MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model

<p dir="ltr">Accurate demand forecasting is vital for optimizing supply chain management and enhancing organizational resilience. Traditional forecasting methods, relying on simple arithmetic, often fail to capture complex patterns caused by seasonal variability and special events. A...

Full description

Saved in:
Bibliographic Details
Main Author: Md Abrar Jahin (20108252) (author)
Other Authors: Asef Shahriar (22504103) (author), Md Al Amin (20923805) (author)
Published: 2025
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1864513533570449408
author Md Abrar Jahin (20108252)
author2 Asef Shahriar (22504103)
Md Al Amin (20923805)
author2_role author
author
author_facet Md Abrar Jahin (20108252)
Asef Shahriar (22504103)
Md Al Amin (20923805)
author_role author
dc.creator.none.fl_str_mv Md Abrar Jahin (20108252)
Asef Shahriar (22504103)
Md Al Amin (20923805)
dc.date.none.fl_str_mv 2025-05-30T09:00:00Z
dc.identifier.none.fl_str_mv 10.1007/s12065-025-01053-7
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/MCDFN_supply_chain_demand_forecasting_via_an_explainable_multi-channel_data_fusion_network_model/30455849
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Commerce, management, tourism and services
Transportation, logistics and supply chains
Information and computing sciences
Artificial intelligence
Data management and data science
Supply chain demand forecasting
Explainable artificial intelligence
Deep learning
Convolutional neural networks
Long short-term memory
Gated recurrent units
dc.title.none.fl_str_mv MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Accurate demand forecasting is vital for optimizing supply chain management and enhancing organizational resilience. Traditional forecasting methods, relying on simple arithmetic, often fail to capture complex patterns caused by seasonal variability and special events. Although deep learning techniques have advanced, the lack of interpretable models hampers understanding and explaining predictions. We introduce the Multi-Channel Data Fusion Network (MCDFN), a novel hybrid deep learning architecture integrating multiple data modalities for superior demand forecasting. MCDFN utilizes Convolutional Neural Networks (CNNs), Long Short-Term Memory networks (LSTMs), and Gated Recurrent Units (GRUs) to extract spatial and temporal features from time series data. Comparative benchmarking against seven other deep-learning models validates MCDFN’s efficacy, showing it outperforms its counterparts across key metrics with a mean squared error (MSE) of 23.5738, root mean squared error (RMSE) of 4.8553, mean absolute error (MAE) of 3.9991, and mean absolute percentage error (MAPE) of 20.1575%. Theil’s U statistic of 0.1181 ( <i>U</i><1 <i>U</i> < 1 ) of MCDFN indicates its superiority over the naive forecasting approach, and a 10-fold cross-validated statistical paired t-test with a <i>p</i>-value of 5% indicated no significant difference between MCDFN’s predictions and actual values. To address the “black box” nature of MCDFN, we employ explainable AI techniques such as ShapTime and Permutation Feature Importance, offering insights into model decision-making processes. This research advances demand forecasting methodologies and provides practical guidelines for integrating MCDFN into existing supply chain systems.</p><h2>Other Information</h2><p dir="ltr">Published in: Evolutionary Intelligence<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1007/s12065-025-01053-7" target="_blank">https://dx.doi.org/10.1007/s12065-025-01053-7</a></p>
eu_rights_str_mv openAccess
id Manara2_0195e39952b5ac8f0edb9a2933e0be1c
identifier_str_mv 10.1007/s12065-025-01053-7
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/30455849
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network modelMd Abrar Jahin (20108252)Asef Shahriar (22504103)Md Al Amin (20923805)Commerce, management, tourism and servicesTransportation, logistics and supply chainsInformation and computing sciencesArtificial intelligenceData management and data scienceSupply chain demand forecastingExplainable artificial intelligenceDeep learningConvolutional neural networksLong short-term memoryGated recurrent units<p dir="ltr">Accurate demand forecasting is vital for optimizing supply chain management and enhancing organizational resilience. Traditional forecasting methods, relying on simple arithmetic, often fail to capture complex patterns caused by seasonal variability and special events. Although deep learning techniques have advanced, the lack of interpretable models hampers understanding and explaining predictions. We introduce the Multi-Channel Data Fusion Network (MCDFN), a novel hybrid deep learning architecture integrating multiple data modalities for superior demand forecasting. MCDFN utilizes Convolutional Neural Networks (CNNs), Long Short-Term Memory networks (LSTMs), and Gated Recurrent Units (GRUs) to extract spatial and temporal features from time series data. Comparative benchmarking against seven other deep-learning models validates MCDFN’s efficacy, showing it outperforms its counterparts across key metrics with a mean squared error (MSE) of 23.5738, root mean squared error (RMSE) of 4.8553, mean absolute error (MAE) of 3.9991, and mean absolute percentage error (MAPE) of 20.1575%. Theil’s U statistic of 0.1181 ( <i>U</i><1 <i>U</i> < 1 ) of MCDFN indicates its superiority over the naive forecasting approach, and a 10-fold cross-validated statistical paired t-test with a <i>p</i>-value of 5% indicated no significant difference between MCDFN’s predictions and actual values. To address the “black box” nature of MCDFN, we employ explainable AI techniques such as ShapTime and Permutation Feature Importance, offering insights into model decision-making processes. This research advances demand forecasting methodologies and provides practical guidelines for integrating MCDFN into existing supply chain systems.</p><h2>Other Information</h2><p dir="ltr">Published in: Evolutionary Intelligence<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1007/s12065-025-01053-7" target="_blank">https://dx.doi.org/10.1007/s12065-025-01053-7</a></p>2025-05-30T09:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1007/s12065-025-01053-7https://figshare.com/articles/journal_contribution/MCDFN_supply_chain_demand_forecasting_via_an_explainable_multi-channel_data_fusion_network_model/30455849CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/304558492025-05-30T09:00:00Z
spellingShingle MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
Md Abrar Jahin (20108252)
Commerce, management, tourism and services
Transportation, logistics and supply chains
Information and computing sciences
Artificial intelligence
Data management and data science
Supply chain demand forecasting
Explainable artificial intelligence
Deep learning
Convolutional neural networks
Long short-term memory
Gated recurrent units
status_str publishedVersion
title MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
title_full MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
title_fullStr MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
title_full_unstemmed MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
title_short MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
title_sort MCDFN: supply chain demand forecasting via an explainable multi-channel data fusion network model
topic Commerce, management, tourism and services
Transportation, logistics and supply chains
Information and computing sciences
Artificial intelligence
Data management and data science
Supply chain demand forecasting
Explainable artificial intelligence
Deep learning
Convolutional neural networks
Long short-term memory
Gated recurrent units