Journal of Air Transport Management, v. 58, pp. 100-112. Abstract: This paper assesses airport efficiency levels in Senegal under a stochastic environment where the satisficing concept for different performance thresholds is applied. A two-stage satisficing DEA-Support Vector Machine approach is used here to compute the impacts of cost structure on these thresholds. In the first stage, within the ambit of the satisficing DEA model, the probabilities of achieving a minimal performance threshold are computed in a stochastic fashion. In the second stage, Support Vector Machine regression is used to discriminate between high/low efficiency groups within a given performance threshold. This methodology was sufficiently robust to handle small samples. The results reveal that the cost of capital and the cost of labor are the cost structure variables that have the greatest impact on efficiency levels, besides cargo operations.