Conclusion
Firefly Algorithm (FA) was applied to find the lowest
makespan (Cmax) of five benchmarking JSSP datasets adopted
from the OR-Library. Experimental design and analysis were
carried out to investigate the appropriate parameters setting of
Table 6 FA’s results from different parameter settings.
Instances
Using parameter settings used by other research
Optimised parameters setting
from the previous experiment
Best
Known
Solution
Apostolopoulos and Vlachos
[24]
Lukasik and Zak [10]
Horng and Jiang [25]
Min Avg SD Min Avg SD Min Avg SD
DS6×6 58 59.5 0.97 60 61.8 1.87 55 56.5 1.08 55
DS10×5 593 604.4 9.37 602 636.7 22.70 593 593 0.00 593
DS15×5 969 996 14.34 991 1040.6 32.02 958 958 0.00 958
DS20×5 1414 1491.5 36.27 1457 1537.8 43.02 1310 1366.9 32.32 1207
DS15×10 1435 1478.8 32.67 1511 1592.2 40.67 1323 1394.4 36.11 1046
บทความวิจัย ( in press) วารสารวิชาการเทคโนโลยีอุตสาหกรรม ปีที่ 8 ฉบับที่ 1 มกราคม – เมษายน 2555
The Journal of Industrial Technology, Vol. 8, No. 1 January – April 2012
the FA. The one-third fractional factorial experimental design
can reduce the number of experimental runs by 66.67%
compared with the conventional full factorial design. Ranges
of FA parameters used by previous research were reviewed
and investigated. The investigation was aimed to study the
effect of the FA parameter setting on its performance before
comparing the FA results between using and not using
optimised parameter settings. In this research, the optimised
setting of the FA parameters of nG, α, β0, and γ parameters
was suggested at 100*25, 0.5, 1, and 0.1, respectively.
Moreover, the proposed algorithm with appropriate
parameters setting produced the best-so-far schedule better
than the FA without adopting parameter settings. It also found
the best known solution in some cases. It should be noted that
the appropriate parameter settings of the proposed algorithms
may be case specific based on the nature and complexity of
the problem domains.
Conclusion
Firefly Algorithm (FA) was applied to find the lowest
makespan (Cmax) of five benchmarking JSSP datasets adopted
from the OR-Library. Experimental design and analysis were
carried out to investigate the appropriate parameters setting of
Table 6 FA’s results from different parameter settings.
Instances
Using parameter settings used by other research
Optimised parameters setting
from the previous experiment
Best
Known
Solution
Apostolopoulos and Vlachos
[24]
Lukasik and Zak [10]
Horng and Jiang [25]
Min Avg SD Min Avg SD Min Avg SD
DS6×6 58 59.5 0.97 60 61.8 1.87 55 56.5 1.08 55
DS10×5 593 604.4 9.37 602 636.7 22.70 593 593 0.00 593
DS15×5 969 996 14.34 991 1040.6 32.02 958 958 0.00 958
DS20×5 1414 1491.5 36.27 1457 1537.8 43.02 1310 1366.9 32.32 1207
DS15×10 1435 1478.8 32.67 1511 1592.2 40.67 1323 1394.4 36.11 1046
บทความวิจัย ( in press) วารสารวิชาการเทคโนโลยีอุตสาหกรรม ปีที่ 8 ฉบับที่ 1 มกราคม – เมษายน 2555
The Journal of Industrial Technology, Vol. 8, No. 1 January – April 2012
the FA. The one-third fractional factorial experimental design
can reduce the number of experimental runs by 66.67%
compared with the conventional full factorial design. Ranges
of FA parameters used by previous research were reviewed
and investigated. The investigation was aimed to study the
effect of the FA parameter setting on its performance before
comparing the FA results between using and not using
optimised parameter settings. In this research, the optimised
setting of the FA parameters of nG, α, β0, and γ parameters
was suggested at 100*25, 0.5, 1, and 0.1, respectively.
Moreover, the proposed algorithm with appropriate
parameters setting produced the best-so-far schedule better
than the FA without adopting parameter settings. It also found
the best known solution in some cases. It should be noted that
the appropriate parameter settings of the proposed algorithms
may be case specific based on the nature and complexity of
the problem domains.
การแปล กรุณารอสักครู่..