B.J. Vrolijk
Behandeling van een groot containerschip. Een simulatie in Taylor II.
Computer program,
Report 95.3.LT.4565, Transport Engineering and Logistics.
One of the ideas of the 2000-12 project-team to load and unload large
containerships, is to use a crane buffer with a capacity of one or two hours.
Automatic Guided Vehicles (AGV's) are used to transport containers from the
crane to the buffer, vice versa.
This report is about a Taylor II simulation concerning the described process.
First is shown how the model is built. The model is written in such a way
that the process can be followed 'real time' on the screen.
The global working of the model can be summarised as follows:
A container enters the system at the mother stack, according to an arrival
schedule. Passing a 'black box', the container is transported to an elevator.
The (available) elevator with the lowest throughput is served first. The
elevator lifts the container to a plateau, in a TRION (a kind of a Automatic
Stacking Crane). When the TRION receives a container, it puts it in his
stackrow. While the stackrow is not active, the process repeats itself. At a
certain time, an AGV arrives. As soon as the AGV enters the 'loading-path'
(connecting dock-stackrow), the stackrow is activated. The stackrow sends a
container to the TRION, which puts it (via fictious buffers) on the waiting
AGV. After acceptation, the AGV moves to the crane to deliver the container
and after that the AGV retuns to the stack.
After the model is built, several experiments are done. The most important
question asked here is: What is the optimum number of AGV's in the current
situation. First a calculation based on averages is presented. After that a
simulation in Taylor II is done. With a warming-up period and different numbers
of AGV's, an optimal AGV-configuration is found. Obvious is that, because of
the deviation in the crane's service-time, more AGV's than expected are
necessary.
After the experiment, conclusions and recommendations for future research
are given. In the conclusion the model's validation and verification are
presented. Also the author's opinion of working with Taylor II is given. The
recommendations are on possible further extensions to the model and details
concerning the model.
Reports on Transport Engineering and Logistics (in Dutch)
Modified: 2008.01.21;
logistics@3mE.tudelft.nl
, TU Delft
/ 3mE
/ TT
/ LT.