A partial-consensus and unequal-authority fuzzy collaborative intelligence approach for assessing robotic applications amid the COVID-19 pandemic

Assessing and comparing the overall performances of robotic applications amid the COVID-19 pandemic is a key task for local governments and relevant stakeholders, but has yet to be investigated. To accomplish this task, this study proposes a partial-consensus and unequal-authority fuzzy collaborative intelligence approach. In the proposed methodology, each evaluator first uses fuzzy geometric mean (FGM) to derive the fuzzy priorities of criteria for assessing the performance of each robotic application. Subsequently, considering the unequal authority levels of evaluators and the lack of an overall consensus, the partial-consensus fuzzy weighted intersection (PCFWI) operator is proposed to aggregate the derivation results. Finally, alpha-cut operations (ACO)-based fuzzy weighted average (FWA) is applied to evaluate the overall performance of each robotic application. The partial-consensus and unequal-authority fuzzy collaborative intelligence approach have been applied to assess the overall performances of four robotic applications amid the COVID-19 pandemic. Based on the experimental results, the Xenex LightStrike robot was named the #1 robotics application during the COVID-19 pandemic, followed by the Brain Navi Nasal Swab Robot. Furthermore, the proposed methodology outperforms three existing methods by up to 21% in preserving evaluators’ original judgments.