The United States Army is hard at work on an algorithm that will best determine the amount of caffeine necessary to keep a person functioning at peak performance based on factors such as individual physiology.

The caffeine-dosing strategy relies on the algorithm to make predictions about how much caffeine an individual should consume and when to consume it to most effectively stave off sleep while also maintaining peak functioning.

Because caffeine affects individuals differently, the algorithm has to consider consumers’ unique metabolisms and sleep patterns among other factors.

As such, researcher Jaques Reifman conducted a handful of brief response time tests — called psychomotor vigilance tasks (PVTs) — over a couple of days to determine an individual’s reaction to caffeine as well as their sleep schedule.

Based on that data, the algorithm developed “caffeine consumption schedules” that demonstrated peak performance results, in some cases using 65 percent less caffeine or that enhanced episodes of peak performance by 64 percent using comparable amounts of caffeine.

The algorithm is currently being tested on soldiers in training, yet developers see this algorithm one-day being useful for shift workers, doctors, nurses or truck drivers.