Updating Lu Factors of Lp Simplex Bases
نویسنده
چکیده
Methods for updating the LU factors of simplex basis matrices are reviewed. An alternative derivation of the Fletcher and Matthews method is given. This leads to generalizations of their method which avoids problems with both the Bartels and Golub method and the Fletcher and Matthews method. The improvements are to both numerical stability and data access locality. The resulting updating algorithm is preferred to the Reid variant of the Bartels and Golub method for both numerical stability and cost of execution.
منابع مشابه
High performance simplex solver
The dual simplex method is frequently the most efficient technique for solving linear programming (LP) problems. This thesis describes an efficient implementation of the sequential dual simplex method and the design and development of two parallel dual simplex solvers. In serial, many advanced techniques for the (dual) simplex method are implemented, including sparse LU factorization, hyper-spa...
متن کاملDYLP: a dynamic LP code
DYLP is a full implementation of the dynamic simplex algorithm for linear programming. Dynamic simplex attempts to maintain a reduced active constraint system by regularly purging loose constraints and variables with unfavourable reduced costs, and adding violated constraints and variables with favourable reduced costs. In abstract, the code alternates between primal and dual simplex algorithms...
متن کاملDYLP : a dynamic lp code SFU - CMPT TR 1998 - 23
DYLP is a linear programming code implementing a dynamic simplex algorithm similar to that outlined by Padberg in [5]. The code attempts to maintain a reduced active constraint system by regularly purging loose constraints and variables with unfavourable reduced costs, and adding violated constraints and variables with favourable reduced costs. In abstract, the code alternates between primal an...
متن کاملAn efficient approach to updating simplex multipliers in the simplex algorithm
The simplex algorithm computes the simplex multipliers by solving a system (or two triangular systems) at each iteration. This note offers an efficient approach to updating the simplex multipliers in conjunction with the Bartels-Golub and Forrest-Tomlin updates for LU factors of the basis. It only solves one triangular system. The approach was implemented within and tested against MINOS 5.51 on...
متن کاملOn the factorization of simplex basis matrices
In the simplex algorithm, solving linear systems with the basis matrix and its transpose accounts for a large part of the total computation time. The most widely used solution technique is sparse LU factorization, paired with an updating scheme that allows to use the factors over several iterations. Clearly, small number of fill-in elements in the LU factors is critical for the overall performa...
متن کامل