Weather Impact Scheduler Agent Skill

import pandas as pd
from datetime import date, timedelta
from typing import Dict, Any, List, Optional
from dataclasses import dataclass, field
from enum import Enum


class WeatherCondition(Enum):
    CLEAR = "clear"
    CLOUDY = "cloudy"
    RAIN = "rain"
    HEAVY_RAIN = "heavy_rain"
    SNOW = "snow"
    WIND = "wind"
    EXTREME_HEAT = "extreme_heat"
    EXTREME_COLD = "extreme_cold"


class ActivitySensitivity(Enum):
    HIGH = "high"      # Concrete, painting, roofing
    MEDIUM = "medium"  # Excavation, masonry
    LOW = "low"        # Indoor work


@dataclass
class WeatherForecast:
    forecast_date: date
    condition: WeatherCondition
    high_temp: float
    low_temp: float
    precipitation_mm: float
    wind_speed_kmh: float


@dataclass
class ScheduleActivity:
    activity_id: str
    name: str
    start_date: date
    end_date: date
    sensitivity: ActivitySensitivity
    outdoor: bool
    can_work_in_rain: bool = False
    min_temp: float = 5.0
    max_temp: float = 35.0
    max_wind: float = 50.0


@dataclass
class WeatherImpact:
    activity_id: str
    impact_date: date
    reason: str
    delay_hours: float
    recommendation: str


class WeatherImpactScheduler:
    def __init__(self, project_name: str):
        self.project_name = project_name
        self.activities: Dict[str, ScheduleActivity] = {}
        self.forecasts: Dict[date, WeatherForecast] = {}
        self.impacts: List[WeatherImpact] = []

    def add_activity(self, activity_id: str, name: str, start_date: date,
                    end_date: date, sensitivity: ActivitySensitivity,
                    outdoor: bool = True, can_work_in_rain: bool = False) -> ScheduleActivity:
        activity = ScheduleActivity(
            activity_id=activity_id,
            name=name,
            start_date=start_date,
            end_date=end_date,
            sensitivity=sensitivity,
            outdoor=outdoor,
            can_work_in_rain=can_work_in_rain
        )
        self.activities[activity_id] = activity
        return activity

    def add_forecast(self, forecast_date: date, condition: WeatherCondition,
                    high_temp: float, low_temp: float,
                    precipitation_mm: float = 0, wind_speed_kmh: float = 0):
        forecast = WeatherForecast(
            forecast_date=forecast_date,
            condition=condition,
            high_temp=high_temp,
            low_temp=low_temp,
            precipitation_mm=precipitation_mm,
            wind_speed_kmh=wind_speed_kmh
        )
        self.forecasts[forecast_date] = forecast

    def analyze_impacts(self) -> List[WeatherImpact]:
        self.impacts = []

        for activity in self.activities.values():
            if not activity.outdoor:
                continue

            current = activity.start_date
            while current <= activity.end_date:
                forecast = self.forecasts.get(current)
                if forecast:
                    impact = self._check_impact(activity, forecast)
                    if impact:
                        self.impacts.append(impact)
                current += timedelta(days=1)

        return self.impacts

    def _check_impact(self, activity: ScheduleActivity,
                     forecast: WeatherForecast) -> Optional[WeatherImpact]:
        reasons = []
        delay_hours = 0

        # Check precipitation
        if forecast.condition in [WeatherCondition.RAIN, WeatherCondition.HEAVY_RAIN]:
            if not activity.can_work_in_rain:
                if activity.sensitivity == ActivitySensitivity.HIGH:
                    reasons.append("Rain - high sensitivity activity")
                    delay_hours = 8
                else:
                    reasons.append("Rain delays")
                    delay_hours = 4

        # Check temperature
        if forecast.low_temp < activity.min_temp:
            reasons.append(f"Too cold ({forecast.low_temp}°C)")
            delay_hours = max(delay_hours, 8 if activity.sensitivity == ActivitySensitivity.HIGH else 4)

        if forecast.high_temp > activity.max_temp:
            reasons.append(f"Too hot ({forecast.high_temp}°C)")
            delay_hours = max(delay_hours, 4)

        # Check wind
        if forecast.wind_speed_kmh > activity.max_wind:
            reasons.append(f"High wind ({forecast.wind_speed_kmh} km/h)")
            delay_hours = max(delay_hours, 8)

        if reasons:
            return WeatherImpact(
                activity_id=activity.activity_id,
                impact_date=forecast.forecast_date,
                reason="; ".join(reasons),
                delay_hours=delay_hours,
                recommendation=self._get_recommendation(activity, forecast)
            )
        return None

    def _get_recommendation(self, activity: ScheduleActivity,
                           forecast: WeatherForecast) -> str:
        if forecast.condition in [WeatherCondition.RAIN, WeatherCondition.HEAVY_RAIN]:
            return "Reschedule or plan indoor work"
        if forecast.low_temp < activity.min_temp:
            return "Use heating blankets or delay start"
        if forecast.high_temp > activity.max_temp:
            return "Start early, plan heat breaks"
        if forecast.wind_speed_kmh > activity.max_wind:
            return "Secure materials, delay crane work"
        return "Monitor conditions"

    def get_total_delay_forecast(self) -> Dict[str, Any]:
        total_hours = sum(i.delay_hours for i in self.impacts)
        by_activity = {}
        for impact in self.impacts:
            act = impact.activity_id
            by_activity[act] = by_activity.get(act, 0) + impact.delay_hours

        return {
            'total_impact_hours': total_hours,
            'total_impact_days': round(total_hours / 8, 1),
            'affected_activities': len(by_activity),
            'by_activity': by_activity,
            'impact_count': len(self.impacts)
        }

    def export_analysis(self, output_path: str):
        data = [{
            'Activity': i.activity_id,
            'Date': i.impact_date,
            'Reason': i.reason,
            'Delay Hours': i.delay_hours,
            'Recommendation': i.recommendation
        } for i in self.impacts]
        pd.DataFrame(data).to_excel(output_path, index=False)

scheduler = WeatherImpactScheduler("Office Tower")

# Add activities
scheduler.add_activity("CONC-001", "Pour Slab L3", date(2024, 3, 15),
                      date(2024, 3, 20), ActivitySensitivity.HIGH, outdoor=True)

# Add forecasts
scheduler.add_forecast(date(2024, 3, 17), WeatherCondition.RAIN, 15, 8, 25, 20)

# Analyze
impacts = scheduler.analyze_impacts()
summary = scheduler.get_total_delay_forecast()
print(f"Projected delay: {summary['total_impact_days']} days")