This paper describes the development and application of a Decision Support System (DSS-R) as an aid in allocating parallel or stand-by redundancy to improve N-stage series systems reliability. The DSS-R developed stands as an alternative to the known algorithms for systems reliability optimization. It is applicable whenever we face the incompleteness, imprecision, inaccuracy and uncertainty of data and relations in redundant systems modeling. Uncertainty is handled by fuzzy sets and fuzzy variables. The DSS-R combines causal knowledge and surface knowledge, which allows a condensed design of the knowledge base. The DSS-R developed is a simple and efficient applications software programmed for use on a standard microcomputer under MS-DOS. The software is written in Arity/Prolog programming language. The DSS-R output includes: (1) the list of recommended redundant stages, (2) the effectiveness of resource expenditure in redundancy, (3) suggestions concerning increase (decrease) of redundancy depending on budget (weight, volume) available.