Integrating Deep Learning in Project Management: Improving Efficiency and Effectiveness in Project Planning and Execution

The purpose of this research is to explore the use of deep learning technologies, namely the K-Nearest Neighbors model within the textile manufacturing process to test the improvement of the said process. To properly implement the model, we must use real-time data from the process and evaluate the improvement depending on a number of factors: how much does the maintenance cost decrease, how accurate is the inventory management, and how well the program forecasts demand, all of which should lead to an improvement of profit. First, the analysis of the results proved that the implementation of KNN model leads to a significant decrease of maintenance costs: while without the model the reduction was 15%, with it cost maintenance cost a 25% decrease. Second, the usage of a model improves inventory management accuracy: before the use of KNN, it was 85%, while after it rose to 95%. Third, demand forecasting becomes more accurate, as the difference was 75% to the positive without the model and 90% with it. The use of a model improved all aspects, and the most improved was profit: before the implementation of the KNN model, profit was 10%, and it was 40% after, which led to the conclusion that the implementation of deep learning models, namely the KNN model, improves the effectiveness of textile manufacturing process and positively affects the profitability and the production process as a whole.