Publications

Spatially synergizing the urban rail transit (URT) network integrated with its feeder transit system and the urban form plays an important role in improving the effectiveness of URT in mitigating traffic congestion, reducing air pollution, optimizing urban spatial structure, etc. This article mainly focuses on assessing the spatial synergy between the two parts by specifying the assessment criteria that can effectively characterize the spatial synergic mechanism between the two parts and developing a novel multicriteria large-scale group assessment (MCLSGA) model, in which basic uncertain linguistic information (BULI), as an extended form of fuzzy linguistic approach, is used to model and process the subjective assessment information (Assess-Inf) elicited by experts. In order to alleviate the computing complexity, an agglomerative hierarchical clustering algorithm is introduced to cluster the Assess-Inf given by a huge number of experts as per the organizers' expected discrimination level on reliability-related Assess-Inf. A method for weighting the clusters is then proposed to control the roles played by the items of the Assess-Inf, representing each cluster and having different reliability levels in their fusing process to tap into wisdom of crowds (WOC) while accommodating the organizers' trust level in the reliability-related Assess-Inf given by experts. Afterward, the best–worst method is extended to the BULI-based large-scale group assessment context with the aim of accurately weighting the criteria by drawing on WOC. Finally, a case study on assessing the spatial synergy between Chongqing's integrated URT system and urban form is conducted to validate the validity of the proposed MCLSGA model.

The benchmarking of agri-food 4.0 supply chain (Agri4SC) falls under the multiple criteria problem in supply chain visibility (SCV) and supply chain resource integration (SCRI) for improving data analytics capabilities and achieving sustainable performance (SP). It is considered a multiple criteria decision-making (MCDM) problem due to three main concerns, namely, multiple Agri4SC evaluation criteria including the SCV, SCRI and SP criteria. These criteria have relative importance and trade-offs. Despite the tremendous efforts over the last years, none of the developed Agri4SCs have met all of the essential Agri4SC evaluation criteria. Another concern raised in the evaluation and benchmarking of the Agri4SC is the uncertainty of experts. Thus, the main contribution of this research is to propose an Agri4SC benchmarking framework in SCV and SCRI for improving data analytics capabilities and achieving SP based on an extension of the proposed Fermatean probabilistic hesitant fuzzy sets (FPHFSs) and MCDM methods. The methodology process is divided into six main parts. Firstly, an Agri4SC decision matrix is formulated based on the intersection of the Agri4SC alternatives and criteria to cover multiple Agri4SC evaluation criteria issues. Secondly, novel FPHFSs are proposed along with their operational laws, score function, accuracy function, Fermatean probabilistic hesitant fuzzy average mean operator and Fermatean probabilistic hesitant fuzzy weighted average operator. The FPHFS can encompass more sophisticated and uncertain evaluation information. Thirdly, Fermatean probabilistic hesitant fuzzy weighted zero inconsistency is formulated to assign weights to the evaluation criteria. Fourthly, the Fermatean probabilistic hesitant fuzzy decision by opinion score method (FPH-FDOSM) is formulated and used to score the alternatives that were evaluated subjectively based on SCV criteria. Fifthly, the FPH-FDOSM-based multi attributive ideal-real comparative analysis (MAIRCA) scoring method with equal probabilities is proposed to score Agri4SC alternatives that were evaluated subjectively based on weighted economic, environmental and social factors. Lastly, the MAIRCA ranking method with unequal probabilities is introduced to benchmark Agri4SC alternatives that were evaluated objectively based on the weighted subcriteria of SP and the trade-offs amongst the identified criteria. The robustness and reliability of the results are tested via sensitivity analysis and Spearman's correlation coefficient.

In this paper, we propose a novel center-based clustering algorithm based on bidirectional conical information granularity. The main purpose is to fully absorb the semantic information of the ordinal relationship between objects to improve the performance of central clustering in identifying interleaved and imbalanced data. The proposed algorithm includes two main stages: (i) the stage of determining the cluster center and (ii) the division stage. In the stage of determining the cluster center, the first cluster center is determined by using the number of conical information granularity in the data, and the remaining cluster centers are determined by defining the statistical measure of "fuzzy importance degree". In the division stage, we divide the points to be clustered into stable and active areas. The former quickly and accurately identifies and assigns the objects belonging to a cluster by measuring the fuzzy similarity between the objects to be clustered and the cluster center, and the latter assigns the objects in the active area by using the information of the points already assigned. This method describes the position and sorting relationship of objects that are granulated through ordinal relationships more accurately in the global environment, thereby gaining a more comprehensive understanding of the structural characteristics of the data. This helps to improve the accuracy and stability of clustering algorithms in handling interleaved and imbalanced data. This paper uses three clustering validity indicators to test the performance of our algorithm. We compare the results with those of six different types of popular clustering algorithms and new algorithms proposed in recent years. The experimental results show that the algorithm proposed in this paper can identify clusters more accurately on the datasets with a complex and staggered distribution. It is significantly better than the clustering algorithm participating in the comparison and has good robustness on datasets with added noise.