This paper evaluates five sensor network architectures for coastal marine pollution monitoring using a fuzzy multi-criteria decision-making framework. Thirty domain experts assessed the alternatives against five criteria (detection performance, spatial coverage, life-cycle cost, energy demand, and robustness) using linguistic ratings, which were converted into triangular fuzzy numbers. A fuzzy decision matrix and the corresponding criterion weights were derived and analysed using Fuzzy TOPSIS. Spatial coverage (w = 0.26) and detection performance (w = 0.24) emerged as the most influential criteria. The glider-satellite architecture achieved the highest closeness coefficient (CC = 0.76), followed by a combined HF radar-innovative mooring system (CC = 0.71); the IoT node grid, fixed buoy network, and vessel-based system obtained coefficients of 0.63, 0.58, and 0.49, respectively. Sensitivity tests across multiple weight scenarios and a robustness check confirm that the top-ranked option is stable under plausible shifts in decision priorities. Methodologically, this study does not propose a new mathematical decision algorithm. Instead, it contributes a rigorously structured decision-support workflow that integrates established techniques, structured expert elicitation with consistency screening, fuzzy aggregation, and multi-scenario weight-sensitivity into a single, replicable pipeline tailored to the selection of coastal pollution monitoring architectures. The methodological contribution, therefore, lies in the systematic integration, operationalisation, and application of these components, rather than in modifying the underlying Fuzzy TOPSIS formulation. For decision-makers, the study clarifies which criteria drive architecture choice, when lower-cost opportunistic or IoT-grid options become competitive, and how to assemble hybrid monitoring designs that combine wide-area coverage with reliable detection.
🌊
