A Decision Analytical Perspective on Public
Mats Danielson, Love Ekenberg, Mattias Göthe and Aron Larsson
If procurement processes are to be taken seriously, purchase managers need decision support tools beyond those that only ascertain that the formal requirements are met. This chapter demonstrates some fundamental flaws with common models used in procurement situations, flaws that are so serious that the evaluations of tenders often become meaningless and may lead to large and costly miscalculations.
We demonstrate how the equability of the tender evaluations can be significantly improved through the use of multi-criteria decision analysis with numerically imprecise input information. Due to this, the computational part of the evaluation step becomes more complex, and algorithms targeted for decision evaluation with imprecise data are used.
We therefore present a decision tool, DecideIT, implementing such algorithms that can be used as an instrument for a more meaningful procurement process. Of importance is to allow for a more realistic degree of precision in the valuation and ranking of tenders under each evaluation criterion, as well as the associated weighting of the criteria, since the criteria are often of a more qualitative nature. Through this, both quantitative and qualitative statements could be easily managed within the same framework and without the need to introduce ad-hoc and often arbitrary conversion formulas supposed to capture the trade-off between criteria.
Read the paper: