Hesitant bi-fuzzy multi-attribute information fusion with expert reliability

Selecting alternatives wisely is essential in any multi-attribute decision-making (MADM) process, and it requires the participation of multiple experts. Initial decision data and the contributing experts' reliability affect the preciseness of the decision outcomes. However, no expert possesses exhaustive knowledge, and their estimations may exhibit varying preferences. Many times, experts hesitate to give contradictory/ conflicting information. Hence, the information gathered from them includes the hesitancy of the expert and conflict. To account for expert influence while preserving the integrity of the original data during aggregation, this paper proposes a hesitant bi-fuzzy information fusion (HBFIF) method that integrates expert reliability (ER) into the decision-making framework in the conflicting framework. Initially, the impression of expert reliability in the decision-making process is examined, and a reliability metric is constructed based on the degree of similarity among expert opinions. The power average (PA) operator is then used with expert reliability to aggregate expert preferences while preserving as much of the original data as feasible. Subsequently, inspired by the TOPSIS method, the hesitant bi-fuzzy information fusion approach is developed to maintain experts' original opinions and risk preferences depending on their assessed reliability. Finally, the proposed HBFIF method is applied to the multi-attribute decision-making framework in selecting an efficient hydrogen storage method in automobiles to confirm its effectiveness and applicability, underscoring its practical relevance and implementation.