Dynamic Failure Risk Assessment of Wastewater Treatment and Reclamation Plant: An Industrial Case Study

Abstract

Due to the growing scarcity of water resources, wastewater reuse has become one of the most effective solutions for industrial consumption. However, various factors can detrimentally affect the performance of a wastewater treatment plant (WWTP), which is considered a risk of not fulfilling the effluent requirements. Thus, to ensure the quality of treated wastewater, it is essential to analyze system failure causes and their potential outcomes and mitigation measures through a systematic dynamic risk assessment approach. This work shows how a dynamic Bayesian network (DBN) can be effectively used in this context. Like the conventional Bayesian network (BN), the DBN can capture complex interactions between failure contributory factors. Additionally, it can forecast the upcoming failure likelihood using a prediction inference. This proposed methodology was applied to a WWTP of the Moorchekhort Industrial Complex (MIC), located in the center of Iran. A total of 15 years’ time frame (2016–2030) has been considered in this work. The first six years’ data have been used to develop the DBN model and to identify the crucial risk factors that are further used to reduce the risk in the remaining nine years. The risk increased from 21% to 42% in 2016–2021. Applying the proposed risk mitigation measures can decrease the failure risk from 33% to 9% in 2022–2030. The proposed model showed the capability of the DBN in risk management of a WWTP system which can help WWTPs’ managers and operators achieve better performance for higher reclaimed water quality.