Sustainability in operations and supply chains is becoming more popular among academics and practitioners through Sustainable Supplier Selection (SSS). In addition to balancing economic, social, and environmental factors, the awareness of the United Nations Sustainable Development Goals 2030 has affected the selection of long-term suppliers, ensuring green operations and sustainable supply chains. The criteria for SSS have multiple dimensions and are interdependent; this mimics the real-world scenario rather than assuming independently from an analytic hierarchy process. We use the multi-criteria decision-making (MCDM) model, combining DEMATEL-based on ANP (called DANP) with VIKOR, to solve the SSS problem. The DANP method is used to model and assess the interdependent relationships between criteria. Then, ranking the available alternatives and selecting the best one can be accomplished using the VIKOR method. We consider the electronic manufacturing industry in Taiwan as an empirical case. This study, in addition to selecting the best sustainable supplier, demonstrates the use of influential network relationship maps to analyze and improve the gaps in each dimension and criterion.

Proton exchange membrane fuel cell (PEMFC) is a clean and efficient alternative technology for transport applications. The degradation analysis of the PEFMC stack plays a vital role in electric vehicles. We propose a hybrid method based on a deep neural network model, which uses the Monte Carlo dropout approach called MC-DNN and a sparse autoencoder model to analyze the power degradation trend of the PEMFC stack. The sparse autoencoder can map high-dimensional data space to low-dimensional latent space and significantly reduce noise data. Under static and dynamic operating conditions, using two experimental PEMFC stack datasets the predictive performance of our proposed model is compared with some published models. The results show that the MC-DNN model is better than other models. Regarding the remaining useful life (RUL) prediction, the proposed model can obtain more accurate results under different training lengths, and the relative error between 0.19% and 1.82%. In addition, the prediction interval of the predicted RUL is derived by using the MC dropout approach.