This study aimed to investigate mechanisms for enhancing project completion speed through the application of artificial intelligence techniques. The study adopted the approach of "Using LSTM Networks for Predicting Project Delay Times," and the researchers utilized data from 3530 residential units for training, testing, and prediction. Selecting project delay times as a focus was driven by their significant impact on vital project completion. The research problem centers around the main question, "Can LSTM networks be successfully used to predict project delay times?" The significance of the study lies in utilizing Long Short-Term Memory (LSTM) neural network techniques to improve the prediction of project delay times, thereby enhancing project planning and management, reducing delays, and increasing efficiency in execution. Among the key findings, it was revealed that LSTM networks can effectively enhance the prediction of construction project delay times, exhibiting high accuracy and retrieval rates. Introducing this advanced technology to project management can lead to improved scheduling, planning, and reduced delays, ultimately contributing to enhanced work efficiency, productivity, and more accurate strategic decision-making.

Research Type: Research Paper.