Ying Shi

Federated learning (FL), which is a distributed privacy computing technology, has demonstrated strong capabilities in addressing potential privacy leakage for multisource data fusion and has been widely applied in various industries. Existing contribution evaluation mechanisms based on Shapley values uniquely allocate the total utility of a federation based on the marginal contributions of participants. However, in practical engineering applications, participants from different data sources typically exhibit significant differences and uncertainties in terms of their contributions to a federation, thus rendering it difficult to represent their contributions precisely. To evaluate the contribution of each participant to FL more effectively, we propose a novel interval federated Shapley value (IntFedSV) contribution evaluation mechanism. Second, to improve computational efficiency, we utilize a matrix semitensor product‐based method to compute the IntFedSV. Finally, extensive experiments on four public datasets (MNIST, CIFAR10, AG_NEWS, and IMDB) demonstrate its potential in engineering applications. Our proposed mechanism can effectively evaluate the contribution levels of participants. Compared with the case of three advanced baseline methods, the minimum and maximum improvement rates of standard deviation for our proposed mechanism are 11.83% and 99.00%, respectively, thus demonstrating its greater stability and fault tolerance. This study contributes positively to promoting engineering applications of FL.

Over the past decades, multimodal transportation has played a crucial role in modern logistics and transportation systems due to its high capacity and low cost. However, multimodal transportation, which is mainly driven by fossil fuels, potentially contributes to significant carbon emissions. In the context of global sustainable development, reducing carbon emissions from the transportation sector has far-reaching implications for supporting society-wide carbon neutrality. In this paper, we have developed for the first time a many-objective multimodal transportation route optimization (MTRO) model that simultaneously considers economic cost, carbon emission cost, time cost, and customer satisfaction, and solve it using the Non-dominated Sorting Genetic Algorithm Version III (NSGAIII). Second, to further improve the convergence performance, we introduce a fuzzy decision variable framework to improve the NSGAIII algorithm. This framework can reduce the search range of optimization algorithm in the decision space and make it converge better. Finally, we conducted a large number of simulation experiments on test problems to verify the applicability and superiority of the improved algorithm, and applied it to MTRO problems under uncertain demand. This work fills the research gap for MTRO problems and provides guidance for relevant departments to develop transportation and decarbonization plans.

As a crucial basic natural resource, water resources are the cornerstone for sustainable national economic development. This paper takes the 2014 pilot water rights trading policy (WRT) as an entry point and uses a difference-in-differences (DID) model to test the policy effect of WRT on water-use efficiency (WUE) based on panel data for 30 Chinese provinces from 2005 to 2021. The study shows that WRT can significantly improve the regional WUE, and these results remain valid after a series of robustness tests, such as the parallel trend test, placebo test, and PSM-DID. Mechanistic analysis revealed that WRT can produce the Porter effect, which affects the WUE through technological innovation. The results of the heterogeneity analysis based on the synthetic control method (SCM) showed that WRT effectively improved WUE in Jiangxi, Henan, Ningxia, Hubei, and Guangdong, but did not achieve the expected effect in Inner Mongolia or Gansu. This paper provides solid empirical support for assessing the effectiveness of WRT and accelerating the process of establishing a unified national WRT market in China by 2025.

It is still necessary to regularly investigate the breakdown of socio-economic elements as a starting point for analyzing the effects of diverse human production activities on PM_2.5 intensity from industrial and regional viewpoints. In this paper, the emission factor model was adopted to measure PM_2.5 emissions in the Beijing–Tianjin–Hebei (BTH) region at the regional and industrial levels. The logarithmic mean Divisia index (LMDI) decomposition model was employed to analyze the factors affecting PM_2.5 emissions related to energy consumption in the BTH region at the regional and sectoral levels. Building on this foundation, a system dynamics (SD) model was established to make a prediction regarding PM_2.5 pollution in the BTH region in 2025. This study found that secondary industry was a major source of PM_2.5 emissions in the BTH region. Coal remained the main form of energy consumption in the BTH region. Secondly, the effect size of the factors affecting PM_2.5 intensity ranked in the order of energy intensity, energy structure, and industrial structure. Thirdly, in 2025, PM_2.5 emissions in the BTH region will decline appreciably, but there is still a certain gap in terms of meeting the targets of “the 14th Five-Year Plan” between the three provinces and cities. These results indicate that the BTH region should achieve the effective management of PM_2.5 pollution at the source through the following initiatives: it is necessary to carry out the continuous adjustment of energy structures to gradually increase the proportion of clean energy; we must steadily promote the decline in energy intensity reduction, and gradually strengthen scientific and technological innovation; and we must continue to promote the optimization of the industrial structure and increase the proportion of tertiary industry every year.