datadrivenconstruction/cwicr-rate-updater
Update CWICR resource rates with current market prices. Integrate external price data, apply inflation adjustments, and maintain rate history.
npx skills add https://github.com/datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill cwicr-rate-updaterResource rates become outdated:
Systematic rate updates integrating market data, inflation indices, and regional factors while maintaining audit trail.
import pandas as pd
import numpy as np
from typing import Dict, Any, List, Optional, Tuple, Callable
from dataclasses import dataclass, field
from datetime import datetime, date
from enum import Enum
import json
class RateType(Enum):
"""Types of rates."""
LABOR = "labor"
MATERIAL = "material"
EQUIPMENT = "equipment"
SUBCONTRACT = "subcontract"
class AdjustmentMethod(Enum):
"""Methods for rate adjustment."""
FIXED_AMOUNT = "fixed_amount"
PERCENTAGE = "percentage"
MULTIPLIER = "multiplier"
REPLACEMENT = "replacement"
@dataclass
class RateChange:
"""Record of rate change."""
resource_code: str
rate_type: RateType
old_rate: float
new_rate: float
change_percent: float
change_date: datetime
reason: str
source: str
@dataclass
class RateUpdateResult:
"""Result of rate update operation."""
total_items: int
updated: int
unchanged: int
errors: int
changes: List[RateChange]
summary: Dict[str, Any]
class CWICRRateUpdater:
"""Update resource rates in CWICR data."""
def __init__(self, cwicr_data: pd.DataFrame):
self.data = cwicr_data.copy()
self.change_log: List[RateChange] = []
self.original_data = cwicr_data.copy()
def get_current_rates(self,
rate_type: RateType = None,
category: str = None) -> pd.DataFrame:
"""Get current rates, optionally filtered."""
df = self.data.copy()
# Filter by category if specified
if category and 'category' in df.columns:
df = df[df['category'].str.contains(category, case=False, na=False)]
# Select relevant columns based on rate type
rate_columns = {
RateType.LABOR: ['work_item_code', 'description', 'labor_rate', 'labor_cost'],
RateType.MATERIAL: ['work_item_code', 'description', 'material_cost'],
RateType.EQUIPMENT: ['work_item_code', 'description', 'equipment_cost', 'equipment_rate']
}
if rate_type and rate_type in rate_columns:
cols = [c for c in rate_columns[rate_type] if c in df.columns]
return df[cols]
return df
def update_rate(self,
work_item_code: str,
rate_type: RateType,
new_rate: float,
reason: str = "Manual update",
source: str = "User") -> Optional[RateChange]:
"""Update single rate."""
rate_column = self._get_rate_column(rate_type)
if rate_column not in self.data.columns:
return None
mask = self.data['work_item_code'] == work_item_code
if not mask.any():
return None
old_rate = float(self.data.loc[mask, rate_column].iloc[0])
self.data.loc[mask, rate_column] = new_rate
change_percent = ((new_rate - old_rate) / old_rate * 100) if old_rate > 0 else 0
change = RateChange(
resource_code=work_item_code,
rate_type=rate_type,
old_rate=old_rate,
new_rate=new_rate,
change_percent=round(change_percent, 2),
change_date=datetime.now(),
reason=reason,
source=source
)
self.change_log.append(change)
return change
def _get_rate_column(self, rate_type: RateType) -> str:
"""Get column name for rate type."""
mapping = {
RateType.LABOR: 'labor_rate',
RateType.MATERIAL: 'material_cost',
RateType.EQUIPMENT: 'equipment_cost',
RateType.SUBCONTRACT: 'subcontract_cost'
}
return mapping.get(rate_type, 'labor_rate')
def apply_percentage_adjustment(self,
rate_type: RateType,
percentage: float,
category: str = None,
reason: str = "Percentage adjustment") -> RateUpdateResult:
"""Apply percentage adjustment to rates."""
rate_column = self._get_rate_column(rate_type)
if rate_column not in self.data.columns:
return RateUpdateResult(0, 0, 0, 1, [], {})
# Build mask
mask = pd.Series([True] * len(self.data))
if category and 'category' in self.data.columns:
mask = self.data['category'].str.contains(category, case=False, na=False)
# Store old values
old_values = self.data.loc[mask, rate_column].copy()
# Apply adjustment
multiplier = 1 + (percentage / 100)
self.data.loc[mask, rate_column] = old_values * multiplier
# Record changes
changes = []
for idx in self.data[mask].index:
old_rate = float(old_values.loc[idx])
new_rate = float(self.data.loc[idx, rate_column])
if old_rate != new_rate:
change = RateChange(
resource_code=str(self.data.loc[idx, 'work_item_code']),
rate_type=rate_type,
old_rate=old_rate,
new_rate=new_rate,
change_percent=percentage,
change_date=datetime.now(),
reason=reason,
source=f"Bulk {percentage}%"
)
changes.append(change)
self.change_log.append(change)
return RateUpdateResult(
total_items=len(self.data[mask]),
updated=len(changes),
unchanged=len(self.data[mask]) - len(changes),
errors=0,
changes=changes,
summary={
'rate_type': rate_type.value,
'adjustment_percent': percentage,
'category': category,
'average_new_rate': self.data.loc[mask, rate_column].mean()
}
)
def apply_inflation_index(self,
base_year: int,
current_year: int,
inflation_rates: Dict[int, float],
rate_types: List[RateType] = None) -> RateUpdateResult:
"""Apply inflation index from base year to current."""
if rate_types is None:
rate_types = [RateType.LABOR, RateType.MATERIAL, RateType.EQUIPMENT]
# Calculate cumulative multiplier
cumulative_multiplier = 1.0
for year in range(base_year, current_year):
rate = inflation_rates.get(year, 0.02) # Default 2%
cumulative_multiplier *= (1 + rate)
total_changes = []
for rate_type in rate_types:
result = self.apply_percentage_adjustment(
rate_type=rate_type,
percentage=(cumulative_multiplier - 1) * 100,
reason=f"Inflation {base_year}-{current_year}"
)
total_changes.extend(result.changes)
return RateUpdateResult(
total_items=len(self.data),
updated=len(total_changes),
unchanged=len(self.data) - len(total_changes),
errors=0,
changes=total_changes,
summary={
'base_year': base_year,
'current_year': current_year,
'cumulative_multiplier': round(cumulative_multiplier, 4),
'total_adjustment_percent': round((cumulative_multiplier - 1) * 100, 2)
}
)
def import_external_rates(self,
external_data: pd.DataFrame,
code_column: str,
rate_column: str,
rate_type: RateType,
match_on: str = 'work_item_code') -> RateUpdateResult:
"""Import rates from external data source."""
changes = []
errors = 0
target_column = self._get_rate_column(rate_type)
for _, row in external_data.iterrows():
code = row[code_column]
new_rate = row[rate_column]
try:
change = self.update_rate(
work_item_code=code,
rate_type=rate_type,
new_rate=new_rate,
reason="External import",
source="External data"
)
if change:
changes.append(change)
except Exception:
errors += 1
return RateUpdateResult(
total_items=len(external_data),
updated=len(changes),
unchanged=len(external_data) - len(changes) - errors,
errors=errors,
changes=changes,
summary={
'source': 'External import',
'rate_type': rate_type.value
}
)
def apply_regional_factors(self,
region_factors: Dict[str, float],
default_factor: float = 1.0) -> RateUpdateResult:
"""Apply regional adjustment factors."""
# This assumes region column exists or applies uniformly
factor = region_factors.get('default', default_factor)
labor_result = self.apply_percentage_adjustment(
RateType.LABOR,
(region_factors.get('labor', factor) - 1) * 100,
reason="Regional adjustment"
)
material_result = self.apply_percentage_adjustment(
RateType.MATERIAL,
(region_factors.get('material', factor) - 1) * 100,
reason="Regional adjustment"
)
equipment_result = self.apply_percentage_adjustment(
RateType.EQUIPMENT,
(region_factors.get('equipment', factor) - 1) * 100,
reason="Regional adjustment"
)
all_changes = (labor_result.changes + material_result.changes +
equipment_result.changes)
return RateUpdateResult(
total_items=len(self.data),
updated=len(all_changes),
unchanged=len(self.data) * 3 - len(all_changes),
errors=0,
changes=all_changes,
summary={
'region_factors': region_factors,
'labor_adjusted': len(labor_result.changes),
'material_adjusted': len(material_result.changes),
'equipment_adjusted': len(equipment_result.changes)
}
)
def get_change_log(self,
start_date: datetime = None,
rate_type: RateType = None) -> List[RateChange]:
"""Get change log, optionally filtered."""
changes = self.change_log
if start_date:
changes = [c for c in changes if c.change_date >= start_date]
if rate_type:
changes = [c for c in changes if c.rate_type == rate_type]
return changes
def export_change_log(self, output_path: str) -> str:
"""Export change log to Excel."""
df = pd.DataFrame([
{
'Resource Code': c.resource_code,
'Rate Type': c.rate_type.value,
'Old Rate': c.old_rate,
'New Rate': c.new_rate,
'Change %': c.change_percent,
'Date': c.change_date.strftime('%Y-%m-%d %H:%M'),
'Reason': c.reason,
'Source': c.source
}
for c in self.change_log
])
df.to_excel(output_path, index=False)
return output_path
def rollback_changes(self,
since: datetime = None) -> int:
"""Rollback changes since date (returns to original data)."""
if since is None:
# Full rollback
self.data = self.original_data.copy()
count = len(self.change_log)
self.change_log = []
return count
# Partial rollback - more complex, would need versioning
return 0
def export_updated_data(self, output_path: str) -> str:
"""Export updated CWICR data."""
if output_path.endswith('.parquet'):
self.data.to_parquet(output_path)
else:
self.data.to_excel(output_path, index=False)
return output_path
class RateScheduler:
"""Schedule automatic rate updates."""
def __init__(self, updater: CWICRRateUpdater):
self.updater = updater
self.schedules: List[Dict[str, Any]] = []
def add_annual_labor_increase(self,
percentage: float,
effective_date: date) -> Dict[str, Any]:
"""Schedule annual labor rate increase."""
schedule = {
'id': len(self.schedules) + 1,
'type': 'annual_labor',
'percentage': percentage,
'effective_date': effective_date,
'rate_type': RateType.LABOR,
'status': 'scheduled'
}
self.schedules.append(schedule)
return schedule
def execute_due_updates(self, current_date: date = None) -> List[RateUpdateResult]:
"""Execute all updates that are due."""
if current_date is None:
current_date = date.today()
results = []
for schedule in self.schedules:
if schedule['status'] == 'scheduled' and schedule['effective_date'] <= current_date:
result = self.updater.apply_percentage_adjustment(
rate_type=schedule['rate_type'],
percentage=schedule['percentage'],
reason=f"Scheduled {schedule['type']}"
)
schedule['status'] = 'executed'
schedule['executed_date'] = current_date
results.append(result)
return results
# Load CWICR data
cwicr = pd.read_parquet("ddc_cwicr_en.parquet")
# Initialize updater
updater = CWICRRateUpdater(cwicr)
# Apply 5% labor rate increase
result = updater.apply_percentage_adjustment(
rate_type=RateType.LABOR,
percentage=5.0,
reason="2024 Annual Increase"
)
print(f"Updated {result.updated} labor rates")
print(f"Average adjustment: {result.summary.get('adjustment_percent')}%")
inflation_rates = {
2020: 0.012, 2021: 0.047, 2022: 0.065, 2023: 0.034
}
result = updater.apply_inflation_index(
base_year=2020,
current_year=2024,
inflation_rates=inflation_rates
)
print(f"Cumulative adjustment: {result.summary['total_adjustment_percent']}%")
berlin_factors = {
'labor': 1.15,
'material': 0.95,
'equipment': 1.05
}
result = updater.apply_regional_factors(berlin_factors)
market_prices = pd.read_excel("current_prices.xlsx")
result = updater.import_external_rates(
external_data=market_prices,
code_column='item_code',
rate_column='price',
rate_type=RateType.MATERIAL
)
updater.export_change_log("rate_changes_2024.xlsx")
Take datadrivenconstruction/cwicr-rate-updater from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.