Tesis doctoral de Sasa Jovanovic
Ports and their terminals in general are frequently studied and analysed as systems in the phase of initial planning and design and in the phase of operation as well. In this thesis, the main attention is focused towards the set of defined planning, design and operation parameters of the specialised port terminals whose primary function is handling of ships carrying passenger vehicles, commercial/industrial vehicles and other wheeled cargoes. Nevertheless, the applicability of research results is by no means restricted to the analyzed type of port terminals as they can be applied to any port terminals usually operating under conditions specified in this thesis. the thesis treats mathematical tools applicable for modelling of ro-ro/vehicle terminals which are herewith presented as queueing systems with multiple servers, batch (group, bulk) arrivals and unlimited waiting area. The first approach treats the terminal and its most important ship-berth link as a multi-server queueing system with poisson arrivals in bulk and exponential service (loading/unloading) times, that is, an mx/m/c/¿ system, which serves as a base model for another, more precise and more reliable queueing system. The second approach treats an advanced multi-server queueing system with poisson bulk arrivals and general service (loading/unloading) times, that is, an mx/g/c/¿ system. Statistical testing of available loading/unloading times collected over the years at the ro-ro terminals in the port of barcelona demonstrated that the empirical statistical distribution fits best with the inverse gauss (wald) and beta general theoretical statistical distributions. Therefore, a more precise model has been developed in order to obtain the selected parameters of the analyzed port/terminal queueing system. Port queueing system parameters and analytic parameters obtained from the aforementioned queueing models and under specified and assumed conditions are, for example, probability of finding the terminal idle, average ship waiting time at the anchorage, average ship time spent in the system (waiting and under loading/unloading operations), average number of ships in the queue (waiting at the anchorage), average number of ships in the system (waiting and under loading/unloading operations), ship turnaround time in port, etc. These parameters are used to forecast or evaluate the planning, design and operation parameters of the future or existing port terminal, respectively. although the queueing theory is not new in port planning, the state of the art demonstrated that the above specific queueing models have never been applied in port systems analysis in the way given in the thesis, due to their complexity. Nevertheless, the approximation methods, which are used for cumbersome queueing models such as mx/g/c/¿ model, made this research possible. obtained parameters of the port/terminal queueing system offer a fair overview of the existing operational scenarios at the existing ro-ro terminals or a reliable forecast for the future terminals in terms of planning, design and operational requirements. These parameters can be used for further analysis of terminal capacity, berth optimisation, dock labour scheduling and cargo handling equipment management. Further research directions could be focused on the above mentioned aspects of port planning, design and operation, based on the results obtained from the analyzed queueing systems which are more advanced that the ones used so far.
Datos académicos de la tesis doctoral «Stochastic and analytic models for planning, design and operation of ro-ro/vehicle terminals«
- Título de la tesis: Stochastic and analytic models for planning, design and operation of ro-ro/vehicle terminals
- Autor: Sasa Jovanovic
- Universidad: Politécnica de catalunya
- Fecha de lectura de la tesis: 20/01/2009
Dirección y tribunal
- Director de la tesis
- martos Dauer Rodríguez
- Tribunal
- Presidente del tribunal: francesc Lleal galceran
- zoran Radmilovic (vocal)
- rosa Castejón arqued (vocal)
- beatriz Blanco rojo (vocal)
