Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution Centers

<i>Background</i>: Transportation demand forecasting is an essential activity for logistics operators and carriers. It leverages business operation decisions, infrastructure, management, and resource planning activities. Since 2015, there has been an increase in the use of deep learning...

Full description

Saved in:
Bibliographic Details
Main Authors: Fábio Polola Mamede, Roberto Fray da Silva, Irineu de Brito Junior, Hugo Tsugunobu Yoshida Yoshizaki, Celso Mitsuo Hino, Carlos Eduardo Cugnasca
Format: Article
Language:English
Published: MDPI AG 2023-11-01
Series:Logistics
Subjects:
Online Access:https://www.mdpi.com/2305-6290/7/4/86
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849412004857511936
author Fábio Polola Mamede
Roberto Fray da Silva
Irineu de Brito Junior
Hugo Tsugunobu Yoshida Yoshizaki
Celso Mitsuo Hino
Carlos Eduardo Cugnasca
author_facet Fábio Polola Mamede
Roberto Fray da Silva
Irineu de Brito Junior
Hugo Tsugunobu Yoshida Yoshizaki
Celso Mitsuo Hino
Carlos Eduardo Cugnasca
author_sort Fábio Polola Mamede
collection DOAJ
description <i>Background</i>: Transportation demand forecasting is an essential activity for logistics operators and carriers. It leverages business operation decisions, infrastructure, management, and resource planning activities. Since 2015, there has been an increase in the use of deep learning models in this domain. However, there is a gap in works comparing traditional statistics and deep learning models for transportation demand forecasts. This work aimed to perform a case study of aggregated transportation demand forecasts in 54 distribution centers of a Brazilian carrier. <i>Methods</i>: A computational simulation and case study methods were applied, exploring the characteristics of the datasets through autoregressive integrated moving average (ARIMA) and its variations, in addition to a deep neural network, long short-term memory, known as LSTM. Eight scenarios were explored while considering different data preprocessing methods and evaluating how outliers, training and testing dataset splits during cross-validation, and the relevant hyperparameters of each model can affect the demand forecast. <i>Results</i>: The long short-term memory networks were observed to outperform the statistical methods in ninety-four percent of the dispatching units over the evaluated scenarios, while the autoregressive integrated moving average modeled the remaining five percent. <i>Conclusions</i>: This work found that forecasting transportation demands can address practical issues in supply chains, specially resource planning management.
format Article
id doaj-art-509873d4768c4f4598e36455ce93c835
institution Kabale University
issn 2305-6290
language English
publishDate 2023-11-01
publisher MDPI AG
record_format Article
series Logistics
spelling doaj-art-509873d4768c4f4598e36455ce93c8352025-08-20T03:34:35ZengMDPI AGLogistics2305-62902023-11-01748610.3390/logistics7040086Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution CentersFábio Polola Mamede0Roberto Fray da Silva1Irineu de Brito Junior2Hugo Tsugunobu Yoshida Yoshizaki3Celso Mitsuo Hino4Carlos Eduardo Cugnasca5Graduate Program in Logistics Systems Engineering, University of São Paulo, São Paulo 05508-010, BrazilInstitute of Advanced Studies, University of São Paulo, São Paulo 05508-010, BrazilGraduate Program in Logistics Systems Engineering, University of São Paulo, São Paulo 05508-010, BrazilGraduate Program in Logistics Systems Engineering, University of São Paulo, São Paulo 05508-010, BrazilDepartment of Production Engineering, University of São Paulo, São Paulo 05508-010, BrazilGraduate Program in Logistics Systems Engineering, University of São Paulo, São Paulo 05508-010, Brazil<i>Background</i>: Transportation demand forecasting is an essential activity for logistics operators and carriers. It leverages business operation decisions, infrastructure, management, and resource planning activities. Since 2015, there has been an increase in the use of deep learning models in this domain. However, there is a gap in works comparing traditional statistics and deep learning models for transportation demand forecasts. This work aimed to perform a case study of aggregated transportation demand forecasts in 54 distribution centers of a Brazilian carrier. <i>Methods</i>: A computational simulation and case study methods were applied, exploring the characteristics of the datasets through autoregressive integrated moving average (ARIMA) and its variations, in addition to a deep neural network, long short-term memory, known as LSTM. Eight scenarios were explored while considering different data preprocessing methods and evaluating how outliers, training and testing dataset splits during cross-validation, and the relevant hyperparameters of each model can affect the demand forecast. <i>Results</i>: The long short-term memory networks were observed to outperform the statistical methods in ninety-four percent of the dispatching units over the evaluated scenarios, while the autoregressive integrated moving average modeled the remaining five percent. <i>Conclusions</i>: This work found that forecasting transportation demands can address practical issues in supply chains, specially resource planning management.https://www.mdpi.com/2305-6290/7/4/86transportation demand forecastingsupply chain managementLSTMARIMAdata preprocessing
spellingShingle Fábio Polola Mamede
Roberto Fray da Silva
Irineu de Brito Junior
Hugo Tsugunobu Yoshida Yoshizaki
Celso Mitsuo Hino
Carlos Eduardo Cugnasca
Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution Centers
Logistics
transportation demand forecasting
supply chain management
LSTM
ARIMA
data preprocessing
title Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution Centers
title_full Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution Centers
title_fullStr Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution Centers
title_full_unstemmed Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution Centers
title_short Deep Learning and Statistical Models for Forecasting Transportation Demand: A Case Study of Multiple Distribution Centers
title_sort deep learning and statistical models for forecasting transportation demand a case study of multiple distribution centers
topic transportation demand forecasting
supply chain management
LSTM
ARIMA
data preprocessing
url https://www.mdpi.com/2305-6290/7/4/86
work_keys_str_mv AT fabiopololamamede deeplearningandstatisticalmodelsforforecastingtransportationdemandacasestudyofmultipledistributioncenters
AT robertofraydasilva deeplearningandstatisticalmodelsforforecastingtransportationdemandacasestudyofmultipledistributioncenters
AT irineudebritojunior deeplearningandstatisticalmodelsforforecastingtransportationdemandacasestudyofmultipledistributioncenters
AT hugotsugunobuyoshidayoshizaki deeplearningandstatisticalmodelsforforecastingtransportationdemandacasestudyofmultipledistributioncenters
AT celsomitsuohino deeplearningandstatisticalmodelsforforecastingtransportationdemandacasestudyofmultipledistributioncenters
AT carloseduardocugnasca deeplearningandstatisticalmodelsforforecastingtransportationdemandacasestudyofmultipledistributioncenters