pRTMNSGA-III: a novel multi-objective algorithm for QoS-aware multi-cloud IoT service selection

Abstract

The Internet of Things (IoT) is an emerging technology expected to play a significant role. The integration of IoT with cloud computing (CC) has enabled the creation of large-scale networks of interconnected smart devices and services. To provide optimal quality of service (QoS) for users, it is necessary to address the conflicting requirements of IoT services. The service selection problem is considered NP-hard and therefore requires using metaheuristic algorithms for efficient resolution. This paper presents a novel hybrid multi-objective metaheuristic, pRTMNSGA-III, which combines the strengths of RNSGA-III and TMNSGA-III to generate solutions that meet user preferences and eliminate unfavorable ones. To further optimize computational time, a parallel solution evaluation approach is employed. In addition, a more effective fuzzy membership function is proposed to select the best solution based on the requester’s preferred QoS. The proposed algorithm is evaluated on multiple datasets, and the results demonstrate its superiority over existing state-of-the-art algorithms, including RNSGA-III, TMNSGA-III, NSGA-III, pNSGA-II, NSGA-II, and NSPSO Experimental results demonstrate that pRTMNSGA-III outperforms existing algorithms by up to 37% in terms of the number of non-dominated solutions generated.