Multi-phase matching mechanism for stable and optimal resource allocation in cloud manufacturing platforms Using IF-VIKOR method and deferred acceptance algorithm

Abstract

ABSTRACT Public platforms are one of the most important Cloud Manufacturing modes. Public platforms enable an environment for manufacturers and demanders to freely and directly connect with each other. Exploiting the high potentials of public platforms depends on final matching. This paper has developed a multi-phase matching mechanism for stable and optimal resource allocation in public platform. The proposed mechanism grades the demanders using an intuitionistic fuzzy VIKOR method using three measures of quality, time, and sustainability. Then, the mechanism clusters the demanders based on these three measures; and finally, allocates the clusters using the Deferred Acceptance (DA) algorithm to the manufacturers. The mechanism is examined using a case study from the Iranian automotive industry. The paper has extended and examined the model under three directions of: the grading method impact, clustering analysis impact, and the platform mode impact. Based on the experiments, the intuitionistic fuzzy TOPSIS-DA results on average 4.34% better outcome rather than the intuitionistic fuzzy VIKOR-DA. The proposed heuristic and optimized clustering method results on average 1.09% better solution rather than the KM clustering method. Also, the analysis of the PoS reveals that a private platform yields on average 10% better utility rather than a public platform.