Small Island Developing States (SIDS) face vulnerability to supply chain disruptions due to their geographical isolation, infrastructural limitations, and heavy dependence on imported goods. Despite advancements in logistics and resilience modeling, existing frameworks are typic
...
Small Island Developing States (SIDS) face vulnerability to supply chain disruptions due to their geographical isolation, infrastructural limitations, and heavy dependence on imported goods. Despite advancements in logistics and resilience modeling, existing frameworks are typically designed for larger, more connected systems and fail to account for the unique logistical constraints of SIDS. This thesis addresses that gap by developing a, discrete event model grounded in a four phase freight framework. The model simulates the flow of goods of like food, medicine, and fuel under conditions of uncertainty, using logic tailored to island infrastructure and behavior.
The model builds upon the framework proposed by the four step transport modeling approach, generation, distribution, mode choice, and assignment, but is adapted for SIDS through simplifications and extensions suited to low data contexts. Cold chain prioritization, fragmented demand generation, and congestion sensitive dispatching are all explicitly modeled to hold true to SIDS specific operationalization. a discrete base event queue manages system operations, enabling simulation of time based disruptions and network delays. By introducing perishability, transport constraints, and batch scheduling, the model balances simplicity with the complexity required to represent real world island conditions.
Insights from fieldwork in the Seychelles and expert conversations inform key behavioral parameters, such as the informal handling of cold goods, the limited separation of goods in transport, and the absence of formal distribution networks. bring empirical insight into storage constraints, inter island transport scheduling, and vehicle distribution rules to simulate realistic disruptions and recovery behavior.
With "service level" as performance indicator for the model an Monte Carlo simulation is done to bring insight in to the working of the model. which showcase that Storm duration and timing a larger effect have on performance the operational variables in the model. These correlations are further investigated during different disruptions showcasing showcasing that in the current model setup the build up of disruption have a larger effect then longer sustaining disruptions on service level. Different behavior for different islands groups (Main/Inner/Outer) are identified and showcase that the network wide approach is key for SIDS.
Different adaptations strategies are tested that showed promise for resilience building in the constraint environment that are SIDS. Through limitation in the model/approach not a conclusive answer was found however recommendations are made that a balancing of adaptations strategies is key in order to create network wide resilience in SIDS.