Generating OWA Weights from Individual Assessments

In this contribution we propose a method for generating OWA weighting vectors from the individual assessments on a set of alternatives in such a way that these weights minimize the disagreement among individual assessments and the outcome provided by the OWA operator. For measuring that disagreement we have aggregated distances between individual and collective assessments by using a metric and an aggregation function. We have paid attention to Manhattan and Chebyshev metrics and arithmetic mean and maximum as aggregation functions. In this setting, we have proven that medians and the mid-range are the solutions for some cases. When a general solution is not available, we have provided some mathematical programs for solving the problem.