Abstract
The timely and cost-effective completion of construction projects pushes on the implementation of optimized scheduling strategies. This study investigates the integration of Agent-Based Modeling (ABM) with Proximal Policy Optimization (PPO) within a Critical Chain Project Management environment aiming to enhance construction scheduling. This scenario-based environment uses an experimental and case study research methodology aiming to advance scheduling by using a reinforcement learning technique that could be renowned for its efficacy in such as dynamic environments. This proposed hybrid algorithm to offer a more realistic way of the construction process uses a gamificative environment (simulation) on a 3D model named as “Dragonfly Tower”. Findings indicate that the proposed hybrid model significantly improves scheduling by reducing unpredictability and volatility, outperforming traditional scheduling methods due its unique function. The results highlight the critical role of hybrid algorithms, particularly PPO within CCPM frameworks, in optimizing resource and schedule efficiency to ensure successful project delivery.