Dynamic weighted idle time heuristic for flowshop scheduling
The constructive heuristic of Nawaz, Enscore and Ham (NEH) has been introduced in 1983 to solve flowshop scheduling. Many researchers have continued to improve the NEH by adding new steps and procedures to the existing algorithm. Thus, this study has developed a new heuristic known as Dynamic Weight...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 语言: | English English English |
| 出版: |
2017
|
| 主题: | |
| 在线阅读: | http://eprints.uthm.edu.my/742/1/24p%20AMIRA%20SYUHADA%20ZAINUDIN.pdf http://eprints.uthm.edu.my/742/2/AMIRA%20SYUHADA%20ZAINUDIN%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/742/3/AMIRA%20SYUHADA%20ZAINUDIN%20%20WATERMARK.pdf |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| 总结: | The constructive heuristic of Nawaz, Enscore and Ham (NEH) has been introduced in 1983 to solve flowshop scheduling. Many researchers have continued to improve the NEH by adding new steps and procedures to the existing algorithm. Thus, this study has developed a new heuristic known as Dynamic Weighted Idle Time (DWIT) method by adding dynamic weight factors for solving the partial solution with purpose to obtain optimal makespan and improve the NEH heuristic. The objective of this study are to develop a DWIT heuristic to solve flowshop scheduling problem and to assess the performance of the new DWIT heuristic against the current best scheduling heuristic, ie the NEH. This research developed a computer programming in Microsoft Excel to measure the flowshop scheduling performance for every change of weight factors. The performance measure is done by using n jobs (n=6,10 and 20) and 4 machines. The weight factors were applied with numerical method within the range of zero to one. Different weight factors and machines idle time were used at different problem sizes. For 6 jobs and 4 machines, only idle time before and in between two jobs were used while for 10 jobs and 20 jobs the consideration of idle time was idle time before, in between two jobs and after completion of the last job. In 6 jobs problem, the result was compared between DWIT against Optimum and NEH against Optimum. While in 10 jobs and 20 jobs problem the result was compared between DWIT against the NEH. Overall result shows that the result on 6 and 10 jobs problem the DWIT heuristic obtained better results than NEH heuristic. However, in 20 jobs problem, the result shows that the NEH was better than DWIT. The result of this study can be used for further research in modifying the weight factors and idle time selections in order to improve the NEH heuristic. |
|---|