Premium
Iterative techniques in optimization: II. The linearity difficulty in dynamic programming and quasilinearization
Author(s) -
McDermott Colin,
Lee E. S.,
Erickson L. E.
Publication year - 1970
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.690160422
Subject(s) - curse of dimensionality , computation , mathematical optimization , dynamic programming , computer science , linearity , dimensionality reduction , reduction (mathematics) , algorithm , iterative method , mathematics , artificial intelligence , engineering , electrical engineering , geometry
The quasilinearization technique is used to overcome the dimensionality difficulty in dynamic programming. The advantage of this approach is that not only the dimensionality or the fast memory requirement can be reduced, but the computation time required can also be reduced considerably. A technique is also devised to overcome the linearity difficulty encountered in this reduction in dimensionality. To illustrate the approach, the three dimensional crosscurrent extraction problem is solved as a one‐ and a two‐dimensional problem in the dynamic programming algorithm. Only approximately 1 min. is required to solve this one‐dimensional problem. However, if this problem were solved as a three‐dimensional problem, computation time of the order of hours would be required.