Multicriteria decision optimization for the design and manufacture of structural aircraft parts
- PDF / 2,029,173 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 11 Downloads / 159 Views
ORIGINAL PAPER
Multicriteria decision optimization for the design and manufacture of structural aircraft parts C. Fortunet1 · S. Durieux1 · H. Chanal1 · E. Duc1 Received: 18 May 2020 / Accepted: 25 July 2020 © Springer-Verlag France SAS, part of Springer Nature 2020
Abstract This paper concerns an optimization method applied to the design for manufacturing (DFM) of aircraft structural parts. Today, the aerospace industry has to produce aircraft less and less costly. Usual design and manufacturing process are not sufficiently efficient. A dedicated DFM method ensures that all manufacturing requirements are taken into account at the design stage. However, all requirements cannot be satisfied together; thus the best compromise must be found. The proposed approach is based on the formulation of 5 design and manufacturing performance indicators to be satisfied. From the geometrical modelling of the problem, an NSGA II genetic algorithm computes a population of one thousand permissible solutions. Thus, a decision process is applied to identify the best compromise according to the behaviour of the decision maker, using Topsis and the AHP method. This methodology is applied in an industrial context to an aircraft structural part manufactured by stamping and machining. The optimal part geometry is then calculated for three different airplane configurations. Such tests are used to extract geometric design rules. In addition, the paper highlights the impact of the user’s behaviour on the computed results. Keywords Design for manufacturing · Aircraft structural parts · Multicriteria optimization · Genetic algorithm · Decision aid method
1 Introduction Today, the aviation industry is faced with two major requirements: the large increase in the number of aircraft to be manufactured and the necessity to reduce costs. To meet these expectations, the industry must design cheaper and simpler aircraft, in a strongly competitive context. Currently, the design and manufacturing process remains traditional, with poor connections between the design, the forming of raw materials and the machining of finished parts. Reducing costs and simplifying manufacturing processes entails implementing a common design approach to take into account all the constraints arising from the manufacturing process right from the design stage. Thus, the problem of the mechanical parts design becomes a problem of optimization under constraints. Optimization is functionally multi-criteria but can become * E. Duc emmanuel.duc@sigma‑clermont.fr 1
Université Clermont Auvergne, CNRS, SIGMA Clermont, Institut Pascal, 63000 Clermont‑Ferrand, France
single-criterion to simplify the process. The constraints are imposed by the functional requirements or by the technological processes implemented. The literature presents many studies to solve this problem. The goal is usually to identify the best solution from among a candidate population. The population is computed using, for example, a genetic algorithm (GA), while the best solution (which best meets the criter
Data Loading...
