Metaheuristic approaches to grouping problems in high-throughput cryopreservation operations for fish sperm

详细信息	查看全文 | 推荐本文 |

• 作者：T.W. Liao^a ; ^{ieliao@lsu.edu} ; E Hu^b ; T.R. Tiersch^b
• 关键词：Heuristic ; Metaheuristic ; Simulated annealing ; Tabu search ; Ant colony optimization ; Grouping problem ; Hybrid metaheuristic ; Differential evolution ; Threshold accepting ; High-throughput cryopreservation ; Blue catfish
• 刊名：Applied Soft Computing Journal
• 出版年：2012
• 期刊代码：9_15684946
• 类别：cp
• 出版时间：August, 2012
• 卷：12
• 期：8
• 页码：2040-2052
• 文件大小：642 K

摘要

High-throughput cryopreservation operations of fish sperm is a technology being developed by researchers today. This paper first formulates a grouping problem in high-throughput cryopreservation operations of fish sperm and then develops a heuristic and four metaheuristic algorithms for its solution. The heuristic is modified from one originally proposed for the assembly line balancing problem. The four metaheuristic algorithms include simulated annealing (SA), tabu search (TS), ant colony optimization (ACO), and a hybrid differential evolution (hDE). For each metaheuristic algorithm, four different initialization methods were used. For both SA and TS, five different neighborhood solution generation methods were also studied. Real world data collected from a high-throughput cryopreservation operation was used to test the effectiveness of algorithms with different initialization and neighborhood solution generation methods. For comparison, a base line of grouping by processing order was also established. The results indicate that: (i) all algorithms performed better than the base line; (ii) using the result of the modified heuristic as the initial solution of metaheuristic algorithms lead to a better solution; the amount of improvement varied from algorithm to algorithm; (iii) among the five neighborhood solution generation operators, insertion operator was the best; (iv) among all algorithms tested, the hybrid differential evolution is the best, followed by tabu search in terms of average objective value.