Reducing Truck Emissions in Import Operations at Container Terminal—A Case Study in a Singaporean Port - Technology Management for Sustainable Production and Logistics - page 141

After all arrival times have been determined, the arrival time of each truck is gen-
erated. In each period, every truck arrival is randomly generated based on the uni-
form distribution.
When a truck arrives at the entrance, it is assigned to the lane that has the short-
est queue length. Then the time for entrance process for each truck is generated
following an Erlang distribution.
The last process is the pick-up process at the block. The truck usually parks
on the aisle along the block and next to the bay of the container to be picked up
by the particular truck. Thus, the processing time for picking containers can be
estimated as the summation of the total shuffling time of upper containers and the
time moving the required container from its position to the truck. A heuristic algo-
rithm is applied to shuffle the upper containers to other positions. The container on
the top will be shuffled. Then, the second-top container is shuffled. This shuffling
procedure is repeated until there is no container above the required container. For
moving a container, the position which locates at the lower tier will have higher
priority. If more than one position, we select the one that takes shortest time for
moving containers. After picking up the container, the truck exits the container
terminal immediately. In this simulation process, the truck waiting time at gate
and in yard is recorded. The total waiting time is the summation of waiting times
of all trucks. In this study, the simulation is executed on Visual basic language.
Compared with using standard simulation software, it is more convenient and effi-
cient to program GA and simulation in the same programming language.

4.3 Procedure of Solution Algorithm

The procedure of whole solution algorithm is shown as below

Step 0 Generate an initial population

Step 1 For each chromosome in the population

Step 1.1 For each substring of chromosome

Generate a random number between 0 and 1 to select the arrival period the
truck corresponding to the container represented by this substring

Step 1.2 For each period

• Randomly generate the arrival pattern of truck in this period
• Randomly generate arrival time for each truck arriving in this period

Step 1.3 Simulate the gate and the yard system operations

Step 1.4 Calculate the total waiting time and record it

Step 1.5 Repeat from Step 1.2 to Step 1.4 for a sufficient number of times

and then calculate the average total waiting time

Step 1.6 Calculate the objective function value of this chromosome

Step 2 Calculate the fitness function of each chromosome