二つの「交換型突然変異」の発想の必然性

概要

論文の詳細を見る
In the Bulletin of Niigata University of International and Information Studies [No.4], I proposed the character-preserving mutation and announced that genetic manipulation was effective in the scheduling of a certain synthetic resin factory. The character-preserving mutation was applied to a Job Shop scheduling problem which was a sample problem for the measurement of efficiency. And, it was confirmed to be statistically significant, as reported in JCKBSE-2002. This research object prohibits including the same elements in one schedule. Therefore, an individual phenotype has to be created as a case of genetic manipulation by a genetic algorithm for the order expression. However, if crossover and mutation are done using the order expression, a good gene sequence in the previous generation may not leave the prototype. Many reports have indicated that it is not suitable to apply a genetic algorithm to a scheduling problem. However, other reports issued at almost the same time as mine proposed the same idea that I proposed, using a different operational method. One of the purposes of the mutation is that the characteristics of the previous generation are not inherited by the next generation. I became interested in the idea of the character-preserving mutation coming out of other researcher, and I examined the mutation in the scheduling problem and the differences between the two experimental techniques.
新潟国際情報大学の論文
2003-03-20