R Vectorize FOR loop using previous iteration values

Is it possible to vectorize / speed up the execution of a FOR loop that is using the previous iteration values ?

In the reproductive example below :

• the current production is computed from the current stock
• the current production updates the NEXT stock
• the next iteration used the updated stock to determine the current production, etc...

So I need to compute the stock at each iteration, in order to compute the production setpoint... Is it possible to avoid (slow) for loop ?

The current implementation takes about 45 seconds for 50k lines.

# Dummy functions for the examples. Real code is more complicated
function1 <- function(energy, stock, critical) {
    if (stock < critical) {
        return (energy)
    } else {
        return(0)
    }
}
function2 <- function(power) {
  return(round(power/100))
}
# Dummy data
d <- data.frame( "energy"= c(660, 660, 660, 660),
                 "stock" = c(20,   0,    0, 0),
                 "delivery" = c(0, 0, 2, 0),
                 "critical" = c(50, 50 ,50, 50),
                 "power" = c(0, 0, 0, 0),
                 "production" = c(0, 0, 0, 0) )

for (i in 1:length(d$energy)) {

  # Computing power, based on CUURENT stock
  d$power[i] <- function1(d$energy[i], d$stock[i], d$critical[i])

  # Computing production
  d$production[i] <- function2(d$power[i])

  # Updating NEXT stock with current production / delivery
  if (i < length(d$energy)) {
    d$stock[i+1] <- d$stock[i] + d$production[i] - d$delivery[i]
  }
}

View(d)