Testing Implementations of PPS-methods for Interval Linear Systems

This paper compares various implementations of the parameter partitioning methods (PPS-methods) for optimal outer estimation of the solution sets to interval linear systems. Special attention is focused on the analysis of the modification based on Rohn’s technique, which is the most complex, laborious, but the most efficient one. We present results of numerical experiments with several computational schemes as well as their analysis, and we formulate practical recommendations on how to optimize the computational schemes of the parameter partitioning methods.