Deciding which gas mains to replace, repair, or retire

US gas utility planners used to ask which leaking mains they should replace first. They ranked pipe by safety risk, worked down the list, and replaced as much as the annual budget allowed. Now they also have to ask which mains they should replace at all.

Consider a neighborhood with three aging mains. Engineers find a hazardous cast iron main beneath a busy street, a second that crews can repair and monitor, and a third where customers could switch to electric heat. The utility replaces the first and must justify why either of the other two needs new gas pipe.

Each mile of new main costs around $1 million and is built to last 30 to 50 years. If the neighborhood switches off gas sooner, the utility may retire that pipe long before customers have finished paying for it. Regulators can refuse cost recovery if staff cannot show why replacement was more prudent than repair or retirement. Safety still determines which mains cannot wait. For every other mile, prioritization has become a choice about whether to replace the pipe at all.

The annual replacement budget

Utilities cannot treat every aging mile as an eventual replacement when the budget pays for only a fraction of the backlog. State regulators often let them recover eligible work between rate cases through an infrastructure replacement surcharge on customer bills. Regulators cap annual recovery and can reject spending they consider imprudent.

Across the US, operators manage roughly 1.3 million miles of distribution main. Cast iron and bare steel account for under 3% of those miles, concentrated mainly in older Northeast and Mid-Atlantic systems. The Pipeline and Hazardous Materials Safety Administration (PHMSA) reported about 14,000 miles of cast and wrought iron. Crews focus on these materials because unprotected bare steel corrodes, while cast iron can crack under ground movement and weaken over time.

Some utilities now have less room within the cap. The Massachusetts Department of Public Utilities reduced the Gas System Enhancement Program recovery cap from 3% of prior-year revenue to 2.5% in April 2025. The department will lower it to 2% in 2026 and 1.5% in 2027. The Pennsylvania Public Utility Commission caps its Distribution System Improvement Charge at 5% of billed distribution revenue. Planners must fund the dangerous pipe first, then decide whether each remaining candidate justifies another $1 million mile of gas infrastructure.

The mains that cannot wait

Safety comes first exactly as it did under the established approach. Engineers replace a segment promptly when crews cannot control an immediate hazard through repair, regardless of what may happen to gas demand in that neighborhood.

PHMSA requires every distribution operator to maintain a Distribution Integrity Management Program (DIMP), which engineers use to estimate each segment's chance of failure and likely consequences. They consider material, age, pressure, leak history, corrosion, soil conditions, and ground movement. Engineers assign greater consequences to mains near homes, schools, and workplaces because a failure could harm more people.

A cast iron main beneath a busy street belongs in the non-deferrable group when crews repeatedly find serious leaks or surveyors classify one as Grade 1, meaning an immediate hazard. Engineers update the order as conditions deteriorate. Planners coordinate lower risk work with road resurfacing and other utility projects to avoid excavating a newly paved street, but they cannot delay an immediate safety job for those savings.

Once planners have funded that safety tier, they face the part of the decision that has changed: how long customers will need each remaining main and whether replacement is worth the commitment.

The case for another gas main

For a lower risk segment, planners now have to prove that another gas main will serve customers long enough to justify its cost. If the neighborhood moves off gas before the asset reaches the end of its 30 to 50 year book life, customers or shareholders may have to absorb the unpaid balance when the utility retires it.

Regulators in Massachusetts and New York now require utilities to examine that possibility. In December 2023, Massachusetts Order 20-80 directed local gas distribution companies to consider ways to serve customers without new gas pipe when seeking cost recovery. The department's April 2025 orders also directed utilities to give more attention to repairs, more sensitive leak detection, and alternatives to replacement.

Under its 2022 order in Case 20-G-0131, the New York Public Service Commission requires gas utilities to evaluate a no-infrastructure option and other non-pipe alternatives. For each marginal segment, planners need to explain who a failure could harm, how crews can manage the pipe, how long customers are likely to use it, and why the selected response is prudent. A network-wide worst-first list cannot provide that justification on its own.

Repair and retirement options

Repair or retirement can avoid a long-lived investment on the lower risk miles. Crews can keep some mains in service through sensitive leak surveys, targeted repairs, and closer monitoring while staff establish how long customers will need gas service. If surveyors later find a serious leak or worsening corrosion, engineers move the segment into the replacement group.

On a street where nearly every customer can heat a building another way, staff could connect buildings to a shared networked geothermal loop, help customers switch to electric heating, or depressurize the segment after customers leave the gas system. The utility can then retire the main instead of installing pipe it expects to abandon early.

Utilities have limited evidence about using these options across whole service territories. Eversource began operating a utility-owned networked geothermal pilot in Framingham, Massachusetts, in June 2024 for about 135 customers. National Grid later dropped a Lowell project after projected costs came in higher than expected. Planners need both outcomes when they estimate whether a similar project can serve a street at an acceptable cost.

The utility can retire a segment only if nearly every customer participates because one remaining gas customer may require it to maintain the pipe. Staff need to test customer interest, building suitability, cost, and delivery timing for the annual plan. Crews must also be able to manage the pipe safely while customers move to another heat source.

A defensible replacement sequence

The utility now needs a segment-level justification as well as a safety ranking. Planners first fund the mains that engineers judge too dangerous to defer. For the remaining candidates, they compare full replacement with targeted repair and monitoring or retirement after customers switch heat sources.

Staff record the pipe's condition, the selected response, the alternatives they rejected, and how long customers are expected to need the main. Engineers update that record when surveyors find a leak, a municipality changes its paving schedule, or customers advance an electrification project. A lower risk bare steel main may stay in service after a repair this year and enter the replacement group next year if crews find worsening conditions.

US utilities will continue removing dangerous cast iron and bare steel as soon as safety requires it. The operators adapting to the new regulatory and demand outlook can also show, mile by mile, which pipe they chose not to replace and why.