Increasing competition in the container shipping sector has meant that terminals are having to equip themselves with increasingly accurate analytical and governance tools. A transhipment terminal is an extremely complex system in terms of both organisation and management. Added to the uncertainty surrounding ships’ arrival time in port and the costs resulting from over-underestimation of resources is the large number of constraints and variables involved in port activities. Predicting ships delays in advance means that the relative demand for each shift can be determined with greater accuracy, and the basic resources then allocated to satisfy that demand. To this end, in this article we propose two algorithms: a dynamic learning predictive algorithm based on neural networks and an optimisation algorithm for resource allocation. The use of these two algorithms permits on the one hand to reduce the uncertainty interval surrounding ships’ arrival in port, ensuring that human resources can be planned around just two shifts. On the other hand, operators can be optimally allocated for the entire workday, taking into account actual demand and operations of the terminal. Moreover, as these algorithms are based on general variables they can be applied to any transhipment terminal. Future integration of the two models within a broader decision support system will provide an important support tool for planners for fast, flexible planning of the terminal’s operations management.

Prediction of arrival times and human resources allocation for container terminal

FANCELLO, GIANFRANCO;PANI, CLAUDIA;SERRA, PATRIZIA;ZUDDAS, PAOLA;FADDA, PAOLO
2011-01-01

Abstract

Increasing competition in the container shipping sector has meant that terminals are having to equip themselves with increasingly accurate analytical and governance tools. A transhipment terminal is an extremely complex system in terms of both organisation and management. Added to the uncertainty surrounding ships’ arrival time in port and the costs resulting from over-underestimation of resources is the large number of constraints and variables involved in port activities. Predicting ships delays in advance means that the relative demand for each shift can be determined with greater accuracy, and the basic resources then allocated to satisfy that demand. To this end, in this article we propose two algorithms: a dynamic learning predictive algorithm based on neural networks and an optimisation algorithm for resource allocation. The use of these two algorithms permits on the one hand to reduce the uncertainty interval surrounding ships’ arrival in port, ensuring that human resources can be planned around just two shifts. On the other hand, operators can be optimally allocated for the entire workday, taking into account actual demand and operations of the terminal. Moreover, as these algorithms are based on general variables they can be applied to any transhipment terminal. Future integration of the two models within a broader decision support system will provide an important support tool for planners for fast, flexible planning of the terminal’s operations management.
File in questo prodotto:
File Dimensione Formato  
MEL_mel20113.pdf

Solo gestori archivio

Dimensione 297.86 kB
Formato Adobe PDF
297.86 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11584/108306
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 63
  • ???jsp.display-item.citation.isi??? 50
social impact