A flexible cost model for seaport-hinterland decisions in container shipping

Thiago de Almeida Rodrigues, Caroline Maria de Miranda Mota, Udechukwu Ojiako, Maxwell Chipulu, Alasdair Marshall, Fikri Dweiri

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)


A seaport–hinterland system offers a wide range of service options for shippers and consignees to export or import containerized cargo. The present study, set within the context of Brazil, seeks to propose a flexible cost model to aid joint seaport–hinterland network decisions by freight forwarders and clients. More specifically, employing simulation and sensibility analysis, the proposed model is employed to explore: (i) decisions on whether to perform logistics services in seaports or dry ports; (ii) decisions on whether to transport containerized cargo by road or by multimodal transportation (road and rail); and (iii) decisions on whether cargo is best delivered or received in either consolidated or deconsolidated form. Simulations are undertaken against seven transportation strategy permutations via which service operators might deliver containers from seaports or dry ports to clients. We find five factors—(i) cargo value, (ii) cargo accommodation type, (iii) transportation mode, (iv) delivery distance and (v) shipping process time—as critical to joint seaport–hinterland operations impacting shippers and consignees. Furthermore, dry ports are shown to be particularly cost efficient for imports requiring long-term storage, even where the dry ports are not directly connected to seaports or intermodal facilities.

Original languageEnglish
Article number101016
Number of pages44
JournalResearch in Transportation Business and Management
Early online date26 Jul 2023
Publication statusPublished - 31 Aug 2023


  • client's choice
  • container shipping
  • cost model
  • Seaport–hinterland network


Dive into the research topics of 'A flexible cost model for seaport-hinterland decisions in container shipping'. Together they form a unique fingerprint.

Cite this