Cryogenic Storage and LNG KPIs: Benchmarks for Boil-Off, Vacuum, Transfer Loss and Uptime
The seven KPIs that run a cryogenic storage operation, how to measure each one, the levers ranked by payback, and which figures have a cited source.
A cryogenic storage operation can be run from seven numbers: boil-off rate, transfer loss, vacuum annulus pressure trend, specific pumping energy, vaporizer utilization, weld first-pass acceptance during construction, and system availability. Each has a measurable definition, a realistic range and known levers. Use the ranges below as reference points for spotting a gap worth investigating, not as targets.
Every figure shows its basis. A superscript number links to a public source in the Sources list, and the text says when a cited figure is a design range, a manufacturer rating or a survey average. A figure marked no cited source is kept exactly as originally published because no public source stating it was found, so treat it as illustrative.
At a glance
Cryogenic storage and LNG KPI benchmarks at a glance
KPI
Typical
Strong
Basis
Boil-off, large full-containment LNG tanks
0.05% per day
0.02 to 0.03% per day (world class)
Typical: Dobrota et al. design figure1; world class no cited source
Words in parentheses keep each figure's original framing, and Not stated means no figure was given for that column. The cited boil-off figure is a design value for terminal storage tanks, and the weld repair rates are survey averages for whole industry sectors; their sections give each source's scope.
Method: How the figures in this guide are sourced
A figure with a superscript number is stated by the cited source, listed under Sources with its publisher, location in the document and access date. Several cited figures are design ranges, manufacturer ratings for new equipment or dated survey averages, and the text says so where they appear. A figure marked no cited source is kept exactly as originally published because no public source stating it was found. It is illustrative, not measured industry data. Worked examples are arithmetic, not benchmarks.
Boil-off rate
Boil-off rate (BOR) targets depend entirely on scale. For large full-containment LNG tanks, world class is 0.02 to 0.03 percent per day no cited source, typical is 0.05, the level tanks are typically designed to stay below1, and anything above 0.08 says the insulation or base heating system has a problem no cited source.
The published figures are design ranges, not performance tiers, and none uses a world-class label. Dobrota and colleagues (2013, citing BP and the International Gas Union) report that LNG storage tanks are typically designed so boil-off stays below 0.05 percent of tank content per day, although it can vary from 0.02 to 0.1 percent1.
A 2019 National Energy Technology Laboratory (NETL) study models terminal storage boil-off at 0.02 to 0.1 percent per day4, and Argonne National Laboratory's GREET 2023 model assumes 0.1 percent per day at the receiving terminal5. Readings up to 0.1 percent per day are therefore within the range these sources describe1, 4.
Transport and smaller storage equipment
Transport equipment and smaller vessels boil off a larger share of their contents each day than a terminal tank. Manufacturers publish ratings for new equipment, and LNG carriers are built to a design limit.
Published boil-off ratings for transport and smaller storage equipment
MicroBulk ratings are on a gross-capacity basis and generally fall as tank size rises. Chart rates its LNG ISO containers for hold times of 44 to 75 days on 20 ft models and 53 to 65 days on 40 ft models6. The same DOE/NETL study notes that actual carrier boil-off is usually lower than the design limit, as low as 0.10 percent per day8.
Measure BOR directly: isolate the tank, log level and pressure for 48 to 72 hours no cited source, and normalize for fill level.
For example, at identical heat leak a 30 percent full tank shows a higher percentage BOR than a 90 percent full one, because the same boil-off is divided by a smaller liquid inventory.
Chart's field test for its bulk tanks controls for the same effect. The tank stabilizes for four days with no withdrawal, must be at least 2/3 full, and its vent flow is then metered for 30 minutes and compared with the rated value9.
How to measure
Isolate the tank, with no withdrawal, before you start logging.
Log level and pressure over a fixed window and convert the loss to a share of contents per day.
Normalize for fill level, or always test at a consistent fill.
What moves it
Insulation condition, which the vacuum trend predicts.
Vacuum annulus pressure is the leading indicator that predicts boil-off drift 6 to 18 months early no cited source. World class vessels hold below 10 microns cold and rise less than 2 microns per month no cited source; typical fleets alarm at 50 to 100 microns no cited source.
Typical fleets re-evacuate every 2 to 3 years no cited source, while a tight vessel goes 5 to 10 years between pump downs no cited source. A vessel that needs re-evacuation twice a year has a leak, not a maintenance schedule no cited source.
Published manufacturer limits give some reference points. Chart tells installers to consult the factory if the first vacuum reading on a new VS or HS bulk tank is above 20 microns, or if readings rise steadily10. Chart also warrants that its 01 Series bulk vessels will maintain all vacuum and performance standards for five years11.
A 1987 Fermilab purchase specification for a 20,000 gallon liquid nitrogen tank set evaporation at no more than 0.30 percent of capacity per day at 50 millitorr (50 microns) of vacuum12. Its warranty clause capped the vacuum at 200 microns cold, 12 months after start-up12.
How to measure
Read annulus vacuum monthly, with the tank cold and at similar fill.
Plot the trend, not just the latest reading.
Record the date of every re-evacuation for each vessel.
What moves it
Finding and fixing vacuum leaks.
Re-evacuation before the trend reaches the insulation performance knee.
Transfer loss per movement is the second loss KPI. World class terminals with vapor return and cold lines hold 0.10 to 0.20 percent of transferred mass no cited source, typical operations run 0.3 to 0.8 percent no cited source, and a warm-line offload without vapor recovery can exceed 2 percent no cited source.
The gap between typical and world class performance on boil-off and transfer loss together is commonly worth 1 to 2 percent of annual throughput no cited source. On a mid-size LNG terminal, that is millions of dollars of product per year.
Measure it as custody meter in minus tank gain, or by before-and-after level surveys on both vessels. Log it per transfer rather than per month, so a bad hose or a lazy precool shows up immediately.
The LNG Transfer Loss calculator gives the achievable floor for your line size and length. If actuals sit more than 0.2 percentage points above that floor, the gap is procedure, not physics no cited source.
How to measure
Subtract tank gain from the custody meter quantity for every transfer.
Cross-check with before-and-after level surveys on both vessels.
Log each transfer separately rather than as a monthly total.
What moves it
Vapor return on transfers.
Cold lines and a disciplined precool.
Hose condition.
Pumping and vaporizer energy
Specific energy KPIs keep the utility bill honest. In-tank and transfer pumps should move product at 1 to 2 kWh per tonne in world class service no cited source; 3 to 5 kWh per tonne usually means throttled discharge valves or excessive recirculation no cited source, both fixable with variable frequency drives (VFDs) or impeller trims.
On regasification, ambient air vaporizers cost nearly nothing to run but need 60 to 85 percent utilization to justify their footprint no cited source. Units running chronically below 50 percent are consolidation candidates no cited source.
Submerged combustion vaporizers burn part of the product as fuel. A 2005 DOE/NETL-funded LNG terminal study states that this fuel gas typically represents 1.25 to 1.5 percent of the total LNG vaporized2.
How to measure
Divide pump kWh by tonnes moved, every month.
Compare actuals with the ideal case and track the ratio.
Log vaporizer run hours against available hours.
What moves it
VFDs or impeller trims in place of throttled discharge valves.
Construction-phase KPIs decide whether the asset starts life healthy. On weld first-pass acceptance for primary containment, world class crews hold 98 percent no cited source, and below 90 the schedule is already gone no cited source.
TWI's 2011 industry survey found that average repair rates for the oil and gas and power sectors ranged from 1 to 3 percent, with peaks up to 25 percent in specific locations3. By product, counting repaired welds over total welds, pressure vessels averaged 2.0 percent and cryogenic storage tanks 0.5 percent, the latter based on one entry3.
Read as a share of welds, a repair rate of 1 to 3 percent means 97 to 99 percent of welds needed no repair.
Erection productivity of 25 to 40 direct hours per tonne of erected steel is a strong field benchmark no cited source; track it weekly against your own baseline. Plan inspection so nondestructive examination (NDE) stays within one shift of welding no cited source, so defects surface while the crew that made them is still on the joint.
At commissioning, hold cooldown ramps within plan, with zero excursions past the 50 K/h stress limit no cited source.
How to measure
Divide welds accepted at first inspection by welds inspected on primary containment.
Track direct hours per tonne of erected steel every week.
Log the lag between welding and NDE, in shifts.
What moves it
NDE close behind welding, so the crew that made a defect fixes it.
Cooldown ramps held to plan at commissioning.
Availability and relief valve health
Round out the board with availability and protection health. Storage and send-out availability should sit at 98 to 99.5 percent excluding planned outages no cited source.
Relief valve pop-test pass rate belongs above 95 percent on a 1 to 5 year cycle no cited source, with every failed test triggering a resize review. Chart's inspection table for its VS and HS bulk tanks calls for verifying relief valve settings every 2 years, with shorter intervals in extreme climates13.
How to measure
Divide available hours by period hours, excluding planned outages.
Record a pass or fail for every relief valve pop test.
What moves it
A resize review after every failed relief valve test.
Restore vapor return on transfers, which pays back in months no cited source.
Retrofit pump VFDs, which pays back in 12 to 24 months no cited source.
Upgrade insulation, which pays back in 2 to 5 years no cited source.
For one published reference point, a refinery pump variable speed drive retrofit described in a 2004 U.S. Department of Energy guide paid back in about 1.6 years overall14. That is a single case study, not a typical range.
A Novel Process to Use Salt Caverns to Receive Ship Borne LNG, Final Report (DOE Award DE-FC26-02NT41653). U.S. Department of Energy, National Energy Technology Laboratory (final report by Conversion Gas Imports, L.P.; McCall, Bishop, Krekel, Davis, Scherz). published August 2005. Appendix 'LNG Vaporization Report', Section III.1 'Submerged Combustion Vaporizers' (PDF p. 1123); see also HiLoad study section 4.3 (PDF p. 1023). Accessed September 30, 2026.
LNG ISO Intermodal Containers (product literature PN 14732510). Chart Inc. (Chart Industries), manufacturer technical data. published 2013. Page 2, specification tables for 20 ft and 40 ft TVAC ISO containers, columns 'Hold Time Days (LNG)' and 'Evaporation LNG%/day'. Accessed September 30, 2026.
Perma-Cyl On-Site Storage System: MicroBulk Solutions (product literature PN 11000938). Chart Inc. (Chart Industries), manufacturer technical data. published 2025. Page 2, specification table row 'Thermal Performance (2) (NER%/Day)', first (230L) and last (7000L) model columns; footnote (2): values based on gross capacity. Accessed September 30, 2026.