A Study of Indicators for Identifying Zero Variables in Interior-Point Methods
The ability to identify zero variables early on in an iterative method is of considerable value and can be used to computational advantage. In this work we first give a formal presentation of the notion of indicators for identifying zero variables, and then study various indicators proposed in the literature for use with interior-point methods for linear programming. We present both theory and experimentation that speaks strongly against the use of the variables as indicators; perhaps the most frequently used indicator in the literature. Our study implies that an indicator proposed by Tapia in 1980 is particularly effective in the context of primal-dual interior-point methods.
Citable link to this pagehttps://hdl.handle.net/1911/101719
MetadataShow full item record
- CAAM Technical Reports