Dutch ports face significant environmental challenges due to high greenhouse gas and air pollutant emissions from berthed ships engaged in essential operations. Entities like the Port of Rotterdam and research institutes produced emission inventories to gain insight into ports' environmental impact. 

However, existing emission inventories rely on fixed factors and overlook berthing duration and activities: they assume the same emission factors regardless of the operations performed by vessels. Thus hindering effective emission reduction strategies and infrastructure development. 

Consequently, the following research studies areas of improvement for the current emission calculation methods by including berthing time patterns. Then, it will apply this information to the Port of Rotterdam to test the new methodology.

The principal result that emerged from this work is that reality is far more complex than the most widely used models to represent it; luckily, the available data can still grasp this complexity. First, not all berthing events are equal, and the rarest ones still hold a considerable impact on the total berthing time. Additionally, the differences do not exist only as a matter of frequency and duration but also imply different energy requirements, as suggested by the interviewed parties. Consequently, a model based on a standard stop only reliably captures part of the behaviour of moored ships. Second, there is no universal relationship between standard industry size classes and time spent in port. Some ship types show a positive linear behaviour (the larger the ship, the longer the visit berthing time), while others show more complex relationships. Moreover, some fleet types belong to only a few classes; therefore, no size-related difference can be observed in the behaviour of berthed vessels. These considerations apply to metrics of volume (GT), maximum transportable weight (DWT), and also cargo capacity (TEU). Consequently, the size classes commonly used in the industry - which reflect the physical limits of channels, ports, and other waterways - might be insufficient to portray the in-port behaviour of all the fleet types. Third, most ships berth at a single location during a port visit, proving that the duration of a port visit could be the only way to identify different types of berthing events. Finally, even the interviews suggested that the system is much more complex than captured by the currently used models. The division by fleet type might be an oversimplification as the product transported and the ship type considerably influence the behaviour in port and energy consumption.

These results contrast with the emission calculation methodologies commonly used in literature and policy documents, indicating a widespread risk of policy and investment failure because partial data and oversimplified models of reality might have driven policy and economic decisions. Thus, institutions may have enacted regulations with potentially inefficient economic, environmental, and social returns. Consequently, improving the system's knowledge is essential for informed decision-making and risk minimization. This research has shown that this objective is not only reachable but could be achievable without exponentially increasing the required data.

In April 2018, the International Maritime Organization (IMO) agreed upon the target to reduce the CO₂ emissions from international shipping in 2050 by 50 percent with respect to 2008 levels. The strategy to reach this target consists of the ambition to reduce CO₂ emissions per transport work, as an average across international shipping, by at least 40% by 2030, compared to 2008 levels. However, due to uncertainties involved in the decision making on the implementation of CO₂ reducing measures, shipowners are uncertain which actions to take regarding their existing fleet in order to deal with those upcoming requirements.
For rail transport, road transport and inland waterway transport, TNO has developed the Multi-level Energy Optimization tool (MEO). For these modalities, MEO is capable of calculating the CO₂ emissions for a certain voyage and assessing the effect of CO₂ reducing measures on the carbon footprint of those voyages. The objective of the research presented in this report is to extend TNO's MEO tool with the maritime modality and develop a tool supporting shipowners in the decision making process regarding the pathway of implementing CO₂ reducing measures in sea-going vessels.
The tool presented in this report is developed to meet this objective and is based on a Markov Decision Processes (MDP) model. The MDP method has the ability of optimizing a time dependent decision policy in the face of uncertainty. In this case, the optimal policy resulting from the MDP model is optimized to obtain the highest cost savings by implementing CO₂ reducing measures. For the input of the model, the vessel has to be described by an operational profile and the vessel and engine specifics such as a speed-load curve and the propulsive efficiency. Also, the environment has to be described by fuel properties, economical parameters and a regulatory framework. With this input, the MDP method can assess the optimal policy regarding the implementation of a set of CO₂ reducing measures with the objective of minimizing costs. The output of the tool consists of the expected values of total costs, CO₂ emissions and CO₂ efficiency for when the optimal policy is followed, in comparison to maintain the business as usual.

To put this tool in practice, a case study has been carried out on a 93000 DWT bulk carrier of which operational data has been provided by the shipowner. A data analysis on the daily reports of the vessel has been done to obtain the values for the vessel input and literature has been used for the environment input. The optimal policy for a set of measures is assessed and the influence on the results by varying fuel prices, the regulatory framework and the dry dock schedule has been studied. The key result of the case study is that with the implementation of several fuel saving measures and a dual fuel engine the total costs with respect to maintaining the business as usual is expected to be reduced by 12%. Also, the total accumulated carbon emission reduction over 15 years is expected to be 29% and the CO₂ efficiency is improved by 45% without having to reduce the speed of the vessel.