{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-Rf5Cy_o5PR2",
        "outputId": "1c9441b8-3385-49e4-aba1-34c3f4819f92"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Python analysis complete.\n",
            "Saved:\n",
            "- vehicle_performance_summary.csv\n",
            "- weather_performance_summary.csv\n",
            "- region_performance_summary.csv\n",
            "- delivery_mode_performance_summary.csv\n",
            "- distance_performance_summary.csv\n",
            "- qualitative_insights.csv\n"
          ]
        }
      ],
      "source": [
        "import pandas as pd\n",
        "\n",
        "# Load synthetic dataset created in datacreation.ipynb\n",
        "synthetic_delivery_data = pd.read_csv(\"synthetic_delivery_data.csv\")\n",
        "\n",
        "# Preview the columns we will use\n",
        "synthetic_delivery_data[\n",
        "    [\n",
        "        \"vehicle_type\",\n",
        "        \"weather_condition\",\n",
        "        \"delivery_mode\",\n",
        "        \"region\",\n",
        "        \"distance_km\",\n",
        "        \"distance_category\",\n",
        "        \"delay_hours\",\n",
        "        \"delay_score\",\n",
        "        \"performance_label\"\n",
        "    ]\n",
        "].head()\n",
        "\n",
        "# Quantitative analysis: vehicle performance\n",
        "vehicle_performance = (\n",
        "    synthetic_delivery_data\n",
        "    .groupby(\"vehicle_type\")[[\"delay_hours\", \"delay_score\"]]\n",
        "    .mean()\n",
        "    .sort_values(by=\"delay_score\", ascending=False)\n",
        ")\n",
        "\n",
        "vehicle_performance\n",
        "\n",
        "# Quantitative analysis: weather performance\n",
        "weather_performance = (\n",
        "    synthetic_delivery_data\n",
        "    .groupby(\"weather_condition\")[[\"delay_hours\", \"delay_score\"]]\n",
        "    .mean()\n",
        "    .sort_values(by=\"delay_score\", ascending=False)\n",
        ")\n",
        "\n",
        "weather_performance\n",
        "\n",
        "# Quantitative analysis: region performance\n",
        "region_performance = (\n",
        "    synthetic_delivery_data\n",
        "    .groupby(\"region\")[[\"delay_hours\", \"delay_score\"]]\n",
        "    .mean()\n",
        "    .sort_values(by=\"delay_score\", ascending=False)\n",
        ")\n",
        "\n",
        "region_performance\n",
        "\n",
        "# Quantitative analysis: delivery mode performance\n",
        "mode_performance = (\n",
        "    synthetic_delivery_data\n",
        "    .groupby(\"delivery_mode\")[[\"delay_hours\", \"delay_score\"]]\n",
        "    .mean()\n",
        "    .sort_values(by=\"delay_score\", ascending=False)\n",
        ")\n",
        "\n",
        "mode_performance\n",
        "\n",
        "# Quantitative analysis: distance performance\n",
        "distance_performance = (\n",
        "    synthetic_delivery_data\n",
        "    .groupby(\"distance_category\")[[\"delay_hours\", \"delay_score\"]]\n",
        "    .mean()\n",
        "    .sort_values(by=\"delay_score\", ascending=False)\n",
        ")\n",
        "\n",
        "distance_performance\n",
        "\n",
        "# Best conditions summary\n",
        "best_conditions_summary = {\n",
        "    \"Best vehicle type\": vehicle_performance.index[0],\n",
        "    \"Best weather condition\": weather_performance.index[0],\n",
        "    \"Best region\": region_performance.index[0],\n",
        "    \"Best delivery mode\": mode_performance.index[0],\n",
        "    \"Best distance category\": distance_performance.index[0]\n",
        "}\n",
        "\n",
        "best_conditions_summary\n",
        "\n",
        "# Worst conditions summary\n",
        "worst_conditions_summary = {\n",
        "    \"Worst vehicle type\": vehicle_performance.index[-1],\n",
        "    \"Worst weather condition\": weather_performance.index[-1],\n",
        "    \"Worst region\": region_performance.index[-1],\n",
        "    \"Worst delivery mode\": mode_performance.index[-1],\n",
        "    \"Worst distance category\": distance_performance.index[-1]\n",
        "}\n",
        "\n",
        "worst_conditions_summary\n",
        "\n",
        "# Qualitative analysis\n",
        "qualitative_insights = pd.DataFrame({\n",
        "    \"Qualitative Theme\": [\n",
        "        \"Weather disruption\",\n",
        "        \"Urban congestion\",\n",
        "        \"Vehicle suitability\",\n",
        "        \"Customer satisfaction\",\n",
        "        \"Delivery mode pressure\"\n",
        "    ],\n",
        "    \"Business Interpretation\": [\n",
        "        \"Bad weather can increase delivery time and reduce delivery reliability.\",\n",
        "        \"Central regions may create delays because of traffic and route complexity.\",\n",
        "        \"Different vehicle types perform better or worse depending on distance, weather, and region.\",\n",
        "        \"Late deliveries can reduce customer ratings and damage customer trust.\",\n",
        "        \"Express and same-day deliveries create higher operational pressure.\"\n",
        "    ],\n",
        "    \"Suggested Action\": [\n",
        "        \"Use weather-aware route planning and adjust delivery expectations during bad weather.\",\n",
        "        \"Allocate flexible and faster vehicles to central areas.\",\n",
        "        \"Match vehicle type to delivery distance, package type, and regional conditions.\",\n",
        "        \"Prioritize high-risk deliveries before they become delayed.\",\n",
        "        \"Use AI-based delay prediction before accepting urgent delivery promises.\"\n",
        "    ]\n",
        "})\n",
        "\n",
        "qualitative_insights\n",
        "\n",
        "# Save quantitative analysis files\n",
        "vehicle_performance.to_csv(\"vehicle_performance_summary.csv\")\n",
        "weather_performance.to_csv(\"weather_performance_summary.csv\")\n",
        "region_performance.to_csv(\"region_performance_summary.csv\")\n",
        "mode_performance.to_csv(\"delivery_mode_performance_summary.csv\")\n",
        "distance_performance.to_csv(\"distance_performance_summary.csv\")\n",
        "\n",
        "# Save qualitative analysis file\n",
        "qualitative_insights.to_csv(\"qualitative_insights.csv\", index=False)\n",
        "\n",
        "print(\"Python analysis complete.\")\n",
        "print(\"Saved:\")\n",
        "print(\"- vehicle_performance_summary.csv\")\n",
        "print(\"- weather_performance_summary.csv\")\n",
        "print(\"- region_performance_summary.csv\")\n",
        "print(\"- delivery_mode_performance_summary.csv\")\n",
        "print(\"- distance_performance_summary.csv\")\n",
        "print(\"- qualitative_insights.csv\")"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Create the exact folders many university Hugging Face templates expect\n",
        "os.makedirs(\"artifacts/py/figures\", exist_ok=True)\n",
        "os.makedirs(\"artifacts/py/tables\", exist_ok=True)\n",
        "\n",
        "# Save tables\n",
        "vehicle_performance.to_csv(\"artifacts/py/tables/vehicle_performance_summary.csv\")\n",
        "weather_performance.to_csv(\"artifacts/py/tables/weather_performance_summary.csv\")\n",
        "region_performance.to_csv(\"artifacts/py/tables/region_performance_summary.csv\")\n",
        "mode_performance.to_csv(\"artifacts/py/tables/delivery_mode_performance_summary.csv\")\n",
        "distance_performance.to_csv(\"artifacts/py/tables/distance_performance_summary.csv\")\n",
        "qualitative_insights.to_csv(\"artifacts/py/tables/qualitative_insights.csv\", index=False)\n",
        "\n",
        "# Save charts\n",
        "vehicle_performance[\"delay_score\"].plot(kind=\"bar\", title=\"Average Delay Score by Vehicle Type\")\n",
        "plt.ylabel(\"Average Delay Score\")\n",
        "plt.tight_layout()\n",
        "plt.savefig(\"artifacts/py/figures/vehicle_performance_chart.png\")\n",
        "plt.close()\n",
        "\n",
        "weather_performance[\"delay_score\"].plot(kind=\"bar\", title=\"Average Delay Score by Weather Condition\")\n",
        "plt.ylabel(\"Average Delay Score\")\n",
        "plt.tight_layout()\n",
        "plt.savefig(\"artifacts/py/figures/weather_performance_chart.png\")\n",
        "plt.close()\n",
        "\n",
        "region_performance[\"delay_score\"].plot(kind=\"bar\", title=\"Average Delay Score by Region\")\n",
        "plt.ylabel(\"Average Delay Score\")\n",
        "plt.tight_layout()\n",
        "plt.savefig(\"artifacts/py/figures/region_performance_chart.png\")\n",
        "plt.close()\n",
        "\n",
        "mode_performance[\"delay_score\"].plot(kind=\"bar\", title=\"Average Delay Score by Delivery Mode\")\n",
        "plt.ylabel(\"Average Delay Score\")\n",
        "plt.tight_layout()\n",
        "plt.savefig(\"artifacts/py/figures/delivery_mode_performance_chart.png\")\n",
        "plt.close()\n",
        "\n",
        "distance_performance[\"delay_score\"].plot(kind=\"bar\", title=\"Average Delay Score by Distance Category\")\n",
        "plt.ylabel(\"Average Delay Score\")\n",
        "plt.tight_layout()\n",
        "plt.savefig(\"artifacts/py/figures/distance_performance_chart.png\")\n",
        "plt.close()\n",
        "\n",
        "print(\"Saved outputs for Hugging Face template.\")\n",
        "print(\"Figures saved in: artifacts/py/figures\")\n",
        "print(\"Tables saved in: artifacts/py/tables\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3f3Jj1q2_OTo",
        "outputId": "4748d8e9-fb59-4730-a808-a2a1aa196d83"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saved outputs for Hugging Face template.\n",
            "Figures saved in: artifacts/py/figures\n",
            "Tables saved in: artifacts/py/tables\n"
          ]
        }
      ]
    }
  ]
}