The maritime industry is under increasing pressure to reduce its greenhouse gas (GHG) emissions. European ports, including the Port of Rotterdam, are transitioning to sustainable solutions to reduce their environmental impact. Since the Port of Rotterdam is large, patrol and incident-response (RPA) vessels are needed to ensure a safe port. However, the existing fleet is reaching its operational end of life. These vessels currently rely on conventional propulsion systems, which are not in line with the sustainability goals of the Port of Rotterdam. The Port of Rotterdam performed research with several companies to see what type of renewable energy solutions could be applied to the new RPA vessels. Given the 24/7 operational profile and energy demands of (relatively small) RPA vessels, a battery or hydrogen propulsion system installed onboard the vessels was not an option. Therefore, another possibility was proposed: the (rapid) swapping of energy modules to and from the vessel at swapping stations.

This research addresses the challenge of designing an optimal configuration for such a swapping method based on the operational sailing profiles of RPA vessels. A systematic methodology was applied, starting with a literature review for the technical and operational issues. The findings highlighted the feasibility of implementing a Battery Energy Storage System (BESS) powered by lithium-ion batteries, supported by decentralised swapping and charging stations structure. Different concepts for the Battery Swapping Method (BSM) were considered from which the Shfitr concept was the most promising application for this thesis. The Shiftr concept uses cranes to automatically swap energy modules to and from vessels. Different optimisation methods were investigated to see which one best supports finding an optimal configuration (energy modules, swapping/charging stations and vehicle related). It was concluded that a Mixed Integer Linear Programming method could be best applied, with the objective to minimise downtime of the swapping process for different configurations.

A mathematical model for the optimisation model was developed to minimise downtime during swapping, incorporating historical AIS data to simulate the vessels’ operational profiles. This optimisation model can track both the vessel r and the modules m for each time t. This mathematical model is made in Gurobi (Python) and can, for different configurations, minimise the total swapping time for a given period of data. The configurations are used as input data, including the number of energy modules and the number/placement of swapping and charging stations based on the energy consumption of vessels and their sailing path defined during the pre-processing. Verification and validation confirmed the optimisation model was working, although computational challenges were observed when validating each individual vessel.

After verification and validation, two case studies were performed, which included multiple vessels. Due to the rapid increase in computational time, only two case studies were performed with two and three vessels, respectively. Except for the number of vessels, the configuration from the pre-processing was the same for both the case studies. For both cases, the number of swaps was in line with the expectations of each case and confirmed the working of the optimisation model. As the vessels only lost minutes on swapping while sailing for hours before the modules were depleted, a low swapping frequency was observed in the optimisation model. At the same time, there are too many swapping stations in the configuration, leading to a lot of unused energy modules.

In conclusion, the optimisation model can successfully optimise the sailing path to minimise swapping time, for a specific configuration. However, the research question of finding the optimal configuration could not be answered fully. Therefore, further research is needed to address computational limitations and to optimise module utilisation. This thesis provides a foundation for swapping renewable energy sources in modules, for relatively small vessels, which are operational 24/7.

Offshore Floating Wind Energy (OFWE) presents a promising yet underdeveloped sector due to its higher costs and engineering complexities compared to fixed-bottom systems. Unlike fixed-bottom systems, OFWE isn't restricted by water depth, offering over four times more ocean area for deployment. This flexibility allows for positioning wind farms further offshore where wind speeds are higher and more consistent, maximizing energy generation potential. However, the current viability of OFWE heavily depends on governmental subsidies to remain competitive.
Integrated Systems (IS) have emerged as a promising solution to address the need for affordable, scalable, and transportable renewable energy storage. IS combines offshore wind farms with hydrogen production systems, aiming to enhance the Techno-Economic Performance (TEP) of OFWE and potentially render it economically feasible.
A case study focusing on Japan's Goto City Wind Farm illustrates the challenges and opportunities associated with OFWE and IS integration. Japan, with its ambitious wind energy goals and limited suitable areas for fixed-bottom wind energy, serves as an ideal location for such a study. Additionally, Japan's interest in integrating hydrogen into its energy mix further underscores the relevance of this research.
The initial analysis of the Goto City Wind Farm without IS integration revealed its lack of economic viability over its operational lifetime. However, by integrating additional components for hydrogen production, such as desalination, electrolysis, and hydrogen carrier configuration units, the project's Techno-Economic Performance was significantly enhanced.
One key advantage of the IS approach lies in its flexibility in power allocation between grid supply and hydrogen production. By dynamically adjusting power allocation based on prevailing market conditions and a predetermined switch price, the IS can optimize revenue generation and adapt to fluctuating energy markets.
Further analyses, including examinations of different hydrogen carrier configurations, capacity assessments, scenario analyses, and sensitivity analyses, were conducted to comprehensively evaluate the TEP of the system.
Results indicated that hydrogen production during periods of low power prices substantially improved the TEP of the Goto City Wind Farm, making it economically feasible by the end of its operational lifespan. Moreover, the ability to switch between hydrogen production and grid supply mitigated risks associated with future uncertainties in hydrogen and power prices.
In conclusion, integrating hydrogen production with OFWE through Integrated Systems offers a promising pathway to enhance TEP and improve the economic viability of offshore wind energy projects, contributing to a more sustainable energy future.

With the energy transition taking up speed and strong decarbonisation ambitions offshore wind is becoming a major source of green electricity. European countries are among the leading drivers of the offshore wind expansion on both the wind turbine as well as the vessels side. It is expected that until 2030 170 GW of capacity can be installed, equalling 16000 wind turbines. The wind turbines are serviced using either small vessels or commissioning and service operation vessels (C/SOVs) when parks are larger and in more challenging conditions further away from shore. The C/SOV market is still in development and it is not known how many of these vessels are needed to serve the European offshore wind industry in 2030. It is further unknown until now which factors influence the need for these vessels as both market dynamics as well as operations have not been researched until now.
This research first defines the quantifiable factors influencing the need for C/SOVs in offshore wind parks. These are the park parameters such as the distance to shore and the number of turbines in the park. These data are used in a factor model and Monte Carlo simulation to make an assumption on the required number of vessels. The results are then compared against qualitative factors influencing the need for C/SOVs indirectly.
Out of a high and a low case, the low case was shown to be the most likely fit for the research results. It showed that to serve the offshore wind market in 2030 between 122 and 138 vessels are needed, which is 12 to 28 more than currently are active or on order.
Considering the fact that the industry needs to adapt to a new market, it is crucial to know which factors drive that market and how they influence it. This project allows for researchers to dive further into these factors and research them in more detail. Further the research can assist industry players in their investment decisions and yards can accordingly plan capacity.

With the current focus on climate change, the shipping industry is forced to decarbonise its fleet. The need for decarbonisation is not only fuelled by the aspirations of the sector but also by laws and regulations from countries and organisations. As these laws are gradually tightening and there is not one clear technical or operational solution, there is a need for a strategy to combine multiple operational and technical solutions. These solutions, as well as other aspects, are associated with risks and uncertainties, such as fuel prices, technical aspects and cost. To help the decision makers in this field, this research aims to present a method that can be used as the basis for a decision making tool. The method will mainly focus on finding the lowest total cost per tonne mile but will also consider other aspects that are found to be necessary, such as safety and technological readiness. The difficulty in the decision is the number of factors that could influence the decarbonisation of the vessel and how these aspects influence each other. The addition of CRSs (Carbon Reduction Systems) will affect the operation profile of the vessel and, with that, the fuel use of the vessel. With fuel being one of the main expenses of ships, fuel use will affect the total cost of the vessel. The method will combine several different analyses to consider all aspects and risks. The strategy will be created by adding CRSs every time the vessel does not comply with the regulations anymore, affecting the vessel's operational profile and fuel use. The cost of fuel use, and other OPEX, will be evaluated with an NPC (Net Present Cost) analysis to compare costs in time. These costs will then be entered into an MCA (Multi Criteria Analysis) and combined with the other criteria, such as safety and technological readiness level, leading to a strategy selection. To increase the method's robustness, the whole set of calculations will also be subjected to a Monte Carlo simulation to mitigate the risks that concern future inputs. The method was translated into a model where a case study was carried out to validate the method and to find some early insights. From this came the result that the method creates valid answers and that there is one method that is the most cost-effective manner to comply with the regulations. This is plainly sailing slower; it has almost no expense and significantly reduces the cost of shipping.

The work focuses on evaluating the feasibility of a Sale and Purchase online platform. Drivers and barriers to implement it have been identified performing a qualitative analysis, involving people directly employed in the industry. Later, a concept design for the platform has been analysed, overcoming the barriers and using the drivers as strength points. Finally, an economic evaluation has been performed

Container ships are becoming larger and larger, but also more numerous. As a result, these large ships are pushing their smaller counterparts down the line, until these smaller ships are no longer efficient for trade routes. The focus area of this thesis is on the North Sea and Baltic Sea area, in particular the area between the West coast of Norway to the Russian part of the Gulf of Finland, with most countries in between with feeder activities between them and North European gateway hubs. This area is seeing an increase in port development. As a result, routes that container ships take are being changed. Either larger ships are coming into the region and stopping at more ports, or they are utilised as feeder ships in the region. This thesis aims to investigate the cascading effect of container ships, regarding the choice of feeder ships between German North Sea ports and feeder destinations in Scandinavian countries and countries on the Baltic Sea. To achieve this, an optimisation model is created that calculates differing total costs for container ships with capacities ranging between 200 and 5000 Twenty-foot Equivalent Units (TEU) in the region. This is done by assigning ships to container flows from German North Sea ports to feeder ports in the region, and allowing the model to choose which arcs are used to fulfil the container inflow of the destination ports. To analyse the impact of each ship type regarding the total cost of their routes taken, ship types are tested separately. Experiments have been run for increasing bunker costs based on existing data, for increasing amounts of container inflows, and decreasing the number of ports based on an increase of minimum container inflow.
The results of the experiments varying bunker costs show the total cost to fulfil all container inflows for each feeder ship type is highly dependent on its utilisation of the Kiel Canal. The Kiel Canal restricts the choice of ships with capacities larger than 1250 TEU. The extra time spent for sailing around the Northern tip of Denmark has a negative effect on the largest ships chosen for this thesis. For an increasing bunker cost, a 1500 TEU capacity ship is hindered more negatively than a 5000 TEU ship due to the Kiel Canal. For variations in volume of container flows, the same impact of the Kiel Canal is also seen. However, total costs start to favour the largest ships chosen, with the largest capacity ship (5000 TEU) having the lowest total costs calculated for a 300\% of the original chosen container flows. For decreasing the number of ports from the set based on their container inflow, the largest capacity ships have the lowest total costs from a minimum container inflow of around 400 TEU per week. From this, with current flows and within the near future, it is estimated that the current fleet of around 1000 - 1250 TEU capacity container ships will maintain their position within the North Sea and Baltic Sea region. However, for routes with a substantial container flow volume, larger capacity ships are estimated to be more cost competitive than their smaller counterparts.
This conclusion is based on experiments with only one ship size available for all container flows, with the assumption that all chosen ships are not restricted by berth sizes in ports. Furthermore, all flows originate from German North Sea ports. Recommendations for future research should take all chosen ship types into consideration for arc-flow pairing, and have more origins for container flows. If all ship types are taken into consideration, size restrictions of ports and transshipment between ship types should also be implemented. Origins included should include other major hubs in the region, as well as from outside the region, as an incoming flow. This would allow for an investigation into large container ships sailing between the current chosen origin hub and large ports, and the consequence on the region as a whole.

Port performance is under pressure due to high congestion levels and a need to reduce GHG emissions. The industry is looking for ways to increase efficiency in ports. A promising concept is known as port call optimization where waiting times and turn-around times can be reduced. This research contributes to the body of port call optimization literature by identifying the importance of performance attributes for innovations. Identifying the performance attributes and how stakeholders weight these has never been explored before in literature. The relative importance has been derived using the Bayesian Best-Worst Method. The performance attributes: incidents, near misses, robustness and added value, are most important for evaluating an innovation for port call management. Results on micro level show heterogeneity and homogeneity in preferences for port attributes. Both ports with complex nautical restrictions prefer performance attributes related to safety. The port with centralised management is more commercially driven and there is homogeneity within a port. Furthermore, heterogeneity is seen between stakeholder groups and between port authority (PA). PAs from centralised port value attributes relatively different than a PA with decentralised and licensing activities. Some differences and similarities can be explained by the port contexts. The findings can be used for the design of innovations in port call management. Research could be repeated in other port cases to validate the significance of these results.

Container liner shipping organisations facilitate trade between all regions globally by deploying vessels in a network with a corresponding sailing schedule. Since the container shipping network is part of a larger network, it is favourable for the continuation of the supply chain to maintain this schedule. Although the importance of the continuation is recognised, the industry is not known for a high degree of schedule reliability. This is due to the presence of regular and irregular uncertainties in the network. It is found that more than 90% of these uncertainties find their origin in the port area. By deploying the LSS methodology, the causes and effects of port uncertainties are further analysed. It is found that the port uncertainties lead to transportation waste at sea in the form of fuel costs and emissions. The shipping sector is held accountable for approximately 3% of the global carbon dioxide emissions. A Dynamic Arrival Time (DAT) design is proposed to recover a schedule after a delay takes place to avoid waiting times and arrive JIT in the port. This differentiates from the current state where a First Come First Serve policy is maintained. An Agent-Based Discrete Event Simulation model in Anylogic is proposed to test the effects of a DAT implementation in a Short Sea Shipping (SSS) network that is subject to regular and irregular port uncertainties. It is found that a fuel and emissions reduction of up to 6.1% is achieved when a DAT is implemented under irregular uncertainties in the network. If the DAT incorporation is tested for regular uncertainties, the effect is equal to a 0.5% reduction in fuel and emissions. Furthermore, it is found that the design leads to a decrease in the port turnaround time due to improved arrival predictability. This improved predictability enables the shipping line (or agent) to recover the schedule and avoid drastic fuel increase during the departure transit. This study's findings are important for the policy-making for shipping companies due to the substantial energy savings that are measured. Furthermore, this study contributes to the MEPC.323(74) resolution, where voluntary cooperation between the port and shipping sectors is encouraged to reduce the GHG emissions from the shipping industry.

To mitigate global warming, worldwide CO2-emissions have to be cut. Similarly, the shipping industry has to drastically decrease it's emissions. Rules and regulation on emissions are becoming more stringent. Moreover, the public and clients are demanding less pollution from ship operators. While many shipping operators are looking at a suited fuel alternative, no consensus has been reached on the best alternative to entirely dispose of harmful emissions from shipping in the long-term. Fugro is one of the ship operators interested in entirely bringing down the emissions of their vessels that are built from 2030 onwards and therefore this research is conducted. This thesis will research how and if Fugro can achieve the IMO target of 70 percent CO2 reduction in 2050, by deploying net zero-emission fuels in Fugro vessels as of 2030. An evaluation will be done to assess which alternative fuels are feasible and most suitable, based on current technologies and considering future scenarios around technology. The research will approach this problem by firstly conducting an extensive literature research into alternative fuels and Multiple Criteria Decision Analysis methods. Thereafter the Fugro fleet, potential future vessel developments, future scenarios and alternative fuels are researched to set up a technical framework for this case problem specifically. With this technical framework in place, the method itself is conducted. This is done in two ways. First, by conducting a qualitative approach using the Analytic Hierarchy Process, also involving stakeholders from Fugro and the industry. Second, by carrying out a case application of the four best alternatives on a representative Fugro vessel. This approach, including a qualitative approach extended with a quantitative one results in a substantiated advise. Both methods complement each other on points where the other falls short. The AHP includes stakeholders and takes into account criteria that are difficult to quantify. The application approach tests the feasibility of the proposed alternatives and tests the potentially subjective or intuitive outcome of the AHP method. With methanol scoring well in both approaches, the conclusion of this research is that methanol is the advised alternative fuel for future Fugro vessels. Liquid hydrogen is the highest rated fuel alternative from the AHP. The application example shows that this fuel isn't able to comply with the required Fugro operability however. Contrary to that, ammonia is a highly rated alternative in literature and is also scoring well in the application part of this thesis, but is not rated as a suitable alternative by stakeholders in the AHP. The last considered alternative in both parts of the method is synthetic diesel, which is comparable to MDO. This fuel scores well in both methods but is considered to remain too expensive to become a viable alternative in the coming ten years. In this thesis, estimates of different fuel options are presented and some design concepts and considerations on the different alternative fuel applications are given. The next step for Fugro would be to work out a methanol concept, by going through the design spiral more than once and work out the design choices and challenges this fuel brings. With alternative fuels being more expensive than MDO, this thesis pointed out that the economic speed will change when using alternative fuels, which is another point that requires more research. While the AHP proved to be a suitable way to score fuel alternatives while including stakeholders and assessing criteria qualitatively, it also pointed out that it is difficult to solely base an alternative fuel advise on this method. Potentially, the AHP results would improve when the questionnaire is simplified by reducing the amount of considered criteria.

This study aims to develop a decision support tool which enables shipowners to select the most appropriate alternative fuel technology to comply with possible different imposed emission regulations while ensuring optimal business performance. In this context, most appropriate is defined as a fuel alternative which minimises required freight rate (RFR) while maximising overall performance on technological, environmental and other criteria. A decision support tool was devised combining a decision model based on the simple multi-attribute rating technique (SMART) with a financial model based on discounted cash-flow (DCF). Additionally, an optimisation model was implemented to optimise for minimal required freight rate through slow steaming. The decision tool provides shipowners with a quantified impact on their current business if they do not transition to alternative fuels under a 'market based measure' (MBM) regulatory scenario, as well as best-alternatives if their current fuels do not meet regulations under an 'emission cap' (EC) scenario. The decision tool is evaluated under optimistic, average and pessimistic scenarios in 2020, 2030 and 2050. Under an emission cap scenario, the results showed an overall preference for Fischer-Tropsch diesel (FTD) as the most promising alternative maritime fuel both in terms of SMART performance and required freight rate, followed by upgraded bio-oil (UBO). Nevertheless, the average difference in required freight rate of alternative fuels compared to HFO remains substantial, at 43% and 38% higher for Fischer-Tropsch diesel under a 'no regulation' scenario and 'market based measures' scenario, respectively. It is therefore evident that without regulatory intervention, heavy fuel oil is expected to retain dominance based on cost. Even under a market based measure (MBM) scenario, the average required freight rate of HFO increases only by 3.4% overall. For LNG, market-based measures lead to an average increase in RFR of 4.1%. These results suggest that in order for the maritime industry to transition towards sustainable alternative fuels, policymakers, governments, international organisations and lenders need to align their policies to collectively enable a more sustainable shipping industry - not only by enforcing stricter regulations, but also by providing the correct financial incentives.

The IMO has proposed Just-In-Time arrivals and services as one of the short-term solutions to reduce greenhouse gas emissions in maritime shipping. The Just-In-Time concept allows vessels to optimize their speed during the voyage. If a vessel earlier knows at which time it is requested to arrive at the pilot boarding place, the vessel can adapt its speed to arrive when the berth, nautical services and the fairway are available. Besides environmental reasons, shipping companies as MSC consider the Just-In-Time initiative as a major port call process optimization. It has obvious advantages for MSC since this concept aims to minimise unnecessary ad hoc waiting times and reduces fuel consumption per mile steamed. In the past, it has become clear it is extremely complicated to implement this initiative. Research is required in order to find out what needs to be improved to enable Just-In-Time arrivals and services. The current port call processes must be analysed. As start point, a case study is conducted on the port call processes of the world’s second largest shipping company in the biggest port of Europe. In this study, the current port call processes of MSC container shipping in the Port of Rotterdam are identified and improved in order to enable Just-In-Time arrivals and services. First, an actor-stakeholder analysis is conducted to understand the incentives and relations between actors involved in a port call. Subsequently, this information is used in the process analysis. In this study, the port call processes are analysed by both qualitative and quantitative research. The research has emphasized the complexity of the port call business processes. Actors primarily act on the basis of their own incentives. In order to enable Just-In-Time arrivals and services, collaboration is required among the involved actors. Today, most actors see the benefits of this concept. However, some actors are not willing to share data about their processes since this is considered as sensitive information. In case actors are willing to share data, it can be concluded that the information is not exchanged frequently enough or does often not meet any data standards. As last, this study shows that adaptations are required to the current business process of a port call in the Port of Rotterdam in order to enable Just-In-Time arrivals and services.

Jumbo Maritime is a shipping company active in the heavy lifting market. Jumbo uses stabilisers to enlarge the stability of the ship during lifting. By using stabilisers Jumbo can perform heavy lifts with a relative small ship, this is one of the major advantages. Downside of using stabilisers is that the installation and de-installation is a time-consuming process. Besides that the usage also causes safety issues, more and more port authorities no longer approve the usage of stabilisers. Research is done in finding alternatives for the use of stabilisers, focusing on alternatives for new ships that need to be build. Looking at Jumbo’s position in the market shows the following unique selling points, for the J-type: • Limited draft compared to its competitors • Length < 150 meters On the one hand the usage of smart stabiliser to speed up the installation and de-installation of stabilisers is looked at. Because smart stabilisers are still stabilisers, also alternatives are analysed that do not require a stabiliser at all. By widening the ship, the ship can be stable enough to perform lifts without extra support from stabilisers. A model is created to be able to determine the required width for each concept. This model is generated based on the input and requirements set by literature, market analysis and Jumbo’s specifications. Some of these requirements are: limited dimensions, minimum required stability, anti heeling and deadweight. The four concepts that are created as input for the model are: • Base concept, concept 1..., Concept 2... Concept 3...Concept 4...: A case study is performed to be able to compare the concepts. Three cases are created, which are actual relevant cases for JumboMaritime: For each concept the costs are determined to sail the certain case. Because the heavy lift market consists of somany single jobs that differ a lot on weight, complexity and distance it is difficult to map the revenues. The costs are determined per job, to be able to compare the different concepts. Costs consists of the capital costs, fuel costs and operational related costs. Besides cost per job, also cost per ton/mile is calculated to determine the economical speed. For case 1 the economical speed is 14 knots, which is similar to the required design speed. This means that the ships sails most cost efficient at this speed. The time that is saved by eliminating the stabiliser is used to reduce the sailing speed. In this way the same amount of jobs can done in the same time. This research shows that the saved fuel consumption as a result of slower sailing can compensate the longer sailing time and increased resistance for a wider hull concept. Comparing the concepts in the different cases shows that concepts 3 and 4 are more beneficial in case the stabiliser usages goes up. It can be concluded that awider hull shape can operate at 2-5%lower costs. The fuel consumption of the V-shape decreases more significantly at lower drafts. It is an advantage at ballast sailing or during sailing light cargoes. Besides the lower costs, the increase of deck space for the wider concepts is added value in the heavy lift market. An other important advantage of the wider ship concepts is the elimination of safety issues because the usage of stabilisers is not needed any more to lift heavy cargoes safely.

Maritime transport covers about 90%of world trade and is seen as one of the most cost-effective way of transporting goods. In order to safely accommodate vessels calling into ports, tugs are operated to assist various sized vessels with navigating, manoeuvring and berthing. Competition within the ship handling market has always been fierce. Boskalis Towage, through a network of joint ventures, offers harbour- and terminal towage services around the world, comprised of about 450 vessels. Operating a large fleet comes large Operating Expenses (OPEX). In order to deliver a more cost-effective service, Boskalis turns to business intelligence (BI) to help with the decision-making process. Maintenance is crucial for the operation of the vessels, but also a large part of the expenses. Large deviations in maintenance costs and operational performance exist between the joint ventures. However, the reasons behind these deviations are yet unclear due to the absence of continuous monitoring capabilities. The objective is therefore to increase the management capabilities of detailed performance monitoring of M&R expenses and performance of tugs, through which possible improvements can be determined. The M&R of tugs is influenced by various factors, such as utilisation, operational location and installed equipment. This research has evaluated the current Enterprise Performance Management (EPM) of Boskalis Towage and used literature on Maintenance Performance Management (MPM) in order to erect a multi-criteria hierarchical function-specific framework which is a framework for maintenance performance monitoring. The presented framework covers aspects of alignment with business objectives and plan, strategy and implementation. Moreover, the framework presents the maintenance function and its areas of performance measuring of equipment-, cost- and process performance across the organisational hierarchy. By performing a regression analysis of the maintenance data, this has also resulted in an increased understanding of the relationship between maintenance costs, operational data and vessel characteristics. As a result, an equation with acceptable statistical accuracy for running repairs has been found. The implementing of the MPM framework on the maintenance of tugs has resulted in the MPM model. This model focuses on the running repairs and dry docking direct costs and activities, incorporating basic maintenance types, i.e. Preventive Maintenance (PM) and Corrective Maintenance (CM). Running repairs have been evaluated in the time domain while the evaluation of dry dockings have been evaluated as events due to their relationship with regulatory surveys. The model incorporates operational-, maintenance- and financial data, tug specifications and operational conditions in order to evaluate maintenance performance. Maintenance Performance Indicators (MPIs) have been established to determine deviations and deficiencies among the different aspects of M&R. Challenges were faced on the aspects of data quality and data management, especially in the areas of maintenance process. The model has been validated through a detailed evaluation of five selected tugs. The underlying reasons for the underperformance of these tugs have been found and verified with knowledgable bodies and thus validating the model and its ability to determine underperformance and deviations. An alternative performance analysis method, Data Envelopment Analysis (DEA), has been researched and shows promise. The Slack-Based Measure (SBM) DEA model is a non-parametric frontier approach, based on Linear Programming (LP), to evaluate performance based on slacks (room for improvement) of Decision Making Units (DMUs). The method is capable of determining similarities in inefficiency in maintenance performance, comparing it to results from the MPM model. It, therefore, proves the method is capable of quick evaluation of maintenance performance and of determining new maintenance targets. However, the model is simplistic and requires further detailing and is unable to determine underlying reasons for underperformance. The regression equation and DEA have been used to formulate a so-called performance region which describes the lower- and upper bounds of the running repair costs for individual tugs. Unfortunately, data quality doesn’t allow for the determining of these bounds for all vessels nor for dry docking. New key benchmarking targets have been established through quartile values for respective Maintenance Performance Indicators (MPIs). With this newly obtained knowledge, future maintenance performance evaluation of tugs can now be performed with a higher detail through analysis conducted with the developed MPM model.

In this thesis, it was examined how high-frequency operational data collected from an automated recording system, can be used to develop a vessel’s performance model. Both the traditional and the Machine Learning (ML) approach were examined. In the traditional approach, every component of the propulsion plant was examined separately, while empirical or semi-empirical methods were used for the added resistance due to waves, wind and shallow water. Historical data were used for verification and tuning (where needed) during the development of the model. In the Machine Learning approach, the method of Support Vector Machines was used. Also, it was examined whether a vessel’s performance model can be developed by combining the methods mentioned above. Thus, in total four models were developed. The data regarding the last year of operation of the vessel Schippersgracht were used (45000 data points) to test the models. It was found that with all the four models, 80 % of the examined dataset was calculated with an error of less than 10 %. The daily fuel consumption was calculated for the same dataset and for 90% of the days examined, the daily fuel consumption was calculated with an error of less than 10 %. The main conclusions about the developed performance models is that all the models resulted in more less the same accuracy. The model based on the traditional approach has the advantage that it offers more information and more outputs can be taken, compared to the ML approach which only provides the output of the fuel consumption. On the other hand, the model based on the ML approach is significantly faster and allows many iterations to be performed in a short time. Regarding the models which were developed by combining the approaches mentioned above, they did not result in additional advantages. Finally, two case studies were performed. In the first case study, it was presented how the model, developed following the traditional approach, can be used to examine how modifications in the propulsion system will influence the vessel’s fuel consumption. The case of switching from fixed rpm (that is currently used) to a variable rpm system was examined, which is a real scenario that the company is considering. It was calculated that installing the variable rpm system and a frequency converter will result in annual savings of around 250,000 € while the investment’s payback time is estimated to be around two years. In the second case study, a future voyage simulation algorithm was developed. This algorithm receives as inputs the voyage plan(route and speed) and performs a calculation for the voyage’s fuel consumption by taking into account the weather predictions. The voyage simulation algorithm is making use of the ML model. A few cases were examined in order to examine how many days in advance the calculation is reliable (due to the uncertainty of the weather predictions) and about the time interval that should be used in the algorithm. However, clear conclusions were not derived, and further examination is suggested(by examining more voyages).