Affiliations: [a] Dongling School of Economics and Management, University of Science and Technology Beijing, Beijing, China | [b] Institute of Industrial Engineering and Ergonomics, RWTH Aachen University, Aachen, Germany
Correspondence:
[*]
Corresponding author: Xiaodong Zhang, Dongling School of Economics and Management, University of Science and Technology Beijing, Beijing 100083, China. E-mail: [email protected].
Abstract: Collaborative Product Development (CPD) process is characterized by autonomous task control, dynamic task sequence, and frequent team collaboration, which endow the process with high flexibility and uncertainty. To make the process predictable and improve process efficiency, it is essential to model, simulate, and analyze the process by considering all these characteristics. As most CPD models are developed to simulate task structures and task-flow process, our work focuses on studying the human working behaviors in CPD process by agent-based simulation, which we think is the main source of process uncertainty and flexibility. In this paper, the local task scheduling behavior of designer agents and the resource conflict resolution behavior of the manager agent are studied under the framework of agent-based simulation. In the simulation, the designer agent schedules his local tasks according to his personal priority, and the manager agent resolves resource conflicts according to collective utility function. To make the process more efficient, a collective utility function is developed for the manager agent, in which task priority, spare resource, and waiting time are considered simultaneously. To validate the effectiveness of the collective utility, simulation experiments are designed and carried out among entire-waiting utility, entire-replacement utility, and collective utility strategy. The simulation results indicate that the proposed utility strategy can effectively shorten the project total time and cost of the case. There are two engineering applications of this approach. First, by simulating and evaluating different task scheduling and resource conflict resolution behaviors of the agents, designers and managers can adopt suitable scheduling strategies at the planning stage of the CPD project. Second, in the executing process of the CPD project, the proposed scheduling priority and utility functions can be used in the dynamic scheduling by developing software agents in the CPD management systems.
