A Novel Human-Machine Collaborative Interface for Aero-Engine Pipe Routing

Despite the rapid advance of information technology (IT), a fully automated Enterprise Information System (EIS) for designing and planning of complex products and assemblies is still not available, and human interventions are required in one way or another. Therefore, one of the challenges in developing an EIS for complex multidisciplinary design is how to design human-machine interfaces and integrate experts' knowledge in decision-making. In this paper, a rule-based methodology is proposed to take into consideration experts' experience; the rules are applied based on human designers' visual perception. The strength of IT and human experts can be synthesized to find a suboptimized solution efficiently. To illustrate the performance of the developed methodology, an EIS for aero-engine pipe routing is used as a case study. Pipe routing is a typical no-deterministic polynomial-time hard (NP-hard) problem; in addition, many factors, such as fluid dynamics, sequences of pipes, installation, maintenance, machinability, stability, and vibration, have to be taken into consideration simultaneously. Experts' guidance is essential to find a vital solution in the confined 3-D space. In the implementation, virtual objects have been determined to define feasible spaces for pipes based on the analysis of fluid structure interaction. A new algorithm is developed to find the possible graphs of pipes routes and improve the solutions with the help of human visual perception. The effectiveness of the proposed methodology has been verified.