Failure Analysis Helps SD1 of Northern Kentucky Improve Its Continuous Sewer Assessment Program

Authors:

  • Bo Copeland - Hazen and Sawyer

This graph shows the overall condition score resulting from the first inspection of each pipe. Scores range from 1 - 100, with higher scores representing pipes in worse condition. The orange line shows the “percentile” of each score representing what percentage of inspections receive a lower score. This figure helped SD1 identify that some scores (e.g., 40) occur much more frequently than others, which allows them to investigate the reasons for this and verify that scoring is appropriate.

This figure shows the structural failure rate of pipes based on their structural condition score when last inspected prior to failure. This failure rate is found by normalizing the number of pipes with the given inspection score that failed by the total number of pipes which received that score. Note that failure rates for pipes with a prior score of 100 were low; since most of these pipes were repaired or replaced before they could cause a failure.

This figure shows pipes' average time to structural failure based on each pipe's previous structural condition score (the color shading indicates how many structural failures had previous scores in each range). The previous figure and this one help to determine the likelihood of pipes failing and will help SD1 optimize inspection return frequency for each score range.

This graph shows the average structural failure rate for each observed structural defect type, based on all inspections included in the analysis. The data for this figure were compiled by reviewing each structural failure's previous inspection, identifying the defect type that led to the failure, and normalizing by the number of times that defect type occurred in all inspections. SD1 may utilize this figure to adjust return frequencies for inspections based on individual defects observed, rather than relying solely on the pipe's score.

Sanitation District No. 1 of Northern Kentucky (SD1) is the second-largest public sewer utility in Kentucky, serving more than 290,000 residents throughout Boone, Campbell, and Kenton counties and maintaining 1,600 miles of sanitary sewer, 440 miles of storm sewer, 120 wastewater pumping stations, 15 flood pump stations, 8 package treatment plants, and 3 regional treatment plants.

During implementation of its Capacity, Management, Operation, and Maintenance (CMOM) Program, SD1 developed a Continuous Sewer Assessment Program. The goal of the program was to develop a logical approach to prioritizing inspection, cleaning, and other maintenance of the system. This included the development of a decision tree that would determine the next action on a pipe after inspection. This next action could be re-inspection or cleaning in a certain period of time or placing the pipe on the rehabilitation list.

Sewer pipe structural and maintenance condition are each assessed and scored based on CCTV inspections, and next actions for pipes are determined based on each of these scores separately. The most conservative of these dictates the next action that is actually scheduled. The triggers for re-inspection and the return frequencies were made relatively conservative at the start of the program to get a better understanding of operation and maintenance (O&M) “hotspot” areas.

Objective

The program has been effective in reducing O&M blockage-related overflows in the system. However, SD1 has been conducting a comprehensive review and analysis of the program with the goal to manage risks related to the CSAP more effectively and efficiently by utilizing historical data to better understand the historical behavior of its collection system and, thereby, to revise and improve internal CSAP processes.

This study is a comprehensive review and analysis of the CSAP that utilizes findings from analysis of the last eight years’ worth of data in order to prioritize the inspection and cleaning of gravity sewer assets based on the historical behavior of the sewer system. The goals of this analysis include:


  • Reduce re-inspection and cleaning frequencies overall, while maintaining adequate frequency of re-inspection and cleaning for pipes with highest likelihood of failure.
  • To the extent reasonably practicable, avoid preventable structural and blockage-related failures (especially overflows and building backups).
  • Based on review of historical inspection and failure data:
    • Update scoring thresholds for determining next actions,
    • Optimize frequency of re-inspection and cleaning for each range of scores,
    • Improve use of other considerations (e.g., previous scores) in determining next actions, and refine criteria used to place pipes in the rehabilitation program.

Methodology

As part of this program update, SD1 evaluated the historical sewer inspection and failure records since July 2008 to determine if re-inspection triggers could be further refined and return frequencies lengthened without significantly increasing risk of failure and to make further data-driven recommendations.

Extensive data, including pipe attributes, inspections, overflows, work requests, and work orders were exported from SD1’s computerized maintenance management software (CMMS) and analyzed with a data visualization software tool called Tableau® (Seattle, Washington), which allows large data sets to be manipulated through custom calculations and filters and displayed in intuitive graphs and charts. Separate lists of pipe structural failures (overflows, basement backups, or sinkholes caused by structural-related pipe condition) and blockage failures (overflows or basement backups caused by non-structural pipe blockages) were developed from the data. Queries were used to join and filter the data that were presented in the Tableau® visualizations, which showed changes in individual pipes’ condition scores between inspections, trends in condition scores and failures for the sewer system as a whole, and correlations between pipes’ failures and their attributes and previous condition scores. Filters were created to “drill down” into these visualizations and look at different subsets of the data being displayed.

Results
The results of this evaluation helped SD1 to better understand the historical behavior of the sanitary sewer system and will help with making changes to the CSAP that are backed up by actual data. The defect analysis will allow SD1 to refine defect scoring to ensure that individual defects or combinations of defects will result in a pipe score that accurately reflects the pipe’s condition and probability of failure. The comparison of failure rate and time to failure of the inspection score ranges will assist SD1 in updating the logic that determines next action for the various scores.

In the process of compiling the data for this analysis, SD1 identified areas for improvement in their data collection and management processes. During a review of CCTV videos, significant variations in scores were found between pipes that were actually in similar condition. SD1 plans to refine their training and limit the number of inspectors who assign defect codes to ensure consistency in future scoring. Inconsistencies were also found in the terms and phrases used by SD1’s staff in their field reports submitted after sewer overflows and trouble calls. SD1 will be updating their CMMS to add more standardized options for their field staff to help achieve consistency and collect better data for future analyses.

These improvements to CSAP processes will help SD1 better assess infrastructure condition and life cycle, cost-effectively minimize O&M blockage-related spills, and implement system renewal and rehabilitation in a way that targets dollars where they have the highest benefit.

For more information, please contact the author at bcopeland@hazenandsawyer.com.

Hear about new publications with our email newsletter

We will never share your details with anyone else.

Horizons

Newsletter Newsletter

Horizons Fall 2017 (pdf)

Horizons showcases significant water, wastewater, reuse, and stormwater projects and innovations that help our clients to achieve their goals, and can help you achieve yours. Articles are written by top engineers and process group leaders, demonstrating and explaining the beneficial application of a variety of technologies and tools.

View previous issues »