Tiered prediction models for port vessel emissions inventories

Abstract Albeit its importance, a large number of port authorities do not provide continuous or publicly available air emissions inventories (EIs) and thereby obscure the contribution ports. This is caused by, e.g., economic effort generated by obtaining data. Therefore, performance abatement measures monitored projected, which specifically disadvantageous concerning top contributors such as container ships. To mitigate this issue, in paper we propose vessel EI prediction models exploring combination different machine-learning algorithms, data from one-off application an activity-based bottom-up methodology vessel-characteristics The results for specific case show that enable acceptable trade-offs between requirements, promoting creation EIs.