(635b) Data Analysis and Predictive Modeling for Wastewater Asset Management

Water and wastewater distribution networks (WWDN) are the most valuable asset of a utility company and serves as a critical component of the urban infrastructure (Rahbaralam et al. 2020). The population growth and economic development have recently led to WWDN expansion, followed by the requirement of more sustainable and reliable performance of current water infrastructure (Bakri et al. 2015). In recent years, the American Society of Civil Engineers (ASCE) gave the water and wastewater infrastructure categories a D and D+ rating respectively, indicating a substandard system (ASCE 2017). Many factors contribute to these scores, the most important being asset deterioration and aging WWDNs (Mazumder et al. 2018). The infrastructure systems often experience a myriad of environmental conditions which have a negative impact on the integrity of the system (Mazumder et al. 2018). Besides, the last major investment into the water distribution network was done in the 1960s and 1970s, meaning that most systems are at least 50 years old (Malm et al. 2013). Most pipelines are built with a lifespan of about 80 to 100 years; therefore, many networks have used up over half of their expected lifespan without accounting for deterioration. As a result, many utility companies have been experiencing asset failures and malfunctions more frequently leading to large financial expenditures and negative environmental impacts. The current solution is to use a subjective and reactionary asset management approach, where a utility authority waits until a failure occurs to solve the malfunction (Baah et al. 2015; Fenner 2000). Figure 1 displays the overall process to developing an asset management plan as defined by the United States Environmental Protection Agency (US EPA) (Allbee and Rose 2017). The downfall of this process is the bias that is generated in the step “Assess Performance Failure Modes”. This bias leads to large amounts of uncertainty within the subjective asset management plan and can lead to capital losses and undesired expenses when an unforeseen event occurs. To that end, if we can find an objective-based predictive approach to strengthen this step in the asset management plan, then the negative economic and environmental impacts can be minimized.

Methodology

The asset management problem is solved using machine learning classification algorithms and Bayesian inference methods to generate a predictive deterioration model for each asset. A predictive model which can forecast the pipeline and manhole failures plays a vital role in the strategic rehabilitation planning of WWDN infrastructures (Rahbaralam et al. 2020). Therefore, these different analysis methods are employed to assess the accuracy of the predictive model. The objective is to minimize any potential failures, environmental hazards, and financial expenditures while providing a list of sound preventative measures as well as corrective actions by analyzing the historical data provided by the partner utility company. The analysis consists of evaluating the existing asset property data to develop correlations for risk factors and failure impacts. Using programming platforms such as Python and MATLAB, each asset is simultaneously analyzed, compared, and ranked. By comparing both analyses, the best method can be used in managing and predicting asset failures.

Results & Summary

The WWDN asset dataset provided by the partner company contains information about numerous variables such as the pipe diameter, pipe material type, pipe segment length, expected remaining life of pipe, and the type of flow within the pipe. After preprocessing the data, the different analysis methods are implemented to generate the best predictive deterioration model based on pipeline properties. By coupling the best analysis with accurate historical data, the model generates a failure score for each asset. This failure score allows for the prediction of asset malfunctions, allowing for each asset to be ranked on the basis of potential failures. The score is then investigated for each asset to identify the pipeline property that is the major contributor (bottleneck). Maintenance and corrective actions are then devised to bring the score to an acceptable level, which in turn minimizes the probability of asset malfunctions. The corrective actions then are compiled into a comprehensive maintenance schedule to provide an objective-based maintenance schedule. Thus, this objective-based predictive approach to asset management can minimize major economic expenditures as well as many negative environmental effects.

Acknowledgments

We thank the Atlantic County Utilities Authority, NJ for supporting and funding the research for this project. We also thank the Department of Chemical Engineering at Rowan University for their continued assistance with acquiring the tools required for this research.

References:

Allbee, Steve, and Duncan Rose. 2017. “Fundamentals of Asset Management Session 0 - Exectutive Overview.” US EPA. https://www.epa.gov/sustainable-water-infrastructure/asset-management-wo....

ASCE. 2017. “2017 Infrastructure Report Card.” ASCE. https://www.infrastructurereportcard.org/cat-item/wastewater/.

Baah, Kelly, Brajesh Dubey, Richard Harvey, and Edward McBean. 2015. “A Risk-Based Approach to Sanitary Sewer Pipe Asset Management.” Science of The Total Environment 505 (February): 1011–17. https://doi.org/10.1016/j.scitotenv.2014.10.040.

Bakri, B., Y. Arai, T. Inakazu, A. Koizumi, H. Yoda, and S. Pallu. 2015. “Selection and Concentration of Pipeline Mains for Rehabilitation and Expansion of Water Distribution Network.” Procedia Environmental Sciences, The 5th Sustainable Future for Human Security (SustaiN 2014), 28 (January): 732–42. https://doi.org/10.1016/j.proenv.2015.07.086.

Fenner, R. A. 2000. “Approaches to Sewer Maintenance: A Review.” Urban Water, Sewer Systems and Processes, 2 (4): 343–56. https://doi.org/10.1016/S1462-0758(00)00065-0.

Malm, A., G. Svensson, H. Bäckman, and Gregory M. Morrison. 2013. “Prediction of Water and Wastewater Networks Rehabilitation Based Current Age and Material Distribution.” Water Supply 13 (2): 227–37. https://doi.org/10.2166/ws.2013.011.

Mazumder, Ram K., Abdullahi M. Salman, Yue Li, and Xiong Yu. 2018. “Performance Evaluation of Water Distribution Systems and Asset Management.” ASCE Library 24 (3): 24.

Rahbaralam, Maryam, David Modesto, Jaume Cardús, Amir Abdollahi, and Fernando M. Cucchietti. 2020. “Predictive Analytics for Water Asset Management: Machine Learning and Survival Analysis.” ArXiv:2007.03744 [Cs, Eess, Stat], July. http://arxiv.org/abs/2007.03744.