Alternate: Sewer Rehabilitation: Pioneering New Planning Tools

Efficiency in planning and executing projects is becoming more and more critical as collection systems age and rehabilitation costs rise. This presentation looks at how machine learning and GIS software helped accelerate both identifying a high priority project area and plan production on a recent sanitary sewer rehabilitation project in Raytown, Missouri. Traditional methods of prioritizing areas for capital improvement projects are time consuming involving city-wide inspections, months of flow monitoring, and complex modelling. These all generate useful information, but communities may not have the budget or time to collect and analyze the data. The City of Raytown had a desire for a quicker start in identifying areas for sewer rehabilitation in order to provide their citizens with exceptional, cost-effective service. The City engaged the services of NEER to identify at risk project areas using the company's machine learning software. The two fundamental building blocks for defining asset management (risk) are Likelihood of Failure (LoF) and Consequence of Failure (CoF). LoF describes the chance of an asset failure occurring and can be represented as a numerical ranking ranging from unlikely (1) to highly likely (5). CoF measures the severity of the impacts if an asset were to fail and can also be expressed numerically using the same ranking system as LoF. The overall risk or Business Risk Exposure (BRE) of an asset is a relative measure formulated as a weighted combination of the LoF and CoF scores, that ranges from 1 through 5. NEER software calculated LoF, CoF, and BRE for each asset scoring on a scale from 1 (low risk) to 5 (very high risk). NEER software inputs include: age of infrastructure, pipe material, surface type, maintenance history, tree cover which is indicative of potential root problems, amongst other factors to identify at risk areas. After identifying the project area, Lamp Rynearson was brought on board to use more traditional field inspection and evaluation to determine required rehabilitation and complete design plans. Ace Pipe Cleaning cleaned and CCTV the sanitary sewers, delivering videos and associated PACP database. Lamp Rynearson engineers quickly reviewed the database and associated poor condition videos. Manhole inspections and field checks of open cut sites were also performed by engineers. Both structural defects (cracks, fractures, holes, collapsed pipe) and maintenance defects (roots, debris, corrosion, deposits attached, infiltration) were examined, and cost-effective repairs determined. Defective service lateral connections and the extent of repairs were also evaluated. Recommendations maps were quickly prepared to visualize locations of identified problems with ArcGIS Pro by mapping the PACP database. Cost of repairs were aggregated with associated LoF, calculated in accordance with PACP Appendix D and normalized to a scale of 1-5. Based on the level of cost-effectiveness to rehabilitate each asset, the final asset selection for the project included all line segments with at least one grade 4 PACP defect. After selection of final assets to include in the construction project by the City, final plans were produced with the field inspection and GIS data. The repair types include rehabilitation of defective service connections with CIPP liner or open cut, sewer main total line segment replacement, open cut point repairs, CIPP mainline, new manholes, and manhole rehabilitation. Using existing data, NEER's machine learning process can quickly generate priority areas for Capital Improvement Planning. This project validated NEER's output showing a good correlation to the PACP field inspection data. In addition, more efficient plan production processes were achieved as Lamp Rynearson developed a semi-automated process utilizing ArcGIS Pro and available GIS data resulting in an accelerated delivery schedule for the project.