Traditional single-layer network models exhibit fundamental limitations in characterizing multilayer interaction within complex social systems, leading to inadequately address the evolving complexities of modern social networks. This limitation has driven the extension of the Influence Maximization (IM) problem to multilayer networks, where identifying influential nodes is challenged by intricate structural properties and cross-layer node interactions. In this paper, we propose a novel Degree-Entropy-Delay (DED) algorithm with a hybrid seed selection strategy. The algorithm combines layer weights, node degree, and information entropy to quantify node importance. To mitigate overlapping influence caused by clustered high-degree nodes, we incorporate the delay mechanism to optimize seed node selection. Comprehensive experiments conducted on eight publicly available real-world networks demonstrate the effectiveness of our algorithm in maximizing influence spread within multilayer networks. The DED algorithm achieves up to 33% higher influence coverage than state-of-the-art methods in large-scale dense networks under the SIR model.