1,720,980 research outputs found
Control and Coordination for Automated Container Terminals
No full textFor enhancing the performance of automated container terminals, this PhD thesis focuses on improving energy efficiency and implementing more autonomous equipment (e.g., free-ranging AGVs) at the operational level. On the one hand, due to the increased energy price and environmental stress, energy efficiency needs to be improved. On the other hand, new emerging AGVs allow free-ranging behavior and can shorten the driving distance than using the traditional routing strategy, demanding a novel advanced control algorithm for scheduling and controlling the free-ranging AGVs and the other related machines. For achieving these research goals, both discrete-event dynamics and continuous-time dynamics are considered in this thesis, using a perspective of hybrid systems. Simulation experiments on compact, medium and large-scale terminal case studies show the potential of the proposed new approaches.Maritime and Transportation TechnologyMechanical, Maritime and Materials Engineerin
Het ontwerp van intelligente software voor energiebezuiniging met behulp van patroonherkenning in het elektriciteitsverbruik
No full textThis thesis presents software that controls electrical household appliances. It uses an intelligent algorithm that adapts to the use of these household appliances by recognizing patterns of electricity useage.Electrical Engineering, Mathematics and Computer Scienc
The impact of the implementation of category loading at container terminals: A simulation study into the implementation of a less compelling loading concept in which containers are loaded to deep-sea vessels at maritime container terminals
No full textThis thesis presents and discusses a loading concept to improve the efficiency of loading containers to deep-sea vessels at maritime container terminals. The efficiency includes the overall service time to handle a deep-sea vessel at the quay of a container terminal. Nowadays, the terminals are threatened by overcapacity due to a decreasing growth of container throughput and the vessels becoming larger. Shipping companies benefit from shorter times spend in the harbour and therefore put pressure on the process of handling the vessels. The container terminals have to invest heavily to meet the demands of services requests of the shipping lines. Researching literature about improving the loading process, it turned out that the bottleneck of the service time at container terminals can be found in the loading of the vessels. The loading of containers to the deep-sea vessels is more complex than discharging the containers from the deep-sea vessel. This due to the applied loading sequence that has to be retained during the loading, which is necessary for a couple of reasons: the weight distribution of the containers over the vessel and the locations of special containers, for example container that contain dangerous goods or have to be connected to the power for climate control, have to be placed carefully. An optimal solution for the loading process is not found yet. As a consequence of the complexity and the number of different variables that have to be optimized in this one process. Different situations as loading an airplane or truck and the handlings in warehouses offer partly solutions that could be used to the loading process of a deep-sea vessel. These solutions are helpful in a way of identifying important influences, like the reachability of goods in warehouses and the different moments of arrivals of goods which can cause problems during the loading sequence of the various transport means. Research in the literature focussing on the loading process at maritime container terminals shows that especially the output of the stacking yard crane, that facilitates the yard, delays the loading process. A promising concept of loading that is mentioned in literature classifies containers in categories. An improvement of 5% in performance of the utilization of the quay crane during loading, compared to a situation without categories is mentioned. The loading concept that is researched is a less compelling way of loading compared to the current situation and according to further research improvements of the loading process are shown. However, the research that is done uses assumptions on the loaded containers and only one way of classifying the containers is tested. Furthermore, no attention is payed to the changes in the process of loading the containers on board of a deep-sea vessel. Lastly, the research is done 15 years ago, the performance of the equipment and the size of the vessels is changed over time…Mechanical, Maritime and Materials EngineeringMarine and Transport TechnologyTransport Engineering and Logistics2016.TIL.807
Optimization to reduce waiting times at locks
No full textCargo transport is important for supplies of goods and food worldwide. The goods and food are transported by inland transport modes, like road, railway and inland waterway transport. Due to a growth in world population the demand of goods and food increase, which results in an increase in cargo transport. More transport leads to more traffic on roads, railway and inland waterways. Roads and railways are at their limits of capacity, while inland waterways have the ability for an increase of waterway traffic. The European waterway network for inland transport is wide and connects sea ports and the European hinterland with each other. While inland waterway transport is cheaper and slower compared to road and railway transport, it is also less reliable than the other transport modes. The reliability is the ability to perform and maintain the agreements made between actors involved in the cargo transport during normal circumstances, as well as during unexpected circumstances, like delays. A disadvantage for inland waterway navigation is the locks in the waterway network. Locks maintain the water on a certain level, so that inland vessels can navigate without the risk of grounding. Locks also cause congestions when the lock capacity is lower than the amount of vessels that want to pass the lock during a certain time. These congestions cause delays for the vessels, which do not increase the reliability of the waterway transport mode. To stimulate the inland waterway transport, so that cargo transport over water will become more compatible with road and railway transport, the reliability for inland waterway transport must be increased. This can be done by decreasing the waiting times at the locks in the waterways. The interarrival times of inland vessels have an influence on the waiting times for vessels at locks. So when the arrival times of vessels at a lock are controlled, the waiting times can be controlled. A simulation model is designed to simulate what the effect of controlled arrivals of vessels is on the waiting time at locks. The simulation model is able to adjust the velocities of all vessels that navigate on the waterway towards the lock. With the adjustments in velocity of a vessels, the arrival time of that vessel is controlled by the model. The adjustment in velocity can let the vessel navigate faster and slower. When a vessel slows down it´s travel time on the waterway increases. A lower velocity and a longer travel time means that the fuel costs decrease and the operation time increases. That could result in a higher total travel costs. Therefore the simulation model can minimize the waiting times or the travel costs. With the minimization of costs the waiting time is minimized as well, due to the fixed operating costs a vessel has, which are the travel costs without fuel costs. Variables in the simulation model are the length of the controlled waterway, the service time of the lock and the lock capacity. The length of the controlled waterway determines the size of the system and the area where the vessels are controlled, the service time is the time it takes to pass the lock and the lock capacity determines the maximum number of vessels that can enter the lock chamber during one service run. As an input for the simulation model each vessel has its own entering velocity and entering time during a day. The entering velocities of the vessel are between 10 and 20 km/h and can be adjusted between 5 km/h and 110% of the entering velocity of the vessel. The entering time is distributed by a Weibull distribution. Since inland waterway transport is not yet a full 24/7 industry, the distribution is in such way that the majority of the daily vessels that pass the lock will enter the system between 06:00 and 20:00 hour. The settings used for the variables in the simulations are based on situations at four locks between the two largest sea ports in Europe, which are Rotterdam and Antwerp. The locks are the Volkerak, Krammer, Hansweert and Kreekrak lock and the waiting times at those lock are expected to be as long as 3 hours in the next decade. The output of the simulation model are waiting times at the lock and the total travel costs for each vessel in an optimized situation, which is a situation with minimized waiting time or travel costs. Optimized results of the simulation model are compared with the simulations that represent the four locks with the expected waiting times. The simulation model shows a reduction in waiting times of more than 80% compared to the expected waiting times for the next decade. If this can be achieved in reality the inland waterway transport mode can be stimulated due to smaller unexpected delays and an increased reliability.Mechanical, Maritime and Materials EngineeringMarine and Transport TechnologyTransport Engineering and Logistics2012.TEL.771
Controlled reefers in the banana supply chain: energy reduction and quality preservation
No full textBananas are the fourth most eaten product in the world. In Europe almost 4 million tonnes of bananas are consumed every year. These bananas need to be imported from plantations in, among others, Latin America. Due to the perishable nature of bananas the transport from plantation to Europe is done in cooled reefer containers. During transport the considered ripening stage is the green life period because of its ability to be influenced by external factors. The ripening rate within the green life period is largely dependent on temperature. Energy consumption for cooling is dependent on temperature. The current cooling strategy is keeping temperatures as low as possible without damaging the bananas, which asks for maximum energy consumption. This article proposes a cooling strategy which results in higher temperatures inside the container with the goal to minimize energy consumption which ultimately results in lower transport costs. In this article a model is developed to combine biological and logistical features of the banana supply chain. To assure right quality, continuous monitoring inside a reefer is used to check the ripening process and detect disturbances in ripening rate. To cope with the disturbances and delays in the supply chain a controller is designed which can adjust temperature during the journey. This is done to adjust the ripening rate and make sure the bananas reach the customer at the right quality at the right time while using up to 10% less energy than conventional cooling strategies.Mechanical, Maritime and Materials EngineeringMarine and Transport Technolog
Multi-agent model predictive control with applications to power networks
No full textTransportation networks, such as power networks, road traffic networks, water distribution networks, railway networks, etc., are the corner stones of our modern society. As transportation networks have to operate closer and closer to their capacity limits and as the dynamics of these networks become more and more complex, control of these networks has to be advanced to a higher level using state-of-the-art control techniques. Such control techniques should be able to deal with the large size and distributed nature of the control problems encountered, and should in addition be able to anticipate undesired behavior at an early stage. In this PhD thesis several novel control techniques for the control of transportation networks are proposed. Each of the techniques proposed is based on a combination of ideas from the fields of multi-agent systems and model predictive control. Control problems from the domain of power networks are used to illustrate and assess the performance of the proposed techniques.TrailMechanical, Maritime and Materials Engineerin
Multi-Agent Control for the Transportation Networks of the Future
No full textWhen going from Delft to Rotterdam by train, or when driving over the highway, or biking through the Dutch landscape, did you ever look up and wonder about the network of power lines that span our country? The electricity network, globally on of the largest structures created by mankind, is a complex system consisting of thousands of power transmission lines, power generation stations, transformer substations, and consumption points. Day-in day-out electricity is transmitted from one point to another to enable our modern life. Can you imagine living a whole day with no electricity at all? To ensure efficient and secure operation of power networks, network operators adjust controls in the network to meet certain control objectives. These controls consist of adjusting power generation, changing transformer taps, switching off consumption, etc. Control objectives typically consist of maintaining values of network variables like voltages and frequency at or close to pre-specified values. The values of these network variables can be manipulated by changing the flows of power over the network. Network operators change the flow of power over the network by using the controls available. Although controlling the flows has always been a challenging task, the ever growing increase of energy consumption, the changes in the power market, and the increasing appearance of small scale, so-called embedded generation, make the control of power networks in the future become even more challenging.Delft Center for Systems and ControlMechanical, Maritime and Materials Engineerin
Process improvement to shorten the lead-time for the ACD division in the distribution center of L’Oreal located in Alphen aan den Rijn
No full textThe thesis is about process improvements in L’Oréal’s distribution center. A framework is developed based on the DMAIC approach, process improvement methodologies and solution evaluation methodologies. The current process is discussed and with the Lean Six Sigma philosophy measures of improvement are developed. The requirements for the improvements were creating flow, reducing waiting time and providing insights. Five different improvements were developed and the dynamic wave and resource planning improvement scored the highest. This improvement provides a tool for the supervisors and team-leaders to prioritize work and manage the employees in a better way. A simulation was conducted to quantify the lead-time improvement. This simulation resulted in a reduction in lead-time of 5 till 47% bringing the average lead-time down from 2.2 to 1.5 days.Mechanical, Maritime and Materials EngineeringMarine and Transport TechnologyTransport, Infrastructure and LogisticsTIL506
A comparison of the performance of automated vehicles in container terminals
No full textSince the beginning of the container era in the middle of the last century, the containerization is growing. Nowadays more than 90 percent of the world's non-bulk cargo is being transported in a container. With many shipping liners deploying container vessels on the world seas, the competition is strong. Therefore the need to lower the price per container has resulted in the development of vessels with increasing sizes. Continuing on cost reduction, shipping lines tend to make fewer port calls with their vessels. Consequently the call sizes, i.e. the number of containers that have to be unloaded and loaded per vessel in a port, will increase. Because of the competition between harbours and container terminals, terminal operators are forced to handle container vessels as fast as possible while reducing the operational costs. Especially in high labour regions, the labour costs are a large part of the operational costs of container terminals. Accordingly, in 1993, automation in container terminals has been introduced at ECT's Delta terminal in Rotterdam in order to reduce the operational costs. Nowadays there are several (semi-) automated container terminals worldwide and currently there are four in the planning stage. In the first automated container terminals automated guided vehicles (AGVs) were used for the supply and discharge of containers at the quay cranes. Because the realized quay crane productivities were lower than the theoretical ones, and this is partly caused by the transportation vehicles, other vehicles such as the Lift-AGV (L-AGV) and the automated shuttle carrier (AShC), were developed. In this research a comparison is made between several state-of-the-art automated vehicles to find out which is the best performing vehicle type. The characteristics of the investigated automated vehicles are elaborated to gain insight in the possibilities of the vehicles. These characteristics are inuencing the behaviour of the automated vehicles in container terminals. Whether or not the interchange between the stack or quay cranes is linked and what the vehicles manoeuvrability is. Due to the complexity of the system a simulation model should assist in answering the research questions. Taking into account the requirements, a simulation model has been set up using the simulation library of TBA. A benchmark model has been defined using the dimensions of automated container terminals which are currently being build of already exist. For each type of vehicle, terminal layouts are designed within the benchmark model. The design of these layouts largely depend on the size of the vehicle and the manoeuvrability, which manoeuvres can it perform and how much space is needed. An implementation for each type of vehicle and each terminal layout has been made in the model. Several peak scenarios with a varying number of vehicles has been performed in order to obtain results which assist in answering the research questions. The output of the model is also used to gain more insight in the behaviour of the different vehicle types. The results must also contribute to the validation of the model. Considering the quay crane productivities, it can be concluded that according to the results of the performed experiments the L-AGV is the best performing vehicle, however the differences with the other vehicles are not very large. This result deviates from the expectation, which can be explained using the generated output of the simulation. However the use of another type of quay crane could be beneficial for other vehicle types.Mechanical, Maritime and Materials EngineeringMarine and Transport TechnologyTransport Engineering and Logistics2013.TEL.779
Adaptive control for autonomous ships with uncertain model and unknown propeller dynamics
No full textMotion control is one of the most critical aspects in the design of autonomous ships. During maneuvering, the dynamics of propellers as well as the craft hydrodynamical specifications experience severe uncertainties. In this paper, an adaptive control approach is proposed to control the motion and trajectory tracking of an autonomous vessel by adopting neural networks that is used for estimating the dynamics of the propellers and handling hydrodynamical uncertainties. Considering that the maneuvering model of a vessel resemble a nonlinear non-affine-in-control system, the proposed neural-based adaptive control algorithm is designed to estimate the nonlinear influence of the input function which in this case is the dynamics of propellers and thrusters. It is also shown that the proposed methodology is capable of handling state dependent uncertainties within the ship maneuvering model. A Lyapunov-based technique and Uniform Ultimate Boundedness are used to prove the correctness of the algorithm. To assess the method's performance, several experiments are considered including trajectory tracking simulations in the port of Rotterdam.Accepted Author ManuscriptTransport Engineering and Logistic
- …
