Hyperautomation on fuzzy data dredging on four advanced industrial forecasting models to support sustainable business management

You-Shyang Chen, Arun Kumar Sangaiah () and Yu-Pei Lin
Additional contact information
You-Shyang Chen: National Chin-Yi University of Technology
Arun Kumar Sangaiah: National Yunlin University of Science and Technology
Yu-Pei Lin: Hwa Hsia University of Technology

Annals of Operations Research, 2024, vol. 342, issue 1, No 8, 215-264

Abstract: Abstract Recently, traditional manufacturing industries have faced two serious gaps and problems in line with effective product-line sales forecasting methods to balance the negative impacts on the performance of the subjective experience, including (1) arbitrary judgment, such as growth rate of expectancy, manager’s experiences, and historical sales data, may cause inaccurately predictive results and severe negative effects, and (2) sales forecasting is a key priority and challenge in the context of considerable product lines that have different properties and need specific models for supporting decision analytics. This study is motivated to propose an advanced hybrid model to utilize the advantages of variation for methods of fuzzy time series (FTS), exponential smoothing (ES), moving average (MA), and regression analysis (RA). To analyze the four product lines—stably growing product (SGP), declining product (DP), irregularly growing product (IGP), and special sales product (SSP)—this study is based on empirical sales data from a leading traditional manufacturer to accurately identify the high potentials of decisive key factors and objectively evaluate the model. Two evaluation standards—the mean absolute percentage error (MAPE) and root mean square error (RMSE), a parameter sensitivity analysis, and comparative analysis—are measured. After implementing the data from the case study, four key reports were conclusively identified. (1) Purely for the RMSE, the best one (10.32) is the ES method in the SGP line. (2) In the DP line, the better one is the RA(2) method, with a relatively low MAPE of 17.78% and RMSE of 26.46. (3) The FTS method is the best choice (MAPE 12.41% and RMSE 18.98) for the IGP line. (4) For the SSP line, the better one (MAPE 24.05 and RMSE 29.34) is the MA method. According to the above reports, although the proposed hybrid model has a general performance for the SSP line, it still has a superior predictor when compared to manager subjective prediction. Interestingly, the proposed model is rarely used, has a new trial with an innovative solution for the traditional manufacturer, and thus realizes applicable values. The study concludes with acceptable and satisfactory results and yields seven important findings and three managerial implications that significantly contribute to decision-making reference for complete sales-production planning for interested parties. Thus, this study benefits and values a conventional industry upgrade from novel application techniques.

Keywords: Decision-making practice; Sustainable business management; Decision support model; Sales forecasting model; Fuzzy time series (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10479-024-05882-0 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:annopr:v:342:y:2024:i:1:d:10.1007_s10479-024-05882-0

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10479

DOI: 10.1007/s10479-024-05882-0

Access Statistics for this article

Annals of Operations Research is currently edited by Endre Boros

More articles in Annals of Operations Research from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().