陳彥匡陸萱2025-08-282025-08-282019-06-18U0061-3107201917095300https://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0061-3107201917095300https://nutcir-lib.nutc.edu.tw/handle/123456789/1221研究目的：本研究使用粒子群演算法求解Yang and Chen(1999)的貨架空間配置模型以追求零售商店利潤最大化。為驗證粒子群演算法求解Yang and Chen(1999)貨架空間配置模型的有效性，本研究將求解結果與Yang(2001)的啟發式演算法、基因演算法以及混合型啟發式演算法之求解結果進行比較。 研究方法：近年來族群式萬用啟發式演算法(Population-based Metaheuristic Algorithms)廣泛應用於各領域，其分為兩類：進化式演算法及群體智慧。過去貨架空間配置相關研究多使用進化式演算法，鮮少研究以群體智慧演算法求解。在群體智慧演算法中，由於粒子群演算法具有設定參數少、收斂速度快以及求解時間短等優勢，故本研究嘗試使用粒子群演算法求解Yang and Chen(1999)之貨架空間配置模型，以商店利潤最大化為目標，並透過6個問題集進行測試以瞭解演算法求解效果。 研究結果：結果顯示，粒子群演算法求得之結果優於Yang(2001)的兩種啟發式演算法、基因演算法及混合型啟發式演算法。為進一步驗證粒子群演算法於各問題集中的求解效果是否優於其他四種演算法，本研究採用Wilcoxon Rank-sum Test檢定，結果顯示粒子群演算法在6個問題集中一致優於其他四種啟發式演算法。 研究貢獻：由於過去貨架空間配置之研究多使用進化式演算法求得利潤，鮮少使用群體智慧，本研究嘗試使用粒子群演算法求解貨架空間配置問題，並驗證粒子群演算法求得之結果優於進化式演算法求得之結果。Object: This study investigates shelf space allocation decision of retail store in a manner that maximizes the overall store profit. We propose particle swarm optimization algorithm for the shelf space model of Yang and Chen (1999) to address the shop shelf space allocation problem. To show the validity of the proposed algorithm in addressing the problem, the results of particle swarm optimization algorithm are compared with results of Yang heuristic (2001), modified heuristic, genetic algorithm and hybrid metaheuristics. Methods: In recent years, population-based metaheuristic algorithms are the most selected to find optimal solution in many areas. There are two distinct forms of population-based metaheuristic algorithms which are evolutionary algorithms and swarm intelligence. Research to date has focus on evolutionary algorithms rather than swarm intelligence. Particle swarm optimization algorithm that belongs to the class of swarm intelligence, an attractive feature of which have few algorithmic parameters, converge fast, short computational time, and so on. Thus, this study presents the investigations on the application of particle swarm optimization algorithm to solve the shelf space allocation problem. Results: We compute the difference between each approach (Yang heuristic (2001), modified heuristic, genetic algorithm, hybrid metaheuristics, particle swarm optimization algorithm) and the best solution of the five approaches. The results show that the particle swarm optimization algorithm performs the best. To verify the validity of particle swarm optimization algorithm, we use Wilcoxon Rank-sum Test statistic. We can conclude from the result that the particle swarm optimization algorithm is a very efficient algorithm. Contributions: Most previous studies of the shelf space allocation decision are rarely refer to swarm intelligence. In this study, we apply the swarm intelligence to solve the shelf space allocation problem. The results show that the performance of swarm intelligence are better than evolutionary algorithms.zh貨架空間配置粒子群演算法Shelf SpaceAllocationParticle Swarm Optimizationmaster thesis