Table 2 shows the best and the average
values of each cost component including the total cost obtained for
the ten test runs for each data type. Based on these results, the
relative improvement values were obtained for all test problems
of each data type and are presented in Table 3. It can be seen that
the relative improvement of the total cost ranged from 7.92% to
20.83%, with an average of 14.56% for the hGNG algorithm, and
5.90–17.91%, with an average of 11.14% for the GNG algorithm.
Also, from this table, it can be seen that both of the proposed
algorithms can reduce costs greatly, especially the cost of farm utilization
and loss from hen mixing. However, the proposed hGNG
algorithm gave a lower cost of farm utilization and a lower loss
from hen mixing than the GNG algorithm. Since a lower number
of hen houses used may result in less hen mixing in the same
hen house, it should lead to lo a lower cost of farm utilization
and lower loss from hen mixing. To demonstrate the efficiency of
the hGNG algorithm on these two cost components, Table 4 shows
a comparison of the results of the current practice and those of the
GNG and hGNG algorithms for each data type. We can see that the
hGNG algorithm yields the lowest number of hen houses used.
However, the relative improvement in the transportation cost is
low for the algorithms. This is because, currently, the company
attempts to reduce this cost by allocating hens from the pullet
houses to the nearest hen houses, resulting in a low possibility
for the proposed algorithms to reduce this cost. However, this
practice will cause a high possibility to occupy more hen houses
and to mix hens with different ages in the same hen house
Table 2 shows the best and the averagevalues of each cost component including the total cost obtained forthe ten test runs for each data type. Based on these results, therelative improvement values were obtained for all test problemsof each data type and are presented in Table 3. It can be seen thatthe relative improvement of the total cost ranged from 7.92% to20.83%, with an average of 14.56% for the hGNG algorithm, and5.90–17.91%, with an average of 11.14% for the GNG algorithm.Also, from this table, it can be seen that both of the proposedalgorithms can reduce costs greatly, especially the cost of farm utilizationand loss from hen mixing. However, the proposed hGNGalgorithm gave a lower cost of farm utilization and a lower lossfrom hen mixing than the GNG algorithm. Since a lower numberof hen houses used may result in less hen mixing in the samehen house, it should lead to lo a lower cost of farm utilizationand lower loss from hen mixing. To demonstrate the efficiency ofthe hGNG algorithm on these two cost components, Table 4 showsa comparison of the results of the current practice and those of theGNG and hGNG algorithms for each data type. We can see that thehGNG algorithm yields the lowest number of hen houses used.However, the relative improvement in the transportation cost islow for the algorithms. This is because, currently, the companyattempts to reduce this cost by allocating hens from the pullethouses to the nearest hen houses, resulting in a low possibilityfor the proposed algorithms to reduce this cost. However, thispractice will cause a high possibility to occupy more hen housesand to mix hens with different ages in the same hen house
การแปล กรุณารอสักครู่..
