Natural gas utilities have always surveyed their distribution networks for leaks. For most of the industry’s history, that meant walking crews with handheld instruments — effective but slow and labor-intensive at scale.
AMLD changes that equation. Vehicles equipped with high-sensitivity methane sensors, typically laser-based and operating at PPB detection thresholds, can continuously scan for methane while driving survey routes. A single vehicle can cover more network miles per day than a walking crew and flag leak indications that warrant follow-up investigation.
But the bigger shift is programmatic. AMLD has evolved from a pilot technology into full program infrastructure. For many utilities, it now sits within field investigation workflows, work order systems, repair tracking, and regulatory conversations.
That changes the evaluation question. Utilities are not only buying survey miles or sensor sensitivity. They are buying confidence that the data is actionable, useful for operations, and defensible when challenged by leadership, auditors, or regulators.
The question has moved from “should we use AMLD?” to “which AMLD program gives us the most confidence, and at what true cost?”
Why Price Per Mile Gets You Halfway There
The default criteria for mobile leak detection are contract price per mile and geographic coverage. Both matter, but neither tells you how the program performs once a vendor’s vehicle is driving on your network.
Price per mile tells you what you pay for the survey. It does not tell you what you pay for the investigation that follows. Coverage tells you how much of your network gets surveyed. It says little about the quality of the data returned.
If a utility bases vendor selection on those two numbers alone, the decision may be easy to justify from a procurement standpoint, but it is incomplete. The missing line item is what happens after the survey: how many indications require follow-up, how many lead to field investigations, how many are real leaks, and how much internal work is needed to document closure.
The better measure is true program cost — contract price per mile, cost per indication investigated, labor associated with dead-end dispatches, and the internal burden of converting detection data into repair decisions and defensible records.
How to Evaluate Advanced Mobile Leak Detection for Your Utility
A stronger AMLD evaluation examines each capability through three lenses: true cost, operational confidence, and defensibility. The question is not whether a system looks strong on paper — it is whether it produces data your teams can act on and explain.
Sensitivity floor and detection threshold
The most basic question is what the system can detect. Mobile methane detection vehicles vary in their minimum detection threshold — the smallest leak concentration they can reliably identify.
A higher detection floor will miss small leaks that a more sensitive instrument would flag. Missed leaks don’t disappear. They surface later, often as larger operational, safety, or regulatory problems.
Sensitivity also supports confidence in the survey record. A program that generates more signals but cannot help teams prioritize them may increase the downstream cost of investigation.
Ethane-to-methane ratio discrimination
Not every methane signal on a distribution network comes from your pipeline. Landfill offgassing, sewer gas, and biogenic sources generate indications that have nothing to do with your infrastructure.
Without ethane-to-methane ratio discrimination, field crews are dispatched to investigate signals the system cannot distinguish from pipeline gas. That drives up investigation labor, reduces crew productivity, and weakens the utility’s basis for explaining why certain indications were prioritized or screened out.
False positive rate and what it actually costs
False positives are one of the biggest drivers of overall program cost and among the easiest variables to underweight in an AMLD procurement process.
Every false positive results in a field crew dispatch. Each investigation can take anywhere from 20 minutes to over an hour. Across hundreds or thousands of network miles per year, the labor cost of chasing false indications becomes real spend on top of the survey contract.
The right framing is cost per mile plus cost per indication investigated. A vehicle that produces fewer dead-end dispatches has a lower true program cost, even if its contract price is higher. Every crew sent to a real leak is money well spent. Every crew sent to a biogenic signal is a cost with no return. We’ll go deeper into how to build that calculation in a follow-up post.
Workflow integration
Detection data stored in a vendor’s portal is only half useful. The bigger question is whether that data flows into your existing work order system, reporting platform, and repair tracking process.
The utilities running the most defensible LDAR programs aren’t just detecting leaks well — they’re closing the loop from indication to repair to documented closure in a traceable, auditable record. That requires the detection system to connect with the rest of the program.
Before signing a contract, ask: how does indication data get into our work order system, and what does that handoff look like?
Data defensibility and regulatory confidence
As AMLD becomes part of standard program infrastructure, data defensibility becomes a core buying criterion. A utility should be able to explain not only what the system detected but also why the program acted on certain indications, how investigations were documented, and what evidence supports closure.
That matters because the audience for AMLD data is expanding. Field teams need clear guidance. Operations leaders need confidence that crews are being used efficiently. Regulators need to see that leak detection programs are systematic, traceable, and tied to action.
A vendor that helps utilities document the full chain of evidence adds value beyond the survey itself.
The Question to Answer Before You Sign
Before finalizing any AMLD contract, a program manager should be able to answer this: “Can I explain to my VP — and if necessary, to our regulator — why this system, at this cost, produces a more defensible program than the alternative?”
If the answer is “it was the lowest price per mile” or “we’ve always used them,” that’s a procurement justification, not a performance one.
The better answer balances cost, defensibility, and operational confidence. The right AMLD program should help a utility cover the network efficiently, reduce unnecessary fieldwork, prioritize actionable indications, integrate with existing workflows, and produce a record that withstands scrutiny.
If you’re working through your AMLD evaluation and want to understand how Project Canary approaches these criteria in practice, reach out to our team.