A Neural Network Approach to Modeling System Integration Sensitivity for Architectural Assessment


Performance sensitivity resulting from system integration is a subject commonly left untreated in the system architecting and early systems engineering phases. Ambiguity in the physical architecture often leads system designers to ignore this emergent behavior even though it can greatly alter the intended system performance. New methods of probing integration sensitivity are being researched. These methods employ computationally intensive methods to produce a set of performance predictions based on design variable iterations. There are two shortcomings with this approach. First, the computational intensity of the process limits the capacity to explore a large number of iterations. Second, a concise and computationally simple method is needed to encode this data and present it to system designers. In this paper, an artificial neural network is proposed to address both of these issues. The multilayer perceptron neural network architecture is developed based on a limited available data set. The results obtained are promising to support RF system development.

  • Abstract
  • Introduction
  • The Integration Sensitivity function
  • Neural Network Architecture and Results
  • Future Work
  • Conclusions
  • References

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In