Akademik Məlumat İdarəetmə Sistemi
Applied Soft Computing, vol.185, 2025 (SCI-Expanded, Scopus)
The smart manufacturing performance in renewable energy projects should be improved. However, it is very necessary to determine the most important indicators to use limited resources of the companies more efficiently. However, there are a limited number of studies in the literature that determine the most important factors. Accordingly, the purpose of this study is to generate appropriate smart manufacturing investment strategies with renewable energy efficiency. For this purpose, a new fuzzy decision-making model has been created. Firstly, the balanced expert dataset is constructed using Q-learning algorithm. After that, the criteria are evaluated with molecular fuzzy Bayesian networks-based weighting. The third stage gives information about the ranking the alternatives with molecular fuzzy multi-objective particle swarm optimization (MOPSO). The main contribution of this study is to identify prior investment strategies regarding smart manufacturing investment projects for renewable energy efficiency by establishing a novel model. The proposed model has some essential superiorities in comparison with the previously created ones. Integration of molecular geometry with fuzzy decision-making modeling helps to manage uncertainties in the calculation process more effectively. The findings demonstrate that conversion process and system integration are the most important factors for renewable energy efficiency in smart manufacturing. These two criteria stand out due to both their central role in energy transformation efficiency and their strong interconnectivity with other operational dimensions of smart manufacturing. Unlike cost efficiency, conversion efficiency and system integration directly influence the performance and adaptability of renewable energy systems. Their prioritization is supported by both the expert evaluations and their dominant weight scores in the fuzzy decision model. While the proposed model involves multiple advanced techniques, such as Q-learning, Bayesian networks, and multi-objective optimization, it is structured in a modular and transparent way. This design ensures that each stage of the decision-making process remains interpretable for practitioners, allowing the model to deliver robust results without sacrificing usability. On the other side, organic ranking cycles and smart energy systems are the most important investment alternatives for smart manufacturing with renewable energy efficiency.