Forecasting worldwide empty container availability with machine learning techniques
Abstract Due to imbalances in the global transport of containerised goods, liner shipping companies go to great lengths to match the regional supply and demand for empty containers by transporting equipment from surplus to deficit regions. Making accurate forecasts of regional empty container availa...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2022-07-01
|
| Series: | Journal of Shipping and Trade |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s41072-022-00120-x |
| _version_ | 1818162708312227840 |
|---|---|
| author | Christoph Martius Lutz Kretschmann Miriam Zacharias Carlos Jahn Ole John |
| author_facet | Christoph Martius Lutz Kretschmann Miriam Zacharias Carlos Jahn Ole John |
| author_sort | Christoph Martius |
| collection | DOAJ |
| description | Abstract Due to imbalances in the global transport of containerised goods, liner shipping companies go to great lengths to match the regional supply and demand for empty containers by transporting equipment from surplus to deficit regions. Making accurate forecasts of regional empty container availability could support liner companies and other involved actors by making better relocation decisions, thus avoiding unnecessary transport costs of empty equipment. Previously proposed container availability prediction models are limited to the application in individual regions and typically characterized by a high degree of temporal aggregation. Against this background, this paper introduces two novel approaches based on machine learning and probabilistic techniques to predict the future weekly availability of empty containers for more than 280 locations worldwide. The machine learning and probabilistic prediction models are built by analysing a unique data set of more than 100 million events from past container journeys. These events represent different stages during the transport process of a container. Both models use a two-step forecast logic. First, the expected future location of a container is predicted. Second, the expected timestamp for arriving at that location is estimated. The machine learning model uses artificial neural networks and mixture density networks to forecast the movements of containers. The models are quantitatively assessed and compared to the actual availability of containers and two more conventional forecasting approaches. The results indicate that the probabilistic prediction approach can keep up with conventional approaches while the neural network approach significantly outperforms the other approaches concerning every evaluation metric. |
| first_indexed | 2024-12-11T16:37:57Z |
| format | Article |
| id | doaj.art-733412f579d844c1bc6ee4e19044e5e7 |
| institution | Directory Open Access Journal |
| issn | 2364-4575 |
| language | English |
| last_indexed | 2024-12-11T16:37:57Z |
| publishDate | 2022-07-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Journal of Shipping and Trade |
| spelling | doaj.art-733412f579d844c1bc6ee4e19044e5e72022-12-22T00:58:24ZengSpringerOpenJournal of Shipping and Trade2364-45752022-07-017112410.1186/s41072-022-00120-xForecasting worldwide empty container availability with machine learning techniquesChristoph Martius0Lutz Kretschmann1Miriam Zacharias2Carlos Jahn3Ole John4Fraunhofer Center for Maritime Logistics and Services CMLFraunhofer Center for Maritime Logistics and Services CMLFraunhofer Center for Maritime Logistics and Services CMLFraunhofer Center for Maritime Logistics and Services CMLFraunhofer Center for Maritime Logistics and Services CMLAbstract Due to imbalances in the global transport of containerised goods, liner shipping companies go to great lengths to match the regional supply and demand for empty containers by transporting equipment from surplus to deficit regions. Making accurate forecasts of regional empty container availability could support liner companies and other involved actors by making better relocation decisions, thus avoiding unnecessary transport costs of empty equipment. Previously proposed container availability prediction models are limited to the application in individual regions and typically characterized by a high degree of temporal aggregation. Against this background, this paper introduces two novel approaches based on machine learning and probabilistic techniques to predict the future weekly availability of empty containers for more than 280 locations worldwide. The machine learning and probabilistic prediction models are built by analysing a unique data set of more than 100 million events from past container journeys. These events represent different stages during the transport process of a container. Both models use a two-step forecast logic. First, the expected future location of a container is predicted. Second, the expected timestamp for arriving at that location is estimated. The machine learning model uses artificial neural networks and mixture density networks to forecast the movements of containers. The models are quantitatively assessed and compared to the actual availability of containers and two more conventional forecasting approaches. The results indicate that the probabilistic prediction approach can keep up with conventional approaches while the neural network approach significantly outperforms the other approaches concerning every evaluation metric.https://doi.org/10.1186/s41072-022-00120-xMaritime logistics forecastsEmpty container relocationMachine learningMixture density network |
| spellingShingle | Christoph Martius Lutz Kretschmann Miriam Zacharias Carlos Jahn Ole John Forecasting worldwide empty container availability with machine learning techniques Journal of Shipping and Trade Maritime logistics forecasts Empty container relocation Machine learning Mixture density network |
| title | Forecasting worldwide empty container availability with machine learning techniques |
| title_full | Forecasting worldwide empty container availability with machine learning techniques |
| title_fullStr | Forecasting worldwide empty container availability with machine learning techniques |
| title_full_unstemmed | Forecasting worldwide empty container availability with machine learning techniques |
| title_short | Forecasting worldwide empty container availability with machine learning techniques |
| title_sort | forecasting worldwide empty container availability with machine learning techniques |
| topic | Maritime logistics forecasts Empty container relocation Machine learning Mixture density network |
| url | https://doi.org/10.1186/s41072-022-00120-x |
| work_keys_str_mv | AT christophmartius forecastingworldwideemptycontaineravailabilitywithmachinelearningtechniques AT lutzkretschmann forecastingworldwideemptycontaineravailabilitywithmachinelearningtechniques AT miriamzacharias forecastingworldwideemptycontaineravailabilitywithmachinelearningtechniques AT carlosjahn forecastingworldwideemptycontaineravailabilitywithmachinelearningtechniques AT olejohn forecastingworldwideemptycontaineravailabilitywithmachinelearningtechniques |