Constrained Optimization of a University Administrator's Preference Function

Universities are complex institutions with multidimensional objectives. Lack of understanding of the "production process" for teaching and research, and a lack of approved procedures for making cost allocations, make it difficult to organize formal planning systems at the level of an academic department. At the same time, planning at the top administration level has been hampered by a lack of understanding of how primary planning variables are interrelated both through financial constraints and (usually unspecified) criteria for choice. This paper reports on a field test of various approaches for obtaining and quantifying subjective estimates of preference for academic outcomes in terms of a small set of university planning variables. The resulting preference functions were combined in a mathematical programming model with constraints on budget balance and growth to yield "optimal" university configurations.