Artificial intelligence methods in process plant layout

  • Authors:
  • A. McBrien;J. Madden;N. R. Shadbolt

  • Affiliations:
  • Univ. of Nottingham, Nottingham, UK;-;Univ. of Nottingham, Nottingham, UK

  • Venue:
  • IEA/AIE '89 Proceedings of the 2nd international conference on Industrial and engineering applications of artificial intelligence and expert systems - Volume 1
  • Year:
  • 1989

Quantified Score

Hi-index 0.00

Visualization

Abstract

A constraint-based system, PLS, capable of deriving a technically acceptable three-dimensional layout for a chemical process plant of typical complexity is described. In the application of PLS to an exemplar plant, over five hundred instances of constraints were identified and manipulated.PLS has been implemented in a frame system created in-house. This features recursively structured frames to represent both the initial plant data and the intermediates derived by the system. Reasoning is conducted by applying rules in which the pattern and action are represented as complex frames.A particular feature of PLS is its powerful problem reduction abstraction. A nested hierarchy of groups is formed by PLS. The groups are formed for sound engineering reasons and reflect the underlying experience of layout engineers.