A cloud-rough reliability allocation model using the best-worst method and decision-making trial and evaluation laboratory

Reliability allocation is important in the product design phase, and unreasonable allocation results may affect product reliability, resource utilization , and overall product quality . As product structures become increasingly complex, uncertainties are inherently involved in reliability allocations, and the connections between subsystems become more closely interrelated. However, previous studies have some limitations, such as ignoring the influences of multiple uncertainties and treating subsystems as separate entities for computational simplicity, which may compromise the precision of the allocation results. Hence, this paper develops an improved reliability allocation model that integrates cloud rough numbers (CRNs), the Best-Worst Method (BWM), and the Decision-Making Trial and Evaluation Laboratory (DEMATEL) to fully address these limitations. First, the concept of CRN is introduced by combining the advantages of cloud modeling, which deals with individual uncertainty, and rough numbers, which handle uncertainty within the expert group. Then, the modified BWM is used to determine the weights of the subsystems, and the modified DEMATEL is employed to analyze the interconnection between these weighted subsystems, allowing for a more comprehensive consideration of the importance of the factors and their intrinsic linkages. Finally, an example analysis, sensitivity analysis, and comparative studies are conducted to verify the validity of the developed model.