminski\}, Gregor and David Romero",

booktitle = "Advances in Production Management Systems. Smart Manufacturing and Logistics Systems",

address = "Germany",

}

Wahedi, H, Wrona, K, Heltoft, M, Saleh, S, Knudsen, TR, Bendixen, U, Nielsen, I, Saha, S & Borup, GS 2022, Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model. in DY Kim, G von Cieminski & D Romero (eds), Advances in Production Management Systems. Smart Manufacturing and Logistics Systems: Turning Ideas into Action - IFIP WG 5.7 International Conference, APMS 2022, Proceedings Part I. Springer, I F I P Advances in Information and Communication Technology, vol. 663 IFIP, pp. 3-10, IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2022, Gyeongju, Korea, Republic of, 25/09/2022. https://doi.org/10.1007/978-3-031-16407-1_1

Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model. / Wahedi, Hadid; Wrona, Kacper; Heltoft, Mads et al.
Advances in Production Management Systems. Smart Manufacturing and Logistics Systems: Turning Ideas into Action - IFIP WG 5.7 International Conference, APMS 2022, Proceedings Part I. ed. / Duck Young Kim; Gregor von Cieminski; David Romero. Springer, 2022. p. 3-10 (I F I P Advances in Information and Communication Technology, Vol. 663 IFIP).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model

AU - Wahedi, Hadid

AU - Wrona, Kacper

AU - Heltoft, Mads

AU - Saleh, Sarkaft

AU - Knudsen, Thomas Roum

AU - Bendixen, Ulrik

AU - Nielsen, Izabela

AU - Saha, Subrata

AU - Borup, Gregers Sandager

PY - 2022

Y1 - 2022

N2 - Accurate demand forecasting is critical for any small and medium-sized manufacturer. Limited structured data sources commonly prevent small and medium-sized manufacturers from improving forecasting accuracy, affecting overall performance. We classified products, then implemented a hybrid forecasting method and compared the outcome with Exponential smoothing, ARIMA, LSTM, and ANN forecasting techniques. Numerical results demonstrate that a selection of forecasting methods is not independent of product categorization. For slow-moving products, careful consideration is required. The hybrid ARIMA-ANN method can outperform some existing techniques and lead to higher prediction accuracy, by capturing both linear and nonlinear variations.

AB - Accurate demand forecasting is critical for any small and medium-sized manufacturer. Limited structured data sources commonly prevent small and medium-sized manufacturers from improving forecasting accuracy, affecting overall performance. We classified products, then implemented a hybrid forecasting method and compared the outcome with Exponential smoothing, ARIMA, LSTM, and ANN forecasting techniques. Numerical results demonstrate that a selection of forecasting methods is not independent of product categorization. For slow-moving products, careful consideration is required. The hybrid ARIMA-ANN method can outperform some existing techniques and lead to higher prediction accuracy, by capturing both linear and nonlinear variations.

KW - ARIMA-ANN

KW - Forecasting

KW - Machine learning

UR - https://www.scopus.com/pages/publications/85140454640

U2 - 10.1007/978-3-031-16407-1_1

DO - 10.1007/978-3-031-16407-1_1

M3 - Article in proceeding

AN - SCOPUS:85140454640

SN - 978-3-031-16406-4

SN - 978-3-031-16409-5

T3 - I F I P Advances in Information and Communication Technology

SP - 3

EP - 10

BT - Advances in Production Management Systems. Smart Manufacturing and Logistics Systems

A2 - Kim, Duck Young

A2 - von Cieminski, Gregor

A2 - Romero, David

PB - Springer

T2 - IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2022

Y2 - 25 September 2022 through 29 September 2022

ER -

Wahedi H, Wrona K, Heltoft M, Saleh S, Knudsen TR, Bendixen U et al. Improving Accuracy of Time Series Forecasting by Applying an ARIMA-ANN Hybrid Model. In Kim DY, von Cieminski G, Romero D, editors, Advances in Production Management Systems. Smart Manufacturing and Logistics Systems: Turning Ideas into Action - IFIP WG 5.7 International Conference, APMS 2022, Proceedings Part I. Springer. 2022. p. 3-10. (I F I P Advances in Information and Communication Technology, Vol. 663 IFIP). doi: 10.1007/978-3-031-16407-1_1