STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer Goods

Accurately forecasting sales for fast-moving consumer goods (FMCG) remains a significant challenge due to the volatile and multi-faceted nature of sales data. Existing methods often struggle to capture intricate patterns driven by seasonal trends, external factors, and consumer behavior, hindering e...

Full description

Saved in:
Bibliographic Details
Main Authors: Yecheng Ma, Lili He, Junhong Zheng
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/3/1516
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850199314540789760
author Yecheng Ma
Lili He
Junhong Zheng
author_facet Yecheng Ma
Lili He
Junhong Zheng
author_sort Yecheng Ma
collection DOAJ
description Accurately forecasting sales for fast-moving consumer goods (FMCG) remains a significant challenge due to the volatile and multi-faceted nature of sales data. Existing methods often struggle to capture intricate patterns driven by seasonal trends, external factors, and consumer behavior, hindering effective inventory management and strategic decision-making. To overcome these challenges, we propose STL-DCSInformer-ETS, a hybrid model that integrates three complementary components: STL decomposition, an enhanced DCSInformer model, and the ETS model. The model uses monthly sales data from a FMCG company, with key features including sales volume, product prices, promotional activities, and regulatory factors such as holidays, geographical information, consumer behavior, product factors, etc. STL decomposition partitions time-series data into trend, seasonal, and residual components, reducing data complexity and enabling more targeted forecasting. The enhanced DCSInformer employs dilated causal convolution and a multi-scale feature extraction mechanism to capture long-term dependencies and short-term variations effectively. Meanwhile, the ETS model specializes in modeling seasonal patterns, further refining forecasting precision. To further improve predictive performance, the Random Forest-based Recursive Feature Elimination (RF-RFE) method is applied to optimize feature selection. RF-RFE identifies key predictive factors from multiple dimensions, such as time, geography, and economy, which significantly influence forecasting accuracy. Through numerical experiments, the method demonstrates excellent performance by achieving a 35.9% reduction in Mean Squared Error and a 21.4% decrease in Mean Absolute Percentage Error, significantly outperforming traditional methods. Furthermore, the model effectively captures both medium- and long-term sales trends while addressing short-term fluctuations, leading to more accurate forecasting and improved decision-making for fast-moving consumer goods. This research provides new theoretical insights into hybrid forecasting models and practical solutions for optimizing inventory management and strategic planning in the FMCG industry.
format Article
id doaj-art-69e60861696f420f8e7a2d298189ef3d
institution OA Journals
issn 2076-3417
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-69e60861696f420f8e7a2d298189ef3d2025-08-20T02:12:38ZengMDPI AGApplied Sciences2076-34172025-02-01153151610.3390/app15031516STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer GoodsYecheng Ma0Lili He1Junhong Zheng2College of Computer Science and Technology (College of Artificial Intelligence), Zhejiang Sci-Tech University, Hangzhou 310018, ChinaCollege of Computer Science and Technology (College of Artificial Intelligence), Zhejiang Sci-Tech University, Hangzhou 310018, ChinaCollege of Computer Science and Technology (College of Artificial Intelligence), Zhejiang Sci-Tech University, Hangzhou 310018, ChinaAccurately forecasting sales for fast-moving consumer goods (FMCG) remains a significant challenge due to the volatile and multi-faceted nature of sales data. Existing methods often struggle to capture intricate patterns driven by seasonal trends, external factors, and consumer behavior, hindering effective inventory management and strategic decision-making. To overcome these challenges, we propose STL-DCSInformer-ETS, a hybrid model that integrates three complementary components: STL decomposition, an enhanced DCSInformer model, and the ETS model. The model uses monthly sales data from a FMCG company, with key features including sales volume, product prices, promotional activities, and regulatory factors such as holidays, geographical information, consumer behavior, product factors, etc. STL decomposition partitions time-series data into trend, seasonal, and residual components, reducing data complexity and enabling more targeted forecasting. The enhanced DCSInformer employs dilated causal convolution and a multi-scale feature extraction mechanism to capture long-term dependencies and short-term variations effectively. Meanwhile, the ETS model specializes in modeling seasonal patterns, further refining forecasting precision. To further improve predictive performance, the Random Forest-based Recursive Feature Elimination (RF-RFE) method is applied to optimize feature selection. RF-RFE identifies key predictive factors from multiple dimensions, such as time, geography, and economy, which significantly influence forecasting accuracy. Through numerical experiments, the method demonstrates excellent performance by achieving a 35.9% reduction in Mean Squared Error and a 21.4% decrease in Mean Absolute Percentage Error, significantly outperforming traditional methods. Furthermore, the model effectively captures both medium- and long-term sales trends while addressing short-term fluctuations, leading to more accurate forecasting and improved decision-making for fast-moving consumer goods. This research provides new theoretical insights into hybrid forecasting models and practical solutions for optimizing inventory management and strategic planning in the FMCG industry.https://www.mdpi.com/2076-3417/15/3/1516STL-DCSInformer-ETSmedium- and long-term sales forecastingfast-moving consumer goodsrandom forest recursive feature eliminationmulti-scale feature extractiondilated causal convolution
spellingShingle Yecheng Ma
Lili He
Junhong Zheng
STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer Goods
Applied Sciences
STL-DCSInformer-ETS
medium- and long-term sales forecasting
fast-moving consumer goods
random forest recursive feature elimination
multi-scale feature extraction
dilated causal convolution
title STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer Goods
title_full STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer Goods
title_fullStr STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer Goods
title_full_unstemmed STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer Goods
title_short STL-DCSInformer-ETS: A Hybrid Model for Medium- and Long-Term Sales Forecasting of Fast-Moving Consumer Goods
title_sort stl dcsinformer ets a hybrid model for medium and long term sales forecasting of fast moving consumer goods
topic STL-DCSInformer-ETS
medium- and long-term sales forecasting
fast-moving consumer goods
random forest recursive feature elimination
multi-scale feature extraction
dilated causal convolution
url https://www.mdpi.com/2076-3417/15/3/1516
work_keys_str_mv AT yechengma stldcsinformeretsahybridmodelformediumandlongtermsalesforecastingoffastmovingconsumergoods
AT lilihe stldcsinformeretsahybridmodelformediumandlongtermsalesforecastingoffastmovingconsumergoods
AT junhongzheng stldcsinformeretsahybridmodelformediumandlongtermsalesforecastingoffastmovingconsumergoods