Premium
Geographic and process information for chemical plant layout problems
Author(s) -
Özyurt Derya B.,
Realff Matthew J.
Publication year - 1999
Publication title -
aiche journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.958
H-Index - 167
eISSN - 1547-5905
pISSN - 0001-1541
DOI - 10.1002/aic.690451015
Subject(s) - integer programming , geographic information system , dimension (graph theory) , mathematical optimization , computer science , process (computing) , linear programming , elevation (ballistics) , space (punctuation) , data mining , operations research , mathematics , geography , remote sensing , geometry , pure mathematics , operating system
Geographic information systems (GIS) introduced here offer a source of information for problems with a geographic dimension, such as layout and environmental decision making. A mixed‐integer linear‐programming formulation of a chemical‐plant layout problem was developed, and is presented along with an existing formulation. The effectiveness of these formulations is highly problem‐ and data‐specific, and a combination of the two formulations can be derived that is more effective on certain problems than either formulation alone. The formulations are ineffective for some problems due to the highly redundant search space engendered by symmetries in the layout—symmetries that can be eliminated by a more careful analysis of the situation using location‐specific information such as existing units and facilities, geographic constraints, wind, elevation, and soil conditions. Different classes of such information can be captured from a GIS and represented within the mixed‐integer linear‐programming formulation. Two cases studies presented illustrate the formulations and the use of geographic information for chemical‐plant layout.