Combination of DOE and PIV methods for multi-criteria decision making

Abstract

When performing the multi-criteria decision making to choose the best solution, if some solutions are removed from the list of solutions or some solutions are added to the list of solutions, the decision making must be re-performed from the begining. This study proposes a new method to remove this limitation. The combination of the DOE (Design Of Experimental) method and PIV (Proximity Indexed Value) method is proposed in this paper. This combination is used to build the relationship between the scores of the solutions and the criteria. When the list of solution to be ranked has been removed or have been added some solutions, the ranking of some solutions only needs to use this relationship without having to recalculate from the beginning. Four different examples were applied to evaluate the effectiveness of the proposed method. The obtained results show that the proposed method ensures the required accuracy as well as its outstanding advantages. The limitations of the proposed method that need to be overcome are also pointed out at the end of this paper.

Selecting lubricating oil for two-stroke gasoline engines: a multi-criteria decision-making approach

“…In such cases, the PIPRECIA (PIvot Pairwise RElative Criteria Importance Assessment) method can be a viable alternative [20]. In the case where the number of lubricant types needs to be ranked either increased or decreased, combining the Design of Experiments (DOE) method with the COCOSO method can be applied to quickly rank the lubricant types without the need to redo the entire calculation process [21][22][23].…”