An Agent-Based Modeling Dynamic Hybrid Model for Project Management in Research and Development.

Area of Science:

• Engineering Management
• Innovation Management
• Computational Modeling

Background:

• Research and Development (R&D) projects face inherent uncertainties affecting technological progress.
• Predicting R&D project evolution and technological maturity is crucial for strategic decision-making.
• Hybrid modeling approaches, integrating System Dynamics (SD) and Agent-based Modeling (ABM), offer advanced capabilities but are underutilized in R&D contexts.

Purpose of the Study:

• To present a novel hybrid SD-ABM framework for predicting the technological maturity of R&D projects.
• To analyze the impact of project structure and team size on R&D project dynamics and efficiency.
• To validate the model's alignment with empirical observations in R&D project management.

Main Methods:

• Developed a multilevel AB-SD hybrid model integrating system-level feedback (work phases, rework) with agent-level interactions (team members, tasks).
• Simulated early-stage innovation projects in the oil and gas sector under various scenarios.
• Compared base-case (parallel tasks) with sequential and mixed parallel-sequential task execution strategies, varying team sizes.

Main Results:

• Sequential task execution reduced rework duration by 88% compared to the base case.
• In parallel configurations, teams of 4-5 members demonstrated optimal performance, reducing project duration and improving task completion.
• Increasing team size beyond optimal levels led to diminishing returns due to communication complexity and management delays.

Conclusions:

• The proposed AB-SD hybrid framework effectively captures R&D project uncertainties and emergent dynamics.
• Quantitative insights into resource allocation, task scheduling, and technology maturity progression are provided.
• Findings support empirical evidence on the impact of team size on R&D project efficiency and coordination.