Synergizing Global and Local Strategies for Dynamic Project Management: An Advanced Machine Learning-Enhanced Framework
George Sklias, Socratis Gkelios, Dimitrios Dimitriou, Maria F. Sartzetaki, Savvas A. Chatzichristofis
Abstract
In this study, we introduce a versatile and scalable optimization tool designed to address several critical project management needs. Our aim is to provide project managers with a robust decision support system that enhances and streamlines decision-making processes. Building upon our previously proposed global scheme—which optimizes project schedules by adjusting dates to match each task’s optimal period—we introduce a novel local scheme. This innovative addition leverages a Machine Learning pipeline, specifically utilizing the Silverkite algorithm, to facilitate long-horizon forecasting. By synergistically combining global and local optimization strategies, we elevate project management efficiency, maximizing potential benefits. This tool is equipped to handle a wide array of variables, offering real-time, consultative support throughout the project’s lifecycle. Through the demonstration of various scenarios, we showcase the effectiveness and adaptability of our optimization tool, underscoring its value in contemporary project management contexts.