A Data-Driven Approach to Determine If New Water Mains Construction Can Cause Neighboring Old Water Mains to Break

The water main replacement program at WSSC Water has been on-going for several years and it now contains data obtained from various sources, including construction practices and condition assessment, that can provide more insights into the root cause of water main breaks. Generally, several factors can contribute toward water main breaks such as internal and external corrosion, load stresses, pressure transients, material properties (including vintage), and construction practices. WSSC Water already includes many of those factors in its water main replacement recommendations process. Recently, based on anecdotal field evidence, WSSC Water sought to determine if there was a correlation between breaks occurring on old cast iron mains that were connected to the newly replaced ductile iron mains. This paper discusses process development and data analysis techniques used in determining any systemic issues of water main failure due to tying new pipes to old water mains. The analysis showed a small count of breaks associated with the tie-in pipes but overall did not find any substantive correlation. Out of the 2,367 counts of the tie-in pipes, there were only 39 counts of tie-in pipes with water main breaks. Although no significant correlation could be established, there are few insights that are recommended for future analysis such as data refinement or determining the effective length of existing water mains impacted by tie-in pipes. The data analysis techniques established here could benefit future similar hypotheses analyzed for validity using similar principles.