Modeling resource allocation and interactions in collaborative product development using Agent Based Simulation

Abstract

The main objective of the proposed research is to study collaborative and distributed development projects. The availability of resources, their properties (i.e., skills, learning-ability, etc.), their improvement (i.e. learning, advancing…), and their interaction dynamics (i.e., predecessor relationships, meetings, etc.) play a major role in the allocation and optimization of resources in such an environment (i.e., collaborative and distributed development environment). Using Agent-Based Simulation, the objective of this research is to estimate and reduce the effort and duration for any product development project using an innovative resource allocation strategy that considers the agents’ technical and managerial skills, their overall experience, and their learning ability. By simply adjusting the Agent-Based Simulation model’s configuration and settings, the model presents itself as a flexible tool to simulate the product development process and covers different approaches from the traditional methodologies such as the waterfall model, to the more recent PD processes like the Agile model. The model implemented was able to successfully demonstrate the usefulness of each of its features in influencing development time and effort. For example, it shows improvement in the development time when using the proposed resource allocation strategy, a decrease in the project time when agents cooperate on tasks, demonstrates the effect of learning on task execution, and the impact of rework on the PD process. Additionally, two case studies were introduced. The case studies show the flexibility of the model and the difference between the Agile model and the waterfall model, from an effort and time perspective. Finally, the model was validated using five different projects from the software development industry. Each of the project’s managers provided feedback and validated the output of the model. The model was successful in simulating several real-life projects related to the software development i