Revised Simplex Method
In real life the linear programming matrices are thin and sparse (there are usually more columns than rows, and most of coe±cients are likely to be zeros). According to the characteristic of the LP problems and the way the simplex method (SM) works, it can be easily be shown that a signi¯cant amount of redundant information is generated at each step. This necessitates extra computational e®ort and computer storage. To overcome the drawback of the standard method of pivoting, a computationally more e±cient procedure `Revised Simplex Method (RSM)' is presented. The RSM is an implementation of the standard simplex method that uses an abbreviated tableau, reconstructing only that data from the complete simplex tableaus absolutely necessary to perform the steps of the simplex method.