Gas Leak Survey and Repair Software: Proving Every Main Was Surveyed on Its Required Cycle

Why survey coverage is harder to prove than to perform

A gas operations director gets an inspection notice. The question is straightforward: show that every main and service in this district was surveyed within its required cycle. The crews did the work. The proof is a stack of completed route sheets, a set of instrument logs on a laptop, a survey map marked up in the field, and a spreadsheet where someone recorded which routes were completed and when. What the inspector wants is coverage by pipe segment, and route sheets are organised by street.

Those are not the same thing. A route walked down a street covers the main under it and, if the surveyor did the job, the services off it. But the record of what a route physically covers was drawn when the route was created, and since then there have been main replacements, new services, an extension into a subdivision and two abandonments. The route sheet still says the same thing. The gap between the route and the pipe is where uncovered footage hides, and nobody discovers it until an inspector or an incident asks.

The tooling around this is usually the utility's GIS holding the pipe with material and vintage, a survey scheduling spreadsheet, instrument logs in a manufacturer format, a work management system holding repair orders, and a separate reporting exercise once a year for regulatory submissions. Every piece works. Nothing joins survey coverage to pipe segment, and that join is the entire compliance question.

Problem 1: detection technology found the leak, but the record is the deliverable

The detection side of this industry has improved dramatically. Picarro mobile analysis, ABB Ability MobileGuard and Bridger Photonics aerial survey all find leaks that a walking survey with an older instrument would miss, and utilities that adopt them are doing the right thing. What none of them is, and none of them claims to be, is the compliance record. They produce indications with locations and concentrations. Turning an indication into a graded leak on a specific pipe segment, with a repair clock and a closure record, is the utility's problem.

SENSIT and Heath build the instruments crews carry, and Heath's OPTIMAIN is the closest thing in the market to a survey management system, though it is naturally oriented around Heath's own instruments and survey services. If your entire program runs through one provider that works. Most utilities run a mix: walking surveys with one instrument, mobile survey under contract with another provider, aerial on a subset of the system, and their own crews on the rest.

A custom build makes the leak the object and the detection method an attribute of it. An indication from any source becomes a candidate, gets investigated and graded in the field by a qualified employee, and attaches to the pipe segment in your GIS with its material, vintage and pressure. Then the whole downstream process, repair clock, re-check schedule, replacement program input and emissions reporting, runs off one record regardless of which technology found it.

Problem 2: grading is a judgement and consistency is the compliance risk

Leak grades drive everything: whether a crew stays on site, how fast the repair must happen, and how often the leak is re-checked while it waits. The grading definitions come from state code and the widely used industry classification guidance, and they turn on readings, location relative to structures, and migration potential. Two competent surveyors can grade the same reading differently, and over a year that inconsistency becomes a pattern an auditor can see.

The usual causes are mundane. A paper form does not prompt for the observations the grade depends on. A surveyor is under time pressure at the end of a route. Nobody reviews grades for consistency because the grades live on paper in a district office. And when a leak is regraded on re-check, the history of that decision is a note rather than a record.

A custom build puts the grading logic into the field capture. The surveyor records the readings and observations the definition depends on, the application proposes a grade from those inputs, and the surveyor confirms or overrides with a reason. The override is the valuable data: a pattern of overrides in one direction from one crew is a training conversation, and it is invisible on paper. Regrades on re-check become versioned events with the reading that justified them.

Problem 3: repair clocks run per grade and nobody sees them together

A Grade 1 leak requires immediate and continuous action until the hazard is eliminated. Lower grades carry longer clocks and scheduled re-checks, with state variations that matter. So a district is simultaneously running dozens of clocks of different lengths, each attached to a leak that may be waiting on a crew, a permit, a paving restriction or a coordination with another utility.

Those clocks live in different places. The immediate ones live in the dispatcher's attention. The longer ones live in a re-check spreadsheet. The repair work order lives in the work management system, where it looks like any other job in a queue rather than a job with a regulatory deadline attached. Leaks get repaired late not because anyone decided to, but because the deadline was not visible at the moment the work was scheduled.

A custom build makes every open leak a clock with a deadline, a grade, an owner and a location, and pushes that into the scheduling decision rather than reporting on it afterwards. The useful view for a supervisor is not a list of leaks, it is the next fourteen days sorted by deadline and grade with crew availability alongside. Re-checks generate automatically and, when a re-check produces a worse reading, the regrade escalates the clock without anyone having to notice.

What a custom leak survey build has to include

• Survey coverage recorded against GIS pipe segments rather than against route definitions, so coverage gaps from replacements, extensions and abandonments surface automatically.
• Offline field capture for survey and leak investigation, because the work happens in places without reliable signal and a tool that needs connectivity will be abandoned.
• Grading logic in the field application, with proposed grade from recorded observations and a reasoned override captured as data.
• Instrument log ingestion from the detection technologies in use, with indications from mobile and aerial providers entering the same candidate queue as walking survey finds.
• Repair clocks per grade with state-specific rules, visible in scheduling rather than only in reporting.
• Automatic re-check generation with regrade handling that escalates clocks when readings worsen.
• Leak history joined to pipe material, vintage and pressure, feeding replacement program prioritisation and emissions reporting.

What it costs and how long it takes

From the utility and field operations work Digital Heroes has delivered, the pattern is this. A first release covering GIS-based coverage proof, offline field capture with grading logic, repair clocks and instrument log ingestion runs $70,000 to $140,000 and ships in 12 to 16 weeks. A full platform adding advanced detection data fusion, unaccounted for gas and methane reporting, integrated repair work management and replacement prioritisation runs $180,000 to $420,000 phased across 6 to 12 months.

Cost drivers specific to gas distribution: GIS quality, because coverage proof is only as good as the pipe data underneath it, and utilities carrying unmapped services or approximate service line locations need that addressed first. Multiple state jurisdictions, since survey cycles, grading definitions and repair timelines differ and each is its own rule model. Detection vendor diversity, because every provider delivers indications in its own structure on its own cadence. And work management integration, which is straightforward technically and slow organisationally because it crosses departments.

What keeps cost down: starting with one operating district, ingesting the one detection source that produces the most volume, and leaving replacement prioritisation to phase two once you have a year of clean leak history to prioritise against.

Build versus buy, and when buying is right

Buy or stay manual if you are a small municipal system under a few hundred miles with one survey crew, one state jurisdiction and a manageable leak backlog. A disciplined spreadsheet and a good supervisor genuinely cover it, and detection equipment is a better use of the budget.

Build when two or more of these are true. You cannot demonstrate coverage by pipe segment without someone reconciling route sheets against the GIS by hand. You use two or more detection technologies and their outputs sit in separate systems. You operate across more than one state and therefore more than one set of survey and repair rules. Repair deadlines are not visible to the people scheduling crews. You have grading inconsistency between crews or districts that nobody can quantify. Or your replacement program is prioritised by pipe vintage alone because leak history is not usable at segment level.

The tipping point is the moment your compliance answer requires reconstruction. Survey work that was performed correctly and cannot be proven cleanly is a finding, and the reconstruction effort is a recurring tax that grows with every system change.

How to choose a developer for leak survey software

Ask how they would prove coverage on a street where a main was replaced mid-year and three services were added. The right answer is coverage recorded against segments with effective dates, so the replacement creates newly uncovered footage automatically. A developer who answers in terms of routes has not understood why the current process fails.

Ask what the field application does with no signal for a full shift. Offline capture with reliable sync is table stakes here, and so is showing prior leak history for the location at the point of investigation, because a surveyor who knows this address had a Grade 2 last year investigates differently.

Ask how grading will be captured. You want the observations that the definition depends on recorded as fields, a proposed grade, and overrides captured with reasons. A free text grade field is how consistency problems become invisible.

Ask which detection providers they have ingested from and settle code and infrastructure ownership in writing before kickoff, which at Digital Heroes means the client owns the repository from the first commit. A good first step inside your own organisation: take one district, export the pipe segments from GIS, and try to prove last cycle's coverage against them. However long that takes is the size of the problem you are buying out of.