Ensemble Approach with Multi-Agent and Swarm Based Algorithms in Quadratic Assignment Problem
Yazarlar (3)
Arş. Gör. Elif Meşeci Zonguldak Bülent Ecevit Üniversitesi, Türkiye
Doç. Dr. Emrullah SONUÇ Karabük Üniversitesi, Türkiye
Doç. Dr. Caner ÖZCAN Karabük Üniversitesi, Türkiye
| Bildiri Türü | Tebliğ/Bildiri | Bildiri Dili | İngilizce |
| Bildiri Alt Türü | Tam Metin Olarak Yayınlanan Tebliğ (Uluslararası Kongre/Sempozyum) | ||
| Bildiri Niteliği | Alanında Hakemli Uluslararası Kongre/Sempozyum | ||
| Kongre Adı | 2nd International Conference on Trends in Advanced Research | ||
| Kongre Tarihi | 22-11-2024 / 23-11-2024 | ||
| Basıldığı Ülke | Türkiye | Basıldığı Şehir | Konya |
| Bildiri Linki | https://drive.google.com/file/d/19YfamIBROfCfP6o2NLkVeQ3KuqWlE9Zw/view?usp=drive_link | ||
| UAK Araştırma Alanları |
Yapay Zeka
|
||
| Özet |
| The quadratic assignment problem (QAP) is the problem of assigning facilities to locations such that each facility is assigned to a precise location and each location is assigned to a precise facility to decrease the overall cost expressed as the sum of flow and distance products. QAP is used to increase efficiency and reduce costs in various areas, such as layout, resource allocation, and inventory management. In his paper, the ensemble approach is proposed to solve QAP effectively. The proposed ensemble method enables agents to acquire information more effectively during the learning process by exploiting the feature of multi-agent reinforcement learning (MARL) to explore different regions of the solution space by running multiple agents in parallel, and the Bayesian updating feature of discrete rat swarm optimization (DRSO). Although the application of both methods individually permitted the analysis of the optimization performance, the ensemble approach, which combined these two methods, provided considerable enhancements in terms of temporal performance. Experimental studies show that multi-agent-based algorithms provide efficient results, especially for quadratic problems. The proposed method achieves excellent results in terms of accuracy and efficiency on samples of different sizes. While the results show the difficulty of ensuring temporal efficiency, they emphasize the compensatory effect of the ensemble method. |
| Anahtar Kelimeler |
