In the construction industry, cost control, resource inefficiency and schedule delays have been persistent challenges especially in large and complicated projects. The available systems of management are inflexible and do not use real-time data, which leads to sub-optimal decisions. To mitigate the drawbacks, the proposed study conceptualizes an AI- and IoT-based building planning management system that makes use of multidimensional planning to enhance forecasting, resources distribution, and risk evaluation. IoT sensors monitor real-time information about the costs, time and resource consumption, and the information undergoes processing by sophisticated models of AI, which are Support Vector Machines and Convolutional Neural Networks. The methodology involves the selection of features through Genetic Algorithm, preprocessing of the data, and the evaluation of performance on the basis of MSE, MAE, and the R 2. It was found in the experiments that the SVM model had the MSE = 0.00093, MAE = 0.022 and R 2 = 0.892, which had a high predictive power and the CNN model had MSE = 0.00107, MAE = 0.024 and R 2 = 0.876. The originality of this study is to integrate real-time IoT information with AI-based multidimensional optimization, and provide a flexible, scalable, and intelligent construction planning system that is highly efficient with minimization of risks in the project.