Optimizing Intelligent Hotel Room Resource Allocation Using Deep Learning Techniques

In light of the accelerated advancement of intelligent hospitality establishments, the effective distribution of room resources has emerged as a pivotal research area, with the objective of enhancing operational efficacy and customer satisfaction. To address this issue, this paper proposes a novel room resource allocation optimisation model based on deep learning, which builds upon the existing Transformer architecture. In particular, an enhanced Transformer model has been devised, which incorporates the time embedding module into the network architecture in order to capture the temporal dynamic characteristics of customer booking behaviour. Furthermore, the ability to identify key features has been enhanced by introducing a weighted self-attention mechanism. Furthermore, the model output layer incorporates a multi-objective optimisation strategy, which serves to balance the relationship between customer satisfaction and hotel revenue. The experimental results demonstrate that the proposed model exhibits superior performance in terms of room utilisation, resource allocation efficiency and customer satisfaction when compared to the traditional method.