3423. Weight Prediction by Use of a Case-Based Reasoning Approach


SAWE Members get a $200 store credit each year.*

Become a SAWE Member

*Store credit coupon available at checkout, click the button in your shopping cart to apply the coupon.
Not applicable to SAWE textbooks and current conference technical papers.


Jan Pfaff, S. Rudolph, M. Dugas, R. Voit-Nitschmann: 3423. Weight Prediction by Use of a Case-Based Reasoning Approach. 2007.



To make weight prediction in the preliminary design phase more exact today, instead of using the well-known empirical approaches, investigations for developing analytical procedures are carried out. For several years, the weight department of the Airbus Future Project Office has been using the analytical weight prediction tool FAME (Fast and Advanced Mass Estimation), which meanwhile produces very good weight estimates, for example, for the wing structure (FAME-Wing). The Project FAME-Fancy is targeting the shortage of data for this tool in the early stages of preliminary design using different approaches. These approaches are being investigated at the Institute of Aircraft Design of the University of Stuttgart (IFB) with the objective of finding methods for generating the unknown parameters for these early stages of the preliminary aircraft design process. During the design process of an aircraft, a growing number of design parameters are defined so that the level of detail constantly increases. Within the FAME-Fancy project, methods are developed that will allow the process to be largely automated. Subsequently, input-data for FAME-Wing is obtained. The emphasis on techniques for the generation of this data is on knowledge-based, case-based, and constraint-based reasoning. The intention here is to show that there is the possibility of creating reliable parameter data in early stages of the preliminary design process with the help of a dimensional analysis approach based on similarity principles. Possible combinations and differences of these approaches are illustrated using several practical applications. The resulting data are validated and compared with existing empirical and half empirical preliminary design methods. Finally, an evaluation is performed.


SKU: Paper3423 Category: