Deep-learning Model Using Hybrid Adaptive Trend Estimated Series for Modelling and Forecasting Sales

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216275%3A25410%2F24%3A39922243" target="_blank" >RIV/00216275:25410/24:39922243 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/article/10.1007/s10479-022-04838-6" target="_blank" >https://link.springer.com/article/10.1007/s10479-022-04838-6</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/s10479-022-04838-6" target="_blank" >10.1007/s10479-022-04838-6</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Deep-learning Model Using Hybrid Adaptive Trend Estimated Series for Modelling and Forecasting Sales

Popis výsledku v původním jazyce

Existing sales forecasting models are not comprehensive and flexible enough to consider dynamic changes and nonlinearities in sales time-series at the store and product levels. To capture different big data characteristics in sales forecasting data, such as seasonal and trend variations, this study develops a hybrid model combining adaptive trend estimated series (ATES) with a deep neural network model. ATES is first used to model seasonal effects and incorporate holiday, weekend, and marketing effects on sales. The deep neural network model is then proposed to model residuals by capturing complex high-level spatiotemporal features from the data. The proposed hybrid model is equipped with a feature-extraction component that automatically detects the patterns and trends in time-series, which makes the forecasting model robust against noise and time-series length. To validate the proposed hybrid model, a large volume of sales data is processed with a three-dimensional data model to effectively support business decisions at the product-specific store level. To demonstrate the effectiveness of the proposed model, a comparative analysis is performed with several state-of-the-art sales forecasting methods. Here, we show that the proposed hybrid model outperforms existing models for forecasting horizons ranging from one to 12 months.

Název v anglickém jazyce

Deep-learning Model Using Hybrid Adaptive Trend Estimated Series for Modelling and Forecasting Sales

Popis výsledku anglicky

Existing sales forecasting models are not comprehensive and flexible enough to consider dynamic changes and nonlinearities in sales time-series at the store and product levels. To capture different big data characteristics in sales forecasting data, such as seasonal and trend variations, this study develops a hybrid model combining adaptive trend estimated series (ATES) with a deep neural network model. ATES is first used to model seasonal effects and incorporate holiday, weekend, and marketing effects on sales. The deep neural network model is then proposed to model residuals by capturing complex high-level spatiotemporal features from the data. The proposed hybrid model is equipped with a feature-extraction component that automatically detects the patterns and trends in time-series, which makes the forecasting model robust against noise and time-series length. To validate the proposed hybrid model, a large volume of sales data is processed with a three-dimensional data model to effectively support business decisions at the product-specific store level. To demonstrate the effectiveness of the proposed model, a comparative analysis is performed with several state-of-the-art sales forecasting methods. Here, we show that the proposed hybrid model outperforms existing models for forecasting horizons ranging from one to 12 months.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

50204 - Business and management

Projekt

<a href="/cs/project/GA19-15498S" target="_blank" >GA19-15498S: Modelování emocí ve verbální a neverbální manažerské komunikaci pro predikci podnikových finančních rizik</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Annals of Operations Research

ISSN

0254-5330

e-ISSN

1572-9338

Svazek periodika

339

Číslo periodika v rámci svazku

1-2

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

32

Strana od-do

297-328

Kód UT WoS článku

000819683700001

EID výsledku v databázi Scopus

2-s2.0-85133272456