Skip to contents

Generate Composite Indicator Validation Flags

get_indicators_flags.Rd

This function validates derived fisheries indicators (CPUE, RPUE, and price per kg) from KEFS survey data. It calculates these performance metrics and flags values that exceed specified maximum thresholds. This validation is applied only to submissions that have passed previous validation checks.

Usage

get_indicators_flags(dat = NULL, limits = NULL, clean_ids = NULL)

Arguments

dat

A data frame containing KEFS survey data with columns:

  • submission_id: Unique identifier for the submission

  • no_of_fishers: Number of fishers on the trip

  • trip_duration: Duration of the trip in hours

  • total_catch_weight: Total weight of catch in kilograms

  • total_catch_price: Total price of catch in Kenyan Shillings

limits

A list containing threshold values for validation:

  • max_cpue: Maximum acceptable catch per unit effort (kg/fisher/hour)

  • max_rpue: Maximum acceptable revenue per unit effort (KSH/fisher/hour)

  • max_price_kg: Maximum acceptable price per kilogram (KSH/kg)

clean_ids

A vector of submission IDs that have passed previous validation checks. Only these submissions will be evaluated for indicator-based anomalies.

Value

A data frame with columns:

  • submission_id: The original submission identifier

  • alert_flag_indicators: Comma-separated string of alert codes (NA if no alerts):

    • "6.1": CPUE (catch per unit effort) exceeds maximum

    • "6.2": RPUE (revenue per unit effort) exceeds maximum

    • "6.3": Price per kilogram exceeds maximum

Details

The function calculates three key fisheries performance indicators:

  1. CPUE (Catch Per Unit Effort): total_catch_weight / no_of_fishers / trip_duration

  2. RPUE (Revenue Per Unit Effort): total_catch_price / no_of_fishers / trip_duration

  3. Price per kg: total_catch_price / total_catch_weight

These indicators help identify submissions with unrealistically high productivity or pricing that may indicate data entry errors or exceptional circumstances requiring review.

This function is typically called after trip-level and catch-level validations have been performed, applying additional scrutiny only to submissions that have already been determined to be generally valid.

Examples

if (FALSE) { # \dontrun{
# Get IDs that passed previous validation
clean_ids <- dplyr::full_join(trip_flags, catch_flags, by = "submission_id") |>
  dplyr::filter(is.na(alert_flag_trip) & is.na(alert_flag_catch)) |>
  dplyr::pull(submission_id) |>
  unique()

indicator_limits <- list(
  max_cpue = 20,
  max_rpue = 3876,
  max_price_kg = 3876
)
indicator_flags <- get_indicators_flags(
  dat = survey_data,
  limits = indicator_limits,
  clean_ids = clean_ids
)
} # }