基于R语言的供需动态计算及净赤字列生成技术问询

Alright, let's work through this problem. You've got 50 days of supply/demand data with a supply-demand difference column, and you need to add a net deficit column while tracking a reservoir with a 200-unit capacity. Here's how to implement this logic cleanly in R, with both base R and tidyverse options:

First, let's set up sample data to test with #

I'll create a dummy dataset that mimics your setup—feel free to swap this out with your actual data:

set.seed(123) # For reproducibility
df <- data.frame(
  Day = 1:50,
  Demand = sample(50:150, 50, replace = TRUE),
  Supply = sample(60:160, 50, replace = TRUE)
)
df$Supply_Demand_Diff <- df$Supply - df$Demand # Your existing difference column

Base R Approach (Using a Loop) #

Since we need to track the reservoir level day-over-day (a cumulative state), a loop works well here for straightforward readability:

# Initialize reservoir and net deficit column
reservoir <- 0
df$net_deficit <- 0

for (i in seq(nrow(df))) {
  daily_gap <- df$Supply[i] - df$Demand[i]
  
  if (daily_gap < 0) {
    # Demand exceeds supply: use reservoir to cover the gap
    required <- abs(daily_gap)
    if (reservoir >= required) {
      # Reservoir has enough to cover the entire deficit
      reservoir <- reservoir - required
      df$net_deficit[i] <- 0
    } else {
      # Reservoir can't cover all demand—net deficit is the unmet amount
      df$net_deficit[i] <- required - reservoir
      reservoir <- 0
    }
  } else {
    # Supply meets or exceeds demand: store excess in reservoir (if there's space)
    excess <- daily_gap
    reservoir <- min(reservoir + excess, 200) # Cap at 200 (full capacity)
    df$net_deficit[i] <- 0
  }
}

Tidyverse Approach (Functional Style with accumulate) #

If you prefer a pipe-based workflow, use purrr::accumulate to carry over the reservoir state between days:

library(dplyr)
library(purrr)

df <- df %>%
  mutate(
    # Track both reservoir level and net deficit with accumulate
    daily_state = accumulate(1:nrow(.), function(prev_state, day_idx) {
      daily_gap <- Supply[day_idx] - Demand[day_idx]
      current_reservoir <- prev_state$reservoir
      
      if (daily_gap < 0) {
        required <- abs(daily_gap)
        if (current_reservoir >= required) {
          new_reservoir <- current_reservoir - required
          net_def <- 0
        } else {
          net_def <- required - current_reservoir
          new_reservoir <- 0
        }
      } else {
        excess <- daily_gap
        new_reservoir <- min(current_reservoir + excess, 200)
        net_def <- 0
      }
      
      list(reservoir = new_reservoir, net_deficit = net_def)
    }, .init = list(reservoir = 0, net_deficit = 0)) %>%
    tail(-1) # Drop the initial empty state
  ) %>%
  # Extract values from the state list into columns
  mutate(
    net_deficit = map_dbl(daily_state, ~.$net_deficit),
    reservoir_level = map_dbl(daily_state, ~.$reservoir),
    daily_state = NULL # Clean up the list column if not needed
  )

Notes on the Logic #

• I added handling for the edge case where the reservoir can't cover the full demand (you didn't mention this, but it's a realistic scenario). If your business rule says net deficit should always be 0 even if demand isn't met, just set df$net_deficit[i] <- 0 in that branch.
• Both methods ensure the reservoir never exceeds its 200-unit capacity—excess supply beyond that is discarded.
• The reservoir_level column is optional, but it's helpful to verify the reservoir is behaving as expected.

内容的提问来源于stack exchange，提问作者89_Simple