In the last decade, with the new situation forced by the Covide-19 pandemic, the information systems are often forced to work remotely, they must communicate and share confidential data with several interlocutors. In such a context, ensuring the confidentiality of communications becomes a complex and difficult task. Hence, the need to have a flexible system that can adapt with different parameters involved in every exchange of information. We recently presented in [1] a new smart approach to data encryption that serves the same purpose. This approach uses the concept of artificial intelligence and apply BNL skyline algorithm to decide about the most suitable algorithm to ensure the best data privacy. However, with the evolution of dimensions and criteria to be considered for this smart encryption, we find that the complexity of the BNL algorithm increase, then, the response time of the application increase and the skyline encryption quality decreases. In this work, we propose a new idea to resolve this problematic. Indeed, this contribution consists in adding another Intelligence brick to dynamically define the Skyline algorithm depending on the type and number of dimensions. Through this paper, we provide an analysis and a comparison of some skyline algorithms for the multidimensional search. The results obtained by this study show the performance of this new approach whether in terms of execution time or in the quality of the dominant encryption solution.