Bilinear Programming: Part II. Application of Bilinear Programming.
Abstract
In the paper a number of new problems such as constrained bematrix game, multi-stage Markovian assignment problem, complementary (orthogonal) planning problem, the problem of reducing a sparse matrix into an almost-triangular matrix by row and column permutations, a location problem on a rectangular network, etc., are defined and formulated as the bilinear programming problem (BLP): maximize C(supt) x + d(supt) y + x(supt) Cy subject to x belongs to X, y belongs to Y. where X and Y are m and n-dimensional polyhedral convex set, respectively. Further, it is shown that several important classical problems such as 0 - 1 integer programs, maximization problem of a convex quadratic function subject to linear constrints, two-move game, etc. are reducible to equivalent BLP's. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1971
- Accession Number
- AD0737650
Entities
People
- Hiroshi Konno
Organizations
- Stanford University