Exploring Congestion in Fuzzy DEA by Solving One Model; Case Study: Hospitals in Tehran

In recent decades, the number of private hospitals has increased in many countries, leading to an oversupply of drugs in the country's medical industry. This leads to hospital congestion and a decline in hospital profits, efficiency, and performance. One of the appropriate methods to detect congestion is Data Envelopment Analysis (DEA). In this study, we identified congested Tehran hospitals with fuzzy and vague data. For this purpose, we developed for the first time the methods proposed by Shadab et al. (Shadab, M., Saati, S., Farzipoor Saen, R., Mostafaee. (2021). Detecting congestion in DEA by solving one model. Operations Research and Decisions. 77–97.) and presented a novel fuzzy nonlinear DEA model for identifying congested hospitals as Decision Making Units (DMUs). We improved the method suggested by Saati et al. (Saati et al. in Fuzzy Optimization Decision making. 1:255–267, 2002) and transformed our proposed possibilistic model into a crisp nonlinear DEA model. The type of congestion (weak and strong) and its magnitude are also determined. Finally, the proposed model is used in an empirical example to identify 15 congested hospitals in Tehran (the capital of Iran) and measure the role of congestion and its magnitude to improve their performance. The result showed that congestion occurred in 39.9% of 15 hospitals.