Report
開發新的粒子群聚法求解流程式排程問題(I) ; A Revised Discrete Particle Swarm Optimization for Permutation Flow-Shop Scheduling Problem
| العنوان: | 開發新的粒子群聚法求解流程式排程問題(I) ; A Revised Discrete Particle Swarm Optimization for Permutation Flow-Shop Scheduling Problem |
|---|---|
| المؤلفون: | 陳春龍 |
| المساهمون: | 資訊管理學系 |
| سنة النشر: | 2012 |
| المجموعة: | National Chengchi University Institutional Repository (NCCUIR) |
| مصطلحات موضوعية: | 粒子群聚法, 流程式生產排程問題, 總完成時間, 基因演算法, 蟻群演算法, Particle Swarm Optimization, Permutation Flow Shop Scheduling, Makespan, Genetic Algorithms, Ant Colony Optimization |
| الوصف: | 本研究擬開發新的粒子群聚法(RDPSO)求解常見的以總完成時間為目標的流程式生產排程問題。我們首先透過完整的探討全體最佳解（global best solution）與個體最佳解（personal best solution）在粒子的搜尋過程中所扮演的引導角色來開發新的群體學習策略(swarm learning strategies)。然後，我們將開發一個新的過濾區域搜尋法（filtered local search）來過濾粒子已經搜尋過的區域，將粒子的搜尋導向尚未開發的區域，以避免粒子過早陷入區域最佳解。如果粒子的搜尋還是陷入區域最佳解，我們將開發三個新的逃離策略(escape strategies)來幫助粒子逃離區域最佳解。本研究將分兩年完成，第一年我們將專注於群體學習策略的研究，以開發一個基本的RDPSO （Basic RDPSO）。第二年我們將專注於過濾區域搜尋法與逃離策略的研究，以開發一個多階的RDPSO（Multi-phase RDPSO，簡稱MRDPSO）來改善Basic RDPSO的效能。我們將使用常用的Taillard 測試問題組，以基因演算法（GA），蟻群演算法，和其它的粒子群聚法（PSO），來評估Basic RDPSO與MRDPSO的效能。 ; This research proposes a revised discrete particle swarm optimization (RDPSO) to solve the permutation flow-shop scheduling problem with the objective of minimizing makespan (PFSP-makespan). The candidate problem is one of the most studied NP-complete scheduling problems. RDPSO proposes new particle swarm learning strategies to thoroughly study how to properly apply the global best solution and the personal best solution to guide the search of RDPSO. A new filtered local search (FLS) is developed to filter the solution regions that have been reviewed and guide the search to new solution regions in order to keep the search from premature convergence. In addition, three escape strategies are proposed to help the search escape if the search becomes trapped at a local optimum. This proposed research will be conducted in two years. The first year will be dedicated to investigating the new particle swarm learning strategies and developing a basic RDPSO heuristic. In the second year, we will study the filtered local search and escape strategies to further improve the results of the basic RDPSO and also develop a multi-phase RDPSO heuristic (MRDPSO). Computational experiments on Taillard`s benchmark data sets will be performed to evaluate the effectiveness of the basic RDPSO and the MRDPSO by comparing their performance with that of Genetic Algorithms (GA), Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO). |
| نوع الوثيقة: | report |
| وصف الملف: | 535191 bytes; application/pdf |
| اللغة: | unknown |
| Relation: | 計畫編號 NSC101-2221-E004-004; https://nccur.lib.nccu.edu.tw//handle/140.119/84761; https://nccur.lib.nccu.edu.tw/bitstream/140.119/84761/1/101-2221-E004-004.pdf |
| الاتاحة: | https://nccur.lib.nccu.edu.tw//handle/140.119/84761 https://nccur.lib.nccu.edu.tw/bitstream/140.119/84761/1/101-2221-E004-004.pdf |
| رقم الانضمام: | edsbas.6D65321A |
| قاعدة البيانات: | BASE |
| الوصف غير متاح. |