The present work analyses the existence of congestion in a set of wastewater treatment plants (WWTPs) for the first time. The study sample included 47 WWTPs managed by the same operator and located in the Valencia region (eastern Spain). The method of analysis consisted in applying a new statistical model to identify and correct congestion situations based on data envelopment analysis (DEA), which takes into account undesirable outputs and multiple projections. The results showed a strong concentration of congested WWTPs (48.94-57.45%). Furthermore, the necessary corrections to avoid the congestion statuses entailed an average input reduction between 57.03 and 74.13% and an average increase in desirable outputs between 46.08% and 50.08%. In addition, only plant size had a significant relationship with the correction levels. The remaining contextual variables (age, overcapacity and type of water and sludge line) failed to show any significant impact on those corrections. To conclude, policy recommendations to improve the management of WWTPs are put forward, notably improving the management of smaller plants, which concentrate the largest percentage of correction.
Keywords: Congestion; Data envelopment analysis (DEA); Multiple projections; Undesirable outputs; Wastewater treatment plants.