A gas-lift field almost never has all the injection gas it could use. Compression capacity, gas supply, and pipeline constraints usually mean the total lift gas available is less than the sum of what every well would happily take. Lift gas allocation is the field-level decision of how to split that limited gas among the wells to produce the most total oil. This guide explains the marginal-response idea behind good allocation, why equal or fixed splits leave oil in the ground, and how SCADA and optimization software reallocate gas as wells change.
Lift Gas Allocation in one line: Lift gas allocation is the process of distributing a limited supply of injection gas among many gas-lift wells to maximize total field oil production. The optimum gives each additional unit of gas to whichever well returns the most extra oil for it, ranking wells by their marginal oil response rather than splitting gas equally.
Each gas-lift well has a gas-lift performance curve: as injection gas increases, oil rate rises, flattens, and eventually falls when too much gas adds friction and lifts less. The slope of that curve, the extra oil you get for one more unit of gas, is the marginal response, and it changes with how much gas the well already has. A gas-starved well has a steep slope; a well near its peak has a shallow one.
The economic rule for splitting a fixed pool of gas is to equalize the marginal response across all wells. As long as one well returns more oil per unit of gas than another, moving gas from the low-response well to the high-response well raises total oil. The field is at its optimum when every well is at the same marginal oil-per-gas value, so no reshuffle can add more oil. This is why the last unit of gas should go where it earns the most, not where the well is biggest.
This is precisely why equal splits and fixed historical setpoints leave production on the table. Giving every well the same rate over-gasses some wells past their peak while starving others that would still respond strongly. Allocation done well concentrates scarce gas on the wells whose curves are still climbing steeply and pulls it back from wells that have already flattened out.
Allocation is a constrained optimization. The objective is to maximize total oil, but the total injection gas is capped by compressor throughput and supply, and each well has its own limits: a minimum rate to keep it lifting and stable, a maximum before it heads or over-gasses, and sometimes a valve or facility constraint. A good allocation respects all of these while pushing gas toward the steepest marginal responses within the cap.
The catch is that the performance curves are not fixed. Reservoir pressure declines, water cut rises, wells foul or clean up, and valves shift their injection point, so a curve that was steep last quarter may be flat today. An allocation computed once and left alone slowly drifts away from optimal as the wells change underneath it. Keeping the split near optimal means re-solving as new well-test and rate data arrive.
There are also softer constraints that matter in practice. Wells prone to heading may need a minimum stabilizing rate even if their marginal response is modest, and operations may cap how often setpoints move to avoid chasing noise. A workable allocation blends the marginal-response math with these operational realities so the recommended rates are ones the field can actually hold.
The data an allocation needs, injection gas rate per well, oil and water rates from well tests, casing and tubing pressures, and the total gas available, are exactly what a cloud SCADA already gathers. With those signals in one place, an optimization engine can estimate each well's current performance curve, compute the split that equalizes marginal response within the gas cap, and produce a target injection rate for each well. As fresh rates come in, it re-solves and updates the targets.
Automation also closes the loop on execution. Each well's injection-control valve or flow controller can be driven to the allocated setpoint, and SCADA verifies that the well actually holds it rather than drifting. When a well changes behavior, its curve is re-estimated from the new data and gas is shifted accordingly, so the field tracks a moving optimum instead of a stale one-time plan. This continuous reallocation is where the real gains show up.
Merobix brings the per-well injection rates, well-test results, and pressures for the whole field into a single browser dashboard, which is the foundation any allocation logic needs. With the field's gas usage and each well's response visible together and updating continuously, an engineer can see where scarce lift gas is earning the most oil and rebalance it deliberately rather than leaving last year's setpoints in place.
You give each additional unit of gas to whichever well returns the most extra oil for it, which means equalizing the marginal oil-per-gas response across all wells within the total gas cap. Wells whose performance curves are still climbing steeply get more gas; wells that have flattened out get less. The field is optimal when no reshuffle of gas between wells can add more total oil.
Because wells have different performance curves, an equal split over-gasses some wells past their peak while starving others that would still respond strongly. Moving gas from a low-response well to a high-response well raises total oil, so equal or fixed setpoints leave production on the table. The optimum concentrates scarce gas where the marginal oil return is highest, not where it is spread evenly.
Often enough to keep up with how the wells change. Reservoir pressure, water cut, and valve behavior all shift the performance curves over time, so an allocation set once slowly drifts from optimal. Re-solving as new well-test and rate data arrive, which a SCADA-fed optimizer can do continuously, keeps the split tracking the moving optimum rather than chasing a stale plan.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.