Modelling Indonesia's National Sea Freight Transport

The main aim of this research is to develop and apply a national freight transport model for Indonesia focusing on the sea freight transport network. The model will be built employing the Aggregate-Disaggregate-Aggregate (ADA) model framework, but the network assignment is performed only for the sea freight network. Indonesia’s sea freight network is chosen as the main focus of this research due to its crucial roles as the backbone of logistics transport for this archipelagic country and due to the policy debate involving two ministries, the Ministry of Transport and the Ministry of Development Planning, with regard to the best sea freight route configuration in Indonesia.
The ADA model framework consists of five steps. Those are PC flows generation (PC flows means the freight flows between production (P) and consumption (C) locations), conversion to firm-to-firm flows (freight flows between suppliers and receivers in the PC locations), a logistics model explaining the transport chain and shipment size decisions of the shippers at the disaggregated level, accumulating disaggregate flows with the same mode into OD flows (freight flows between the origin (O) and destination (D) location classified for each mode type), and network assignment. Therefore, several methodological approaches are needed to build the model, involving aggregate econometric modelling, discrete choice modelling, and transport network analysis.
One particular objective of this research is to implement the ADA logistics model to Indonesia. The logistics model applied in the official ADA model for Sweden (Samgods), and Norway (NEMO), is functioning in a deterministic way. This method assumes that shippers choose the transport chain and shipment size which have a minimum total logistics cost. An optimisation model is employed. The deterministic method is not based on the actual observations, unlike the stochastic method. The stochastic approach in the ADA applies the random utility model to estimate the probability that a particular transport chain and shipment size will be used by shippers. This stochastic method is expected to improve the model's reliability in forecasting the impacts of policy measures.
The key decision within the stochastic logistics model for the transport actors' (in this research defined as manufacturers and wholesalers, i.e. shippers) relates to the transport chain choice for their domestic shipment. A revealed preference (RP) and stated preference (SP) survey is conducted as an approach to collect that particular data. Discrete choice modelling based on the joint RP/SP data is used to improve the initial logistics model that followed the deterministic approach.
The main contribution of this research lies in developing Indonesia's national sea freight transport model, and this model will be the first application of the ADA model for a developing country. By developing this model, this research also contributes to collect data regarding the shipper choices on the transport chain and shipment size for their interregional shipment. In addition, this model is applied to test different sea freight network configurations and determine which one is best, i.e., resulting in the minimum total cost, for Indonesia.