The design and simulation of an autonomous system for aircraft maintenance scheduling. (November 2019)

• Record Type: Journal Article
• Authors: Liu, Yinling; Wang, Tao; Zhang, Haiqing; Cheutet, Vincent; Shen, Guohua
• Is Part Of: Computers & industrial engineering. Volume 137(2019)
• Publication Date: 2019-11
• Subjects: Maintenance system -- Systems Engineering -- Architecture design -- Agent-based modelling and simulation; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2019.106041
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: A novel autonomous system framework for aircraft maintenance is proposed. The framework integrates System Engineering and Agent-Based Modeling and Simulation. A variety of maintenance processes are considered. First time a significant exploration of real-world maintenance data is performed. Maintenance cost and service level are evaluated in the experiments. Abstract: Operational support is a key issue for aircraft maintenance, which aims to improve operational efficiency and reduce operating costs under the premise of ensuring flight safety. Although many works have emerged to achieve this aim, they mostly address the concept of maintenance systems, the relationship between stakeholders and the loop of maintenance information separately. Hence, the cooperation between stakeholders could be impeded especially when urgent decisions should be made, relying on historical data and real-time data. In this paper, we propose an innovative design of an autonomous system supporting the automatic decision-making for maintenance scheduling. The design starts from the proposition of the analysis framework, to concept formulation of the system, to information transitional level interface, and ends with an instance of system module interactions. The underlying architecture illustrates the high-level fusion of technical and business drives; optimizes strategies and plans with regard to maintenance costs, service level and reliability. An agent-based simulation system is developed as a proof to illustrate the feasibility of the system principle and algorithms. Furthermore, the simulation experiment analyzing the impact of maintenance sequence strategies on maintenance costs and service level has demonstrated the algorithm functionality and the feasibility of the proposed approach.