The adequacy of multi-criteria assessment largely depends on how comprehensively an indicator system reflects a phenomenon under consideration. If the number of indicators is large, experts cannot adequately evaluate the indicator weights. As a result, the scope of calculations increases significantly, but the accuracy drops. This problem can be solved by forming a hierarchically structured indicator system. Multi-criteria assessment of such systems is started from the lowest hierarchical level where the values of the related indicator groups are estimated, and then they turn into the indices of a higher hierarchical level. Same procedure is repeated until a value representing the condition of a phenomenon under consideration is obtained.
No takes yet. Share an insight, caveat, or question.
Romualdas Ginevičius (2020) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: