Food Delivery Time Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "369be5c1",
   "metadata": {},
   "source": [
    "# Food Delivery Time Prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c695895f",
   "metadata": {},
   "source": [
    "## Background\n",
    "\n",
    "Delivery service is a form of service that provides services to deliver orders (especially goods) that ordered by customers to a place according to their wishes. Currently, delivery service is one of the services most needed by people in obtaining the goods needed because it saves a lot of time and energy.\n",
    "\n",
    "So, currently there are many types of businesses that have delivery services as one of the services that can be provided to customers. One of them is a business engaged in the food or beverage sector that has a service to deliver food ordered to the place the customer wants.\n",
    "\n",
    "However, of course the timeliness required in delivering food to the customer's place is the main challenge in this service. Accuracy time in delivering the food must be shown to keep transparency with their customers. So, by using historical data on the time it takes to deliver food, the use of machine learning algorithms is one way to predict the accuracy of the time needed to deliver food to location."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e5e7d80",
   "metadata": {},
   "source": [
    "## Objectives\n",
    "\n",
    "1. What factors can affect the time in delivering food from the restaurant to the destination location?\n",
    "2. How much food delivery time prediction accuracy?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59f64870",
   "metadata": {},
   "source": [
    "## Load Data & Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "cabd7df1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: shap in c:\\users\\asus\\anaconda3\\lib\\site-packages (0.41.0)\n",
      "Requirement already satisfied: packaging>20.9 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (21.3)\n",
      "Requirement already satisfied: cloudpickle in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (2.0.0)\n",
      "Requirement already satisfied: scipy in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (1.7.3)\n",
      "Requirement already satisfied: numpy in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (1.21.5)\n",
      "Requirement already satisfied: pandas in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (1.4.2)\n",
      "Requirement already satisfied: numba in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (0.55.1)\n",
      "Requirement already satisfied: slicer==0.0.7 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (0.0.7)\n",
      "Requirement already satisfied: tqdm>4.25.0 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (4.64.0)\n",
      "Requirement already satisfied: scikit-learn in c:\\users\\asus\\anaconda3\\lib\\site-packages (from shap) (1.0.2)\n",
      "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from packaging>20.9->shap) (3.0.4)\n",
      "Requirement already satisfied: colorama in c:\\users\\asus\\anaconda3\\lib\\site-packages (from tqdm>4.25.0->shap) (0.4.4)\n",
      "Requirement already satisfied: setuptools in c:\\users\\asus\\anaconda3\\lib\\site-packages (from numba->shap) (61.2.0)\n",
      "Requirement already satisfied: llvmlite<0.39,>=0.38.0rc1 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from numba->shap) (0.38.0)\n",
      "Requirement already satisfied: pytz>=2020.1 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from pandas->shap) (2021.3)\n",
      "Requirement already satisfied: python-dateutil>=2.8.1 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from pandas->shap) (2.8.2)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from python-dateutil>=2.8.1->pandas->shap) (1.16.0)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from scikit-learn->shap) (2.2.0)\n",
      "Requirement already satisfied: joblib>=0.11 in c:\\users\\asus\\anaconda3\\lib\\site-packages (from scikit-learn->shap) (1.1.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install shap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "57e585a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Standard library\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Train test split library\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Multicollinearity check library\n",
    "\n",
    "from statsmodels.stats.outliers_influence import variance_inflation_factor as vif \n",
    "from statsmodels.tools.tools import add_constant\n",
    "\n",
    "# Regression algorithm library\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "# Evaluation metric library\n",
    "\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "\n",
    "# Hyperparameter tuning library\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# Feature importance library\n",
    "\n",
    "import shap\n",
    "\n",
    "# Save model\n",
    "\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8a0d492d",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Delivery_person_ID</th>\n",
       "      <th>Delivery_person_Age</th>\n",
       "      <th>Delivery_person_Ratings</th>\n",
       "      <th>Restaurant_latitude</th>\n",
       "      <th>Restaurant_longitude</th>\n",
       "      <th>Delivery_location_latitude</th>\n",
       "      <th>Delivery_location_longitude</th>\n",
       "      <th>Type_of_order</th>\n",
       "      <th>Type_of_vehicle</th>\n",
       "      <th>Time_taken(min)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4607</td>\n",
       "      <td>INDORES13DEL02</td>\n",
       "      <td>37</td>\n",
       "      <td>4.9</td>\n",
       "      <td>22.745049</td>\n",
       "      <td>75.892471</td>\n",
       "      <td>22.765049</td>\n",
       "      <td>75.912471</td>\n",
       "      <td>Snack</td>\n",
       "      <td>motorcycle</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>B379</td>\n",
       "      <td>BANGRES18DEL02</td>\n",
       "      <td>34</td>\n",
       "      <td>4.5</td>\n",
       "      <td>12.913041</td>\n",
       "      <td>77.683237</td>\n",
       "      <td>13.043041</td>\n",
       "      <td>77.813237</td>\n",
       "      <td>Snack</td>\n",
       "      <td>scooter</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5D6D</td>\n",
       "      <td>BANGRES19DEL01</td>\n",
       "      <td>23</td>\n",
       "      <td>4.4</td>\n",
       "      <td>12.914264</td>\n",
       "      <td>77.678400</td>\n",
       "      <td>12.924264</td>\n",
       "      <td>77.688400</td>\n",
       "      <td>Drinks</td>\n",
       "      <td>motorcycle</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7A6A</td>\n",
       "      <td>COIMBRES13DEL02</td>\n",
       "      <td>38</td>\n",
       "      <td>4.7</td>\n",
       "      <td>11.003669</td>\n",
       "      <td>76.976494</td>\n",
       "      <td>11.053669</td>\n",
       "      <td>77.026494</td>\n",
       "      <td>Buffet</td>\n",
       "      <td>motorcycle</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>70A2</td>\n",
       "      <td>CHENRES12DEL01</td>\n",
       "      <td>32</td>\n",
       "      <td>4.6</td>\n",
       "      <td>12.972793</td>\n",
       "      <td>80.249982</td>\n",
       "      <td>13.012793</td>\n",
       "      <td>80.289982</td>\n",
       "      <td>Snack</td>\n",
       "      <td>scooter</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     ID Delivery_person_ID  Delivery_person_Age  Delivery_person_Ratings  \\\n",
       "0  4607     INDORES13DEL02                   37                      4.9   \n",
       "1  B379     BANGRES18DEL02                   34                      4.5   \n",
       "2  5D6D     BANGRES19DEL01                   23                      4.4   \n",
       "3  7A6A    COIMBRES13DEL02                   38                      4.7   \n",
       "4  70A2     CHENRES12DEL01                   32                      4.6   \n",
       "\n",
       "   Restaurant_latitude  Restaurant_longitude  Delivery_location_latitude  \\\n",
       "0            22.745049             75.892471                   22.765049   \n",
       "1            12.913041             77.683237                   13.043041   \n",
       "2            12.914264             77.678400                   12.924264   \n",
       "3            11.003669             76.976494                   11.053669   \n",
       "4            12.972793             80.249982                   13.012793   \n",
       "\n",
       "   Delivery_location_longitude Type_of_order Type_of_vehicle  Time_taken(min)  \n",
       "0                    75.912471        Snack      motorcycle                24  \n",
       "1                    77.813237        Snack         scooter                33  \n",
       "2                    77.688400       Drinks      motorcycle                26  \n",
       "3                    77.026494       Buffet      motorcycle                21  \n",
       "4                    80.289982        Snack         scooter                30  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load dataset\n",
    "\n",
    "data = pd.read_csv('deliverytime.txt',sep=\",\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "666a5c6c",
   "metadata": {},
   "source": [
    "## Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "bd967c15",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 45593 entries, 0 to 45592\n",
      "Data columns (total 11 columns):\n",
      " #   Column                       Non-Null Count  Dtype  \n",
      "---  ------                       --------------  -----  \n",
      " 0   ID                           45593 non-null  object \n",
      " 1   Delivery_person_ID           45593 non-null  object \n",
      " 2   Delivery_person_Age          45593 non-null  int64  \n",
      " 3   Delivery_person_Ratings      45593 non-null  float64\n",
      " 4   Restaurant_latitude          45593 non-null  float64\n",
      " 5   Restaurant_longitude         45593 non-null  float64\n",
      " 6   Delivery_location_latitude   45593 non-null  float64\n",
      " 7   Delivery_location_longitude  45593 non-null  float64\n",
      " 8   Type_of_order                45593 non-null  object \n",
      " 9   Type_of_vehicle              45593 non-null  object \n",
      " 10  Time_taken(min)              45593 non-null  int64  \n",
      "dtypes: float64(5), int64(2), object(4)\n",
      "memory usage: 3.8+ MB\n"
     ]
    }
   ],
   "source": [
    "# General info\n",
    "\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "de8a75ba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ID                             0.0\n",
       "Delivery_person_ID             0.0\n",
       "Delivery_person_Age            0.0\n",
       "Delivery_person_Ratings        0.0\n",
       "Restaurant_latitude            0.0\n",
       "Restaurant_longitude           0.0\n",
       "Delivery_location_latitude     0.0\n",
       "Delivery_location_longitude    0.0\n",
       "Type_of_order                  0.0\n",
       "Type_of_vehicle                0.0\n",
       "Time_taken(min)                0.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check ratio missing value\n",
    "\n",
    "data.isna().sum() / len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5e89a786",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check duplicated data\n",
    "\n",
    "data.duplicated().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "517d459a",
   "metadata": {},
   "source": [
    "- Dataset consists 45593 rows and 11 columns. Then the dataset also consists of 7 numerical data and 4 categorical data\n",
    "- The problem faced is a regression problem, namely predicting the time needed to deliver food (`Time taken(min)`)\n",
    "- There are no missing value and duplicated value in this dataset. However, it is necessary to do more checking for categorical data whether there are unreasonable values or not (There is a blank space or something like that)\n",
    "- It seems all numerical values contained in the dataset are quite reasonable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ac432c34",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7E+08    3\n",
      "7E+06    3\n",
      "7E+03    3\n",
      "5E+07    3\n",
      "6E+02    3\n",
      "        ..\n",
      "C3E      1\n",
      "4438     1\n",
      "A8B3     1\n",
      "8537     1\n",
      "5FB2     1\n",
      "Name: ID, Length: 45451, dtype: int64\n",
      "\t\n",
      "PUNERES01DEL01    67\n",
      "JAPRES11DEL02     67\n",
      "HYDRES04DEL02     66\n",
      "JAPRES03DEL01     66\n",
      "VADRES11DEL02     66\n",
      "                  ..\n",
      "DEHRES18DEL03      7\n",
      "AURGRES11DEL03     7\n",
      "KOLRES09DEL03      6\n",
      "KOCRES16DEL03      6\n",
      "BHPRES010DEL03     5\n",
      "Name: Delivery_person_ID, Length: 1320, dtype: int64\n",
      "\t\n",
      "Snack      11533\n",
      "Meal       11458\n",
      "Drinks     11322\n",
      "Buffet     11280\n",
      "Name: Type_of_order, dtype: int64\n",
      "\t\n",
      "motorcycle           26435\n",
      "scooter              15276\n",
      "electric_scooter      3814\n",
      "bicycle                 68\n",
      "Name: Type_of_vehicle, dtype: int64\n",
      "\t\n"
     ]
    }
   ],
   "source": [
    "# Retrieval all categorical data columns\n",
    "\n",
    "cat_col = [col_name for col_name in data.columns if data[col_name].dtypes == 'O']\n",
    "\n",
    "# Check all value in categorical data\n",
    "\n",
    "for col in cat_col:\n",
    "    print(data[col].value_counts())\n",
    "    print('\\t')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ccf9a91e",
   "metadata": {},
   "source": [
    "There are no odd values in all categorical data."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fad8a43f",
   "metadata": {},
   "source": [
    "To get the time needed to deliver food, the distance between the restaurant and the delivery location is needed.To get the required distance, [Haversine Formula](https://en.wikipedia.org/wiki/Haversine_formula) can be used to calculate the distance between 2 locations by utilizing longitudes and latitudes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "970951db",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set the earth's radius (in kilometers)\n",
    "R = 6371\n",
    "\n",
    "# Convert degrees to radians\n",
    "\n",
    "def deg_to_rad(degrees):\n",
    "    return degrees * (np.pi/180)\n",
    "\n",
    "# Function to calculate the distance between two points using the haversine formula\n",
    "\n",
    "def dist_calculate(lat1, lon1, lat2, lon2):\n",
    "    d_lat = deg_to_rad(lat2-lat1)\n",
    "    d_lon = deg_to_rad(lon2-lon1)\n",
    "    a = np.sin(d_lat/2)**2 + np.cos(deg_to_rad(lat1)) * np.cos(deg_to_rad(lat2)) * np.sin(d_lon/2)**2\n",
    "    c = 2 * np.arctan2(np.sqrt(a), np.sqrt(1-a))\n",
    "    return R * c\n",
    "\n",
    "# Calculate the distance between each pair of points\n",
    "\n",
    "data['Distance'] = np.nan\n",
    "\n",
    "for i in range(len(data)):\n",
    "    data.loc[i, 'Distance'] = dist_calculate(data.loc[i, 'Restaurant_latitude'], \n",
    "                                             data.loc[i, 'Restaurant_longitude'], \n",
    "                                             data.loc[i, 'Delivery_location_latitude'], \n",
    "                                             data.loc[i, 'Delivery_location_longitude'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aff899d4",
   "metadata": {},
   "source": [
    "Since dataset now have distance column, latitudes and longitudes data will be dropped."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4e9f8c31",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Delivery_person_ID</th>\n",
       "      <th>Delivery_person_Age</th>\n",
       "      <th>Delivery_person_Ratings</th>\n",
       "      <th>Distance</th>\n",
       "      <th>Type_of_order</th>\n",
       "      <th>Type_of_vehicle</th>\n",
       "      <th>Time_taken(min)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4607</td>\n",
       "      <td>INDORES13DEL02</td>\n",
       "      <td>37</td>\n",
       "      <td>4.9</td>\n",
       "      <td>3.025149</td>\n",
       "      <td>Snack</td>\n",
       "      <td>motorcycle</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>B379</td>\n",
       "      <td>BANGRES18DEL02</td>\n",
       "      <td>34</td>\n",
       "      <td>4.5</td>\n",
       "      <td>20.183530</td>\n",
       "      <td>Snack</td>\n",
       "      <td>scooter</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5D6D</td>\n",
       "      <td>BANGRES19DEL01</td>\n",
       "      <td>23</td>\n",
       "      <td>4.4</td>\n",
       "      <td>1.552758</td>\n",
       "      <td>Drinks</td>\n",
       "      <td>motorcycle</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7A6A</td>\n",
       "      <td>COIMBRES13DEL02</td>\n",
       "      <td>38</td>\n",
       "      <td>4.7</td>\n",
       "      <td>7.790401</td>\n",
       "      <td>Buffet</td>\n",
       "      <td>motorcycle</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>70A2</td>\n",
       "      <td>CHENRES12DEL01</td>\n",
       "      <td>32</td>\n",
       "      <td>4.6</td>\n",
       "      <td>6.210138</td>\n",
       "      <td>Snack</td>\n",
       "      <td>scooter</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     ID Delivery_person_ID  Delivery_person_Age  Delivery_person_Ratings  \\\n",
       "0  4607     INDORES13DEL02                   37                      4.9   \n",
       "1  B379     BANGRES18DEL02                   34                      4.5   \n",
       "2  5D6D     BANGRES19DEL01                   23                      4.4   \n",
       "3  7A6A    COIMBRES13DEL02                   38                      4.7   \n",
       "4  70A2     CHENRES12DEL01                   32                      4.6   \n",
       "\n",
       "    Distance Type_of_order Type_of_vehicle  Time_taken(min)  \n",
       "0   3.025149        Snack      motorcycle                24  \n",
       "1  20.183530        Snack         scooter                33  \n",
       "2   1.552758       Drinks      motorcycle                26  \n",
       "3   7.790401       Buffet      motorcycle                21  \n",
       "4   6.210138        Snack         scooter                30  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Remove unneeded columns\n",
    "data = data.drop(['Restaurant_latitude', 'Restaurant_longitude', \n",
    "                  'Delivery_location_latitude','Delivery_location_longitude'], axis=1)\n",
    "\n",
    "\n",
    "# Move distance column position\n",
    "data.insert(4, 'Distance', (data.pop('Distance')))\n",
    "\n",
    "# Take a look final dataset\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67391110",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f99f6962",
   "metadata": {},
   "source": [
    "In Exploratory Data Analysis (EDA), univariate, bivariate, and multivariate analysis will be carried out."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ede5fa7",
   "metadata": {},
   "source": [
    "### Univariate Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "174a345d",
   "metadata": {},
   "source": [
    "In univariate analysis, data will be analyzed using descriptive statistics like check central tendency (mean, median, mode), outliers and data distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d85bf6f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Group numerical and categorical data\n",
    "\n",
    "num_col = [col_name for col_name in data.columns if data[col_name].dtypes != 'O']\n",
    "cat_col = [col_name for col_name in data.columns if data[col_name].dtypes == 'O']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c30dc190",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Delivery_person_Age</th>\n",
       "      <th>Delivery_person_Ratings</th>\n",
       "      <th>Distance</th>\n",
       "      <th>Time_taken(min)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>45593.000000</td>\n",
       "      <td>45593.000000</td>\n",
       "      <td>45593.000000</td>\n",
       "      <td>45593.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>29.544075</td>\n",
       "      <td>4.632367</td>\n",
       "      <td>99.303911</td>\n",
       "      <td>26.294607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>5.696793</td>\n",
       "      <td>0.327708</td>\n",
       "      <td>1099.731281</td>\n",
       "      <td>9.383806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>15.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.465067</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>25.000000</td>\n",
       "      <td>4.600000</td>\n",
       "      <td>4.663493</td>\n",
       "      <td>19.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>29.000000</td>\n",
       "      <td>4.700000</td>\n",
       "      <td>9.264281</td>\n",
       "      <td>26.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>34.000000</td>\n",
       "      <td>4.800000</td>\n",
       "      <td>13.763977</td>\n",
       "      <td>32.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>19692.674606</td>\n",
       "      <td>54.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Delivery_person_Age  Delivery_person_Ratings      Distance  \\\n",
       "count         45593.000000             45593.000000  45593.000000   \n",
       "mean             29.544075                 4.632367     99.303911   \n",
       "std               5.696793                 0.327708   1099.731281   \n",
       "min              15.000000                 1.000000      1.465067   \n",
       "25%              25.000000                 4.600000      4.663493   \n",
       "50%              29.000000                 4.700000      9.264281   \n",
       "75%              34.000000                 4.800000     13.763977   \n",
       "max              50.000000                 6.000000  19692.674606   \n",
       "\n",
       "       Time_taken(min)  \n",
       "count     45593.000000  \n",
       "mean         26.294607  \n",
       "std           9.383806  \n",
       "min          10.000000  \n",
       "25%          19.000000  \n",
       "50%          26.000000  \n",
       "75%          32.000000  \n",
       "max          54.000000  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Numerical data\n",
    "\n",
    "data[num_col].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "de75c6cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Delivery_person_ID</th>\n",
       "      <th>Type_of_order</th>\n",
       "      <th>Type_of_vehicle</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>45593</td>\n",
       "      <td>45593</td>\n",
       "      <td>45593</td>\n",
       "      <td>45593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>45451</td>\n",
       "      <td>1320</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>7E+08</td>\n",
       "      <td>PUNERES01DEL01</td>\n",
       "      <td>Snack</td>\n",
       "      <td>motorcycle</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>3</td>\n",
       "      <td>67</td>\n",
       "      <td>11533</td>\n",
       "      <td>26435</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           ID Delivery_person_ID Type_of_order Type_of_vehicle\n",
       "count   45593              45593         45593           45593\n",
       "unique  45451               1320             4               4\n",
       "top     7E+08     PUNERES01DEL01        Snack      motorcycle \n",
       "freq        3                 67         11533           26435"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Categorical data\n",
    "\n",
    "data[cat_col].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ceba798",
   "metadata": {},
   "source": [
    "- It looks like `distance` has skew data distribution and has extreme outliers (the difference between the mean and the median is relatively large). For outliers will be checked further using boxplot and IQR, then for data distribution will be checked further using histogram\n",
    "- Mean and median value `Time_taken(min)` column relatively has same value(most likely to have normal data distribution)\n",
    "- Since `ID` and `Delivery_person_ID` are both identities, any analysis result will not gain any insights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "82c4aafe",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Detecting outliers using boxplot\n",
    "\n",
    "plt.figure(figsize=(8,4))\n",
    "for i in range(0, len(num_col)):\n",
    "    plt.subplot(1, len(num_col), i+1)\n",
    "    sns.boxplot(y = data[num_col[i]], color = '#1AD761')\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14168a7e",
   "metadata": {},
   "source": [
    "It looks like `Delivery_person_Ratings` and `distance` columns have extreme outliers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fec3366a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Max value in Delivery_person_Ratings column is 6.0\n",
      "Min value in Delivery_person_Ratings column is 1.0\n",
      "\t\n",
      "Max value in Distance column is 19692.67460647305\n",
      "Min value in Distance column is 1.4650674052308583\n",
      "\t\n"
     ]
    }
   ],
   "source": [
    "# Check Delivery_person_Ratings and distance maximum and minimum value\n",
    "\n",
    "column = ['Delivery_person_Ratings', 'Distance']\n",
    "\n",
    "for col in column:\n",
    "    print('Max value in {a} column is {b}'.format(a=col, b=data[col].max()))\n",
    "    print('Min value in {a} column is {b}'.format(a=col, b=data[col].min()))\n",
    "    print('\\t')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "36f7b361",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Outliers in Delivery_person_Ratings column are < 3.999999999999999 or > 5.2\n",
      "Outliers in Distance column are < -22.63795911360644 or > 31.964945393444165\n"
     ]
    }
   ],
   "source": [
    "# Check lower and upper boundary for Delivery_person_Ratings and distance column\n",
    "\n",
    "for col in column:\n",
    "    IQR = data[col].quantile(0.75) - data[col].quantile(0.25)\n",
    "    lower_fence = data[col].quantile(0.25) - (IQR * 3)\n",
    "    upper_fence = data[col].quantile(0.25) + (IQR * 3)\n",
    "    print('Outliers in {a} column are < {b} or > {c}'.format(a=col, b=lower_fence, c=upper_fence))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25aa2d81",
   "metadata": {},
   "source": [
    "Extreme outliers at `Delivery_person_Ratings` column are on lower boundary, meanwhile `distance` column have extreme outliers in upper boundary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d26e43bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAD5CAYAAAAa5CMeAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAluUlEQVR4nO3de5QW1Z3u8e8jIM1FQKFFoLmpxAsEYSBINBq8RTSZ4JzRCc6MOkfnkBhNzDqZS4zjGTJzWGuSk8QcczGD0RGNUVk6HkmiE/FCYgxqQBFBUIjcWq6iIKBNbPydP2o3/XbT/XYD/fJW089nrVpde9fe9e4qWP3rvXe9uxQRmJmZtbUjyt0AMzM7PDnAmJlZSTjAmJlZSTjAmJlZSTjAmJlZSTjAmJlZSbQYYCRVSHpB0suSlkr6RsqfLulNSYvSdnFBnRslrZT0mqQLC/LHSXolHbtVklJ+V0kPpPznJQ0rwbWamdkh1JoezG7g3Ig4DRgDTJY0MR27JSLGpO1RAEmnAlOBkcBk4EeSOqXytwHTgBFpm5zyrwHeiYgTgVuAbx70lZmZWVm1GGAiszMlu6St2LczpwD3R8TuiFgFrAQmSBoA9IqI+ZF9u/Nu4JKCOrPS/oPAeXW9GzMza586t6ZQ6oEsBE4EfhgRz0u6CLhe0pXAAuCrEfEOMAh4rqB6dcr7IO03zif9XAcQEbWStgN9gbeaa1O/fv1i2LBhrWm+mZmVyMKFC9+KiMqmjrUqwETEHmCMpD7Aw5JGkQ13/StZb+Zfge8AVwNN9TyiSD4tHNtL0jSyITaGDBnCggULWtN8M2uv3nwz+zmo7m/RlN77t6mVm6Q1zR3br6fIImIbMA+YHBGbImJPRHwI3A5MSMWqgcEF1aqA9Sm/qon8BnUkdQZ6A2838fkzI2J8RIyvrGwyYJrZ4eTFF7OtPiNt1h605imyytRzQVI34HxgeZpTqfNnwJK0PweYmp4MG042mf9CRGwAdkiamOZXrgQeKahzVdq/FHgqvAqnmVm71pohsgHArDQPcwQwOyJ+IekeSWPIhrJWA58HiIilkmYDrwK1wHVpiA3gWuAuoBvwWNoA7gDukbSSrOcy9eAvzczMyqnFABMRi4GxTeRfUaTODGBGE/kLgFFN5NcAl7XUFjOzUvrggw+orq6mpqam3E3JnYqKCqqqqujSpUur67Rqkt/MrCOorq7mqKOOYtiwYfibEvUigq1bt1JdXc3w4cNbXc8Bxszy66yzGmeU9ONqamocXJogib59+7Jly5b9qucAY2b51adP44ySf6SDS9MO5L54sUszy681a7KtPiNthy9JXHFF/RR3bW0tlZWVfOYznyljqw6MezBmVjYDBg9jY3XzAaPuV+ov6tIp4xe/aKr0/juuaigb1q1um5O1kR49erBkyRLef/99unXrxty5cxk0qH1+sdQBxszKZmP1GgbPerfZ45UvZd9kGDz2oixdldKXXdQmn7/uql5tcp62dtFFF/HLX/6SSy+9lPvuu4/LL7+cZ555BoBdu3bxpS99iVdeeYXa2lqmT5/OlClTWL16NVdccQW7du0C4Ac/+AFnnHEG8+bNY/r06fTr148lS5Ywbtw4fvrTnx6SoUAPkZmZ5czUqVO5//77qampYfHixZx++ul7j82YMYNzzz2X3//+9zz99NP8/d//Pbt27eLYY49l7ty5vPjiizzwwAN8+ctf3lvnpZde4nvf+x6vvvoqb7zxBs8+++whuQ73YMzMmvPzn++bd8IJcOqpUFsLjz227/GTToKPfARqamDu3IbH/vRPW/Wxo0ePZvXq1dx3331cfPHFDY49/vjjzJkzh29/+9tA9uTb2rVrGThwINdffz2LFi2iU6dOvP7663vrTJgwgaqqbKWuMWPGsHr1aj7xiU+0qi0HwwHGzCyHPvvZz/J3f/d3zJs3j61bt+7NjwgeeughTjrppAblp0+fTv/+/Xn55Zf58MMPqaio2Husa9eue/c7depEbW1t6S8ABxgzy7HfndLwey+/21ja78Hso1iPo3Pn4scrKlrdY2nK1VdfTe/evfnoRz/KvHnz9uZfeOGFfP/73+f73/8+knjppZcYO3Ys27dvp6qqiiOOOIJZs2axZ8+e5k9+iHgOxsxy672KnrxX0bM+XduT92p7Fqlx+KiqquKGG27YJ//mm2/mgw8+YPTo0YwaNYqbb74ZgC9+8YvMmjWLiRMn8vrrr9OjR49D3eR9qL0uWjx+/Pjw+2DM2jdJRZ8iG7rpDQDW9D8+S/dM6Z3Ht8nnr7uqF4W/A5ctW8Ypp5zSJuc+HDV1fyQtjIjxTZX3EJmZ5daI9a8B9QFmRJ+UbqMAY6XlITIzMysJBxgzMysJBxgzMysJBxgzMysJT/KbWW49M/Kchun15zRT0vLIPRgzy63dR1aw+8j6b6Tv/rCC3R9WFKnR/rXVcv2TJk2i7qscF198Mdu2bWvLZrZKiwFGUoWkFyS9LGmppG+k/GMkzZW0Iv08uqDOjZJWSnpN0oUF+eMkvZKO3aq0nKekrpIeSPnPSxpWgms1s3Zm+MYVDN+4oj591AqGH7WiSI32r3C5fqBNlut/9NFH6bPPy9tKrzU9mN3AuRFxGjAGmCxpIvA14MmIGAE8mdJIOhWYCowEJgM/ktQpnes2YBowIm2TU/41wDsRcSJwC/DNg780M2vvTtiwkhM2rKxP917JCb1XFqlxeKhbrh/Yu1x/nV27dnH11VfzsY99jLFjx/LII48A8P777zN16lRGjx7N5z73ub0BCmDYsGG89dZbAFxyySWMGzeOkSNHMnPmzL1levbsyU033cRpp53GxIkT2bRp00FfR4sBJjI7U7JL2gKYAsxK+bOAS9L+FOD+iNgdEauAlcAESQOAXhExP7Kvzt7dqE7duR4Ezqvr3ZiZdTQHslz/bbfdRvfu3Vm8eDE33XQTCxcubPLcd955JwsXLmTBggXceuutexfS3LVrFxMnTuTll1/m7LPP5vbbbz/o62jVJH/qgSwETgR+GBHPS+ofERsAImKDpGNT8UHAcwXVq1PeB2m/cX5dnXXpXLWStgN9gbcO6KrMzNpEE8v1cwJwKlALNLFcPycBHwFqgEbL9VO65fp/85vf7H0HzOjRoxk9enST57711lt5+OGHAVi3bh0rVqygb9++HHnkkXvnecaNG8fcxq8aOACtCjARsQcYI6kP8LCkUUWKN9XziCL5xeo0PLE0jWyIjSFDhhRrsplZu7a/y/UDLb6lct68eTzxxBPMnz+f7t27M2nSJGpqagDo0qXL3vpttaT/fj2mHBHbJM0jmzvZJGlA6r0MADanYtXA4IJqVcD6lF/VRH5hnWpJnYHewNtNfP5MYCZki13uT9vNzPZfsR5H5xaOV7RwvLj9Xa7/7LPP5t577+Wcc85hyZIlLF68eJ9zbt++naOPPpru3buzfPlynnvuuX3KtKXWPEVWmXouSOoGnA8sB+YAV6ViVwGPpP05wNT0ZNhwssn8F9Jw2g5JE9P8ypWN6tSd61LgqWivyzybWZt5evQFPD36gvr0mxfw9JsXFKlx+Njf5fqvvfZadu7cyejRo/nWt77FhAkT9qk7efJkamtrGT16NDfffDMTJ04s6TW0uFy/pNFkE/CdyALS7Ij4F0l9gdnAEGAtcFlEvJ3q3ARcTTZI+ZWIeCzljwfuArqRDV5+KSJCUgVwDzCWrOcyNSLeKNYuL9dv1v61tFx/qXm5/v3T5sv1R8Risl/8jfO3Auc1U2cGMKOJ/AXAPvM3EVEDXNZSW8ysYxnx5nIAVgw6OUv3TuntJ5etTdZ6/ia/meXW0M2rGLp5VX36qFUMPWpVkRqWJw4wZmZWEg4wZmYF/HxR0w7kvjjAmJklFRUVbN261UGmkYhg69atVFTs30KjXq7fzCypqqqiurqaLVu2lLspuVNRUUFVVVXLBQs4wJhZbj0x9qKG6eqLminZNrp06cLw4cNL+hkdiYfIzMysJBxgzCy3Tl67hJPXLqlP91nCyX2WFKlheeIAY2a5VbV1HVVb19Wne66jque6IjUsTxxgzMysJBxgzMysJBxgzMysJPyYspnlVu0RnRqmP+zUTEnLIwcYM8utead9qmF6/aeaKWl55CEyMzMrCQcYM8utUasXMWr1ovr0MYsYdcyiZstbvjjAmFluHffOBo57Z0N9uvsGjuu+oUgNyxMHGDMzKwkHGDMzK4kWA4ykwZKelrRM0lJJN6T86ZLelLQobRcX1LlR0kpJr0m6sCB/nKRX0rFbJSnld5X0QMp/XtKwElyrmZkdQq15TLkW+GpEvCjpKGChpLnp2C0R8e3CwpJOBaYCI4GBwBOSPhIRe4DbgGnAc8CjwGTgMeAa4J2IOFHSVOCbwOcO/vLMrD3b3aVrw/Sers2UtDxqMcBExAZgQ9rfIWkZMKhIlSnA/RGxG1glaSUwQdJqoFdEzAeQdDdwCVmAmQJMT/UfBH4gSeHXypl1aM+MOrdhesO5zZS0PNqvOZg0dDUWeD5lXS9psaQ7JR2d8gYBhcudVqe8QWm/cX6DOhFRC2wH+u5P28zMLF9aHWAk9QQeAr4SEe+SDXedAIwh6+F8p65oE9WjSH6xOo3bME3SAkkL/EpTs8PfaX9YwGl/WFCf7ruA0/ouKFLD8qRVAUZSF7Lgcm9E/CdARGyKiD0R8SFwOzAhFa8GBhdUrwLWp/yqJvIb1JHUGegNvN24HRExMyLGR8T4ysrK1l2hmbVble9uofLd+j8mK7ttobKb/7hsL1rzFJmAO4BlEfHdgvwBBcX+DKh7zdwcYGp6Mmw4MAJ4Ic3l7JA0MZ3zSuCRgjpXpf1Lgac8/2Jm1r615imyM4ErgFckLUp5XwculzSGbChrNfB5gIhYKmk28CrZE2jXpSfIAK4F7gK6kU3uP5by7wDuSQ8EvE32FJqZmbVjrXmK7Lc0PUfyaJE6M4AZTeQvAEY1kV8DXNZSW8zMrP3wcv1mllvvde3eMF3bvZmSlkcOMGaWW7879ZMN0xs/2UxJyyOvRWZmZiXhAGNmuTVuxfOMW/F8fbryecZVPl+khuWJh8jMLLeO3tnw63BHd93n63GWY+7BmJlZSTjAmJlZSTjAmJlZSXgOxsxy691uvRqm/9irmZKWRw4wZpZbL5x8ZsP05jObKWl55CEyMzMrCQcYM8utCcufZcLyZ+vTxz7LhGOfLVLD8sRDZGaWW73ef7dh+sh3mylpeeQejJmZlYQDjJmZlYQDjJmZlYTnYMwst97peUzD9O5jmilpeeQAY2a5tXDE6Q3TW05vpqTlkYfIzMysJBxgzCy3znj115zx6q/r08f9mjOO+3WRGpYnLQYYSYMlPS1pmaSlkm5I+cdImitpRfp5dEGdGyWtlPSapAsL8sdJeiUdu1WSUn5XSQ+k/OclDSvBtZpZO9N993t03/1efbrze3Tv/F6RGpYnrenB1AJfjYhTgInAdZJOBb4GPBkRI4AnU5p0bCowEpgM/EhSp3Su24BpwIi0TU751wDvRMSJwC3AN9vg2szMrIxaDDARsSEiXkz7O4BlwCBgCjArFZsFXJL2pwD3R8TuiFgFrAQmSBoA9IqI+RERwN2N6tSd60HgvLrejZmZtU/7NQeThq7GAs8D/SNiA2RBCDg2FRsErCuoVp3yBqX9xvkN6kRELbAd6NvE50+TtEDSgi1btuxP083M7BBr9WPKknoCDwFfiYh3i3QwmjoQRfKL1WmYETETmAkwfvz4fY6b2eFlS6/Khun3K5spaXnUqgAjqQtZcLk3Iv4zZW+SNCAiNqThr80pvxoYXFC9Clif8quayC+sUy2pM9AbePsArsfMDiMvnzC+YXrr+GZKWh615ikyAXcAyyLiuwWH5gBXpf2rgEcK8qemJ8OGk03mv5CG0XZImpjOeWWjOnXnuhR4Ks3TmJlZO9WaHsyZwBXAK5IWpbyvA/8GzJZ0DbAWuAwgIpZKmg28SvYE2nURsSfVuxa4C+gGPJY2yALYPZJWkvVcph7cZZnZ4eCsJU8B8Myoc7P0gJTecG7Z2mSt12KAiYjf0vQcCcB5zdSZAcxoIn8BMKqJ/BpSgDIzq9P1g90N0512N1PS8sjf5Dczs5JwgDEzs5JwgDEzs5Lwcv1mllsbjx7QMP3egGZKWh45wJhZbi0ZNqZh+u0xTZazfPIQmZmZlYQDjJnl1qSXH2fSy4/Xpwc+zqSBjxepYXniITIzy63OH+5pmD5iTzMlLY/cgzEzs5JwgDEzs5JwgDEzs5LwHIyZ5VZ138EN0zsHN1PS8sgBxsxya/mQhmvjLt+2z1q5lmMeIjMzs5JwgDGz3Dr/pcc4/6XH6tNVj3F+1WNFalieOMCYmVlJOMCYmVlJOMCYmVlJOMCYmVlJtBhgJN0pabOkJQV50yW9KWlR2i4uOHajpJWSXpN0YUH+OEmvpGO3SlLK7yrpgZT/vKRhbXyNZtZOrTl2OGuOHV6f3jGcNTuGF6lhedKaHsxdwOQm8m+JiDFpexRA0qnAVGBkqvMjSZ1S+duAacCItNWd8xrgnYg4EbgF+OYBXouZHWZWDDqZFYNOrk9vP5kV208uUsPypMUAExG/Ad5u5fmmAPdHxO6IWAWsBCZIGgD0ioj5ERHA3cAlBXVmpf0HgfPqejdm1rF12lNLpz219WnV0km1RWpYnhzMHMz1khanIbSjU94gYF1BmeqUNyjtN85vUCciaoHtQN+mPlDSNEkLJC3YsmXLQTTdzNqDcxbP5ZzFc+vTg+ZyzqC5RWpYnhxogLkNOAEYA2wAvpPym+p5RJH8YnX2zYyYGRHjI2J8ZWXlfjXYzMwOrQMKMBGxKSL2RMSHwO3AhHSoGihcja4KWJ/yq5rIb1BHUmegN60fkjMzs5w6oACT5lTq/BlQ94TZHGBqejJsONlk/gsRsQHYIWliml+5EnikoM5Vaf9S4Kk0T2NmZu1Yi6spS7oPmAT0k1QN/DMwSdIYsqGs1cDnASJiqaTZwKtALXBdRNS94/RasifSugGPpQ3gDuAeSSvJei5T2+C6zMyszFoMMBFxeRPZdxQpPwOY0UT+AmCftbYjoga4rKV2mFnH84cBJzZMbz+xmZKWR34fjJnl1qrjRjRM7xjRTEnLIy8VY2a51fWPNXT9Y019+ogauh5RU6SG5YkDjJnl1llLn+aspU/Xpwc+zVkDny5Sw/LEAcbMzErCAcbMzErCAcbMzErCAcbMzErCjymbWW6tGHhSw/S2k5opaXnkAGNmubWm//EN0zuPb6ak5ZGHyMwst7rX7KR7zc76dOeddO+8s0gNyxMHGDPLrTOWPcMZy56pTx/3DGcc90yRGpYnDjBmZlYSDjBmZlYSDjBmZlYSDjBmZlYSfkzZzHJr2eCRDdPvjGympOWRA4yZ5dab/YY0TO8a0kxJyyMPkZlZbvXatY1eu7bVp7tso1eXbc2Wt3xxgDGz3Jrw+nwmvD6/Pt1/PhP6zy9Sw/KkxQAj6U5JmyUtKcg7RtJcSSvSz6MLjt0oaaWk1yRdWJA/TtIr6ditkpTyu0p6IOU/L2lYG1+jmZmVQWt6MHcBkxvlfQ14MiJGAE+mNJJOBaYCI1OdH0nqlOrcBkwDRqSt7pzXAO9ExInALcA3D/RizMwsP1oMMBHxG+DtRtlTgFlpfxZwSUH+/RGxOyJWASuBCZIGAL0iYn5EBHB3ozp153oQOK+ud2NmZu3Xgc7B9I+IDQDp57EpfxCwrqBcdcoblPYb5zeoExG1wHag7wG2y8zMcqKtH1NuqucRRfKL1dn35NI0smE2hgzx44pmh7tXhp7WML31tGZKWh4daA9mUxr2Iv3cnPKrgcEF5aqA9Sm/qon8BnUkdQZ6s++QHAARMTMixkfE+MrKygNsupm1F5uOGcimYwbWp98fyKb3BxapYXlyoAFmDnBV2r8KeKQgf2p6Mmw42WT+C2kYbYekiWl+5cpGderOdSnwVJqnMbMOrs/OrfTZubU+3XUrfbpuLVLD8qTFITJJ9wGTgH6SqoF/Bv4NmC3pGmAtcBlARCyVNBt4FagFrouIPelU15I9kdYNeCxtAHcA90haSdZzmdomV2Zm7d74FS8A8MTYi7J0ZUpXX1S2NlnrtRhgIuLyZg6d10z5GcCMJvIXAKOayK8hBSgzMzt8+Jv8ZmZWEg4wZmZWEg4wZmZWEl6u38xya9Hx4xqm3xrXTEnLIwcYM8utt3of2zBdc2wzJS2PPERmZrnVb/tm+m3fXJ+u2Ey/is1FalieOMCYWW6NeWMhY95YWJ/ut5Ax/RYWqWF54gBjZmYl4QBjZmYl4QBjZmYl4QBjZmYl4ceUzSy3FoyY0DC9ZUIzJS2PHGDMLLe29Wz4ctttu/2y2/bEAcasAxsweBgbq9eUuxnN6v929l7CupeO9e+W0n7pWLvgAGPWgW2sXsPgWe+W7fPXXdWr6PGPrnkZqA8wH+2b0tUOMO2BJ/nNzKwkHGDMzKwkHGDMzKwkHGDMzKwkDmqSX9JqYAewB6iNiPGSjgEeAIYBq4G/iIh3UvkbgWtS+S9HxK9S/jjgLqAb8ChwQ0TEwbTNzNq/Fz7y8YbpTR9vpqTlUVv0YM6JiDERMT6lvwY8GREjgCdTGkmnAlOBkcBk4EeSOqU6twHTgBFpm9wG7TKzdu7dHn14t0ef+vQHfXj3gz7Nlrd8KcUQ2RRgVtqfBVxSkH9/ROyOiFXASmCCpAFAr4iYn3otdxfUMbMObNBbaxn01tr6dI+1DOqxtkgNy5OD/R5MAI9LCuDfI2Im0D8iNgBExAZJda+gGwQ8V1C3OuV9kPYb55tZB3fKuqUAvNlvSJY+OqV3DSlbm6z1DjbAnBkR61MQmStpeZGyaiIviuTvewJpGtlQGkOG+D+YmVmeHdQQWUSsTz83Aw8DE4BNadiL9LPu/abVwOCC6lXA+pRf1UR+U583MyLGR8T4ysrKg2m6mZmV2AEHGEk9JB1Vtw98ClgCzAGuSsWuAh5J+3OAqZK6ShpONpn/QhpO2yFpoiQBVxbUMTOzdupghsj6Aw9nMYHOwM8i4r8k/R6YLekaYC1wGUBELJU0G3gVqAWui4g96VzXUv+Y8mNpMzOzduyAA0xEvAGc1kT+VuC8ZurMAGY0kb8AGHWgbTGzw9PvTjmrYXrjWc2UtDzyaspmllvvVfRsmK7t2UzJA9S5K2kUpmyOqxrKhnWry9qGUnGAMbPcGrrpDQDW9D8+S/dM6Z3Ht80H1O4u6+sKoOVXFrRnDjBmllsj1r8G1AeYEX1Suq0CjJWUF7s0M7OScIAxM7OScIAxM7OScIAxM7OS8CS/meXWMyPPaZhef04zJS2PHGDMLLd2H1nRMP1hRTMlLY88RGZmuTV84wqGb1xRnz5qBcOPWlGkhuWJezBmllsnbFgJwKrjRmTp3im9Y0TZ2mSt5x6MmZmVhAOMmZmVhAOMmZmVhAOMmZmVhCf5zSy3nh59QcP0mxc0U9LyyAHGzHJrT6eGv6L2hH9ltSf+1zIrowGDh7Gxek25m5FbI95cDsCKQSdn6d4pvf3ksrXJWs8BxqyMNlavKesLr/L+squhm1cB9QFm6FEp7QDTLuRmkl/SZEmvSVop6Wvlbo+ZmR2cXAQYSZ2AHwIXAacCl0s6tbytMjOzg5GLAANMAFZGxBsR8UfgfmBKmdtkZmYHIS9zMIOAdQXpauD0MrXFDpFyT3B36tqdPbvfK9vnmx3uFBHlbgOSLgMujIi/TekrgAkR8aVG5aYB01LyJOC1g/jYfsBbB1H/cNDR70FHv37wPejo1w8Hfw+GRkRlUwfy0oOpBgYXpKuA9Y0LRcRMYGZbfKCkBRExvi3O1V519HvQ0a8ffA86+vVDae9BXuZgfg+MkDRc0pHAVGBOmdtkZmYHIRc9mIiolXQ98CugE3BnRCwtc7PMzOwg5CLAAETEo8Cjh/Aj22SorZ3r6Pego18/+B509OuHEt6DXEzym5nZ4ScvczBmZnaY6XABRtKdkjZLWlLutpSDpMGSnpa0TNJSSTeUu02HmqQKSS9Iejndg2+Uu03lIKmTpJck/aLcbSkHSaslvSJpkaQF5W5POUjqI+lBScvT74SPt+n5O9oQmaSzgZ3A3RExqtztOdQkDQAGRMSLko4CFgKXRMSrZW7aISNJQI+I2CmpC/Bb4IaIeK7MTTukJP1PYDzQKyI+U+72HGqSVgPjI6LDfg9G0izgmYj4SXqCt3tEbGur83e4HkxE/AZ4u9ztKJeI2BARL6b9HcAyspUUOozI7EzJLmnrUH9pSaoCPg38pNxtsfKQ1As4G7gDICL+2JbBBTpggLF6koYBY4Hny9yUQy4NDy0CNgNzI6Kj3YPvAf8AfFjmdpRTAI9LWphWCelojge2AP+Rhkp/IqlHW36AA0wHJakn8BDwlYgo3wtJyiQi9kTEGLJVIyZI6jDDpZI+A2yOiIXlbkuZnRkRf0K2ivt1afi8I+kM/AlwW0SMBXYBbfqqFAeYDijNOzwE3BsR/1nu9pRTGhKYB0wub0sOqTOBz6Y5iPuBcyX9tLxNOvQiYn36uRl4mGxV946kGqgu6L0/SBZw2owDTAeTJrjvAJZFxHfL3Z5ykFQpqU/a7wacDywva6MOoYi4MSKqImIY2bJMT0XEX5e5WYeUpB7pIRfSsNCngA71ZGlEbATWSTopZZ0HtOnDPrn5Jv+hIuk+YBLQT1I18M8RcUd5W3VInQlcAbyS5iAAvp5WUugoBgCz0ovujgBmR0SHfFS3A+sPPJz9vUVn4GcR8V/lbVJZfAm4Nz1B9gbw39vy5B3uMWUzMzs0PERmZmYl4QBjZmYl4QBjZmYl4QBjZmYl4QBjZmYl4QBjuSNpUnqEvC69VNKk8rXImiPpryQ9Xu52WD45wFhJpKXQ35e0Q9I2Sb+T9AVJ+/1/LiJGRsS8EjSzw5F0l6Q/Stop6W1JcyWd3Mq6wySFpL3fn4uIeyPiU6VrsbVnDjBWSn8aEUcBQ4F/A/6RtHJruRX+kuyAbfhWRPQkW0X7TXLyb2KHHwcYK7mI2B4Rc4DPAVdJGiWpq6RvS1oraZOkH6dlW/aRekPnSxqYekXHFBwbK+mttL4akq5OL056R9KvJA0tKBuSrpO0Algh6YeSvtPos34u6SvFrie150ZJr6bP+Q9JFQXHP5NeYlXXcxvdqO4/SloM7JLUOaXfTL291ySdl8p2lfQ9SevT9j1JXdOxSZKqJX1V2Qv0Nkjar29hR8T7wGxgTEH7Pp1W1n1X0jpJ0wuq/Cb93JZ6QB+X9DeSftvoHn9B0op0b36YlieqW8H6O+nfa5Wk6wt7ROlcb6T7sErSX+3P9Vj+OMDYIRMRL5AtsHcW8E3gI2S/3E4k+2v6f7VQfz0wH/jzguy/BB6MiA8kXQJ8HfhvQCXwDHBfo9NcApwOnArMAi6vG7aT1I9sPabGdZryV8CFwAnpOv4pneNPgDuBzwN9gX8H5tQFhuRysnex9En1rwc+lnp7FwKrU7mbgIlk9+g0ssUY/6ngPMcBvcnu3TXADyUd3Yq2k9raI7VlZUH2LuDK1LZPA9em+wrZu0MA+kREz4iY38ypPwN8LLX5L9I1AfwPspWLx5Atqlh33rq23ApclO7DGcCi1l6L5VREePPW5hvZL8nzm8h/juwX5y7ghIL8jwOr0v4kslVe9zkX8LdkizMCCFgHnJ3SjwHXFNQ7AngPGJrSAZzbqD3LgAvS/vXAo628ti8UpC8G/pD2bwP+tVH514BPFtS9uuDYiWTvpDkf6NKo3h+AiwvSFwKrC+7R+0DnguObgYkttP0uoAbYRvYumFXA6CLlvwfckvaHpXtY+Jl/A/y2IB3AJwrSs4Gvpf2ngM8XHDu/7nxAj9SmPwe6lfv/r7e22dyDsUNtENkvlO7AwjSMtA34L7JeR0seBD4uaSDZX9RB1lOBbK7n/xac822yIFT4xs51jc43C6hbSfivgXtaeR2F51kDDCxow1fr2pDaMbjgeIO6EbES+AowHdgs6f50baQ6a5r5HICtEVFbkH4P6NmKtn87IvqQBYz3gbrVdJF0uqSnJW2RtB34AtCvFecstLGZNg2k4X0rvA+7yIZQvwBskPTL1j58YPnlAGOHjKSPkf2y/39kv9hGRkSftPWObOK5qMje3/I42dDLXwL3RfpzmOwX1ucLztknIrpFxO8KT9HolD8Fpkg6DTglta01BhfsDwHWF7RhRqM2dI+IwmG3Bm2IiJ9FxCfIglOQDR+Szjm0oGjh5xy0iFgL3EAWlOvmv34GzAEGR0Rv4MdkQXqfdh+ADWQveKtTeA+JiF9FxAVkq10vB24/yM+zMnOAsZKT1EvZWxTvB34aES+T/fK4RdKxqcwgSRcWO0+Bn5HNE/x52q/zY+BGSSPTOXtLuqzYiSKiGvg9Wc/locgmvlvjOklV6YGDrwMPpPzbgS+knoCUvXfk00rvHmlM0kmSzk1zNDVkgXdPOnwf8E/K3l/Tj2yOqk1fDBYRc8mCVt0rg48C3o6IGkkTyIJ4nS1kw2rHH+DHzQZuSP/WfcieKgRAUn9Jn01zMbuBndTfB2unHGCslH4uaQfZX/U3Ad+l/n0T/0g2ufycpHeBJygYqmnBHGAEsCkFKwAi4mGyv/7vT+dcQjap3JJZwEdp/fAYZIHtcbJ3aLwB/O/UhgVkk9k/AN4hu8a/KXKermSPcL9FNrR0LFnAIp1zAbAYeAV4se5z2tj/Af4hBbkvAv+S/t3+F1lQACAi3gNmAM+m4b+J+/k5t5Pds8XAS8CjQC1ZIDkC+CpZsHsb+GRqi7Vjfh+MdXjK3sX+U2BYRHzYivKrgb+NiCdK3bbDmaSLgB9HxNAWC1u75B6MdWjKvj9zA/CT1gQXO3CSukm6OH33ZxDwz8DD5W6XlY4DjHVYkk4hezR2ANnjuHX5Q9IXCZvahpSpua2mbO22ptpe7i8uCvgG2dDhS2SPiBf97pO1bx4iMzOzknAPxszMSsIBxszMSsIBxszMSsIBxszMSsIBxszMSsIBxszMSuL/A8JcyT8b7GQ1AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Data distribution check\n",
    "\n",
    "for i in range(0, len(num_col)):\n",
    "    plt.figure(figsize=(14,8))\n",
    "    plt.subplot(2, len(num_col)//2, i+1)\n",
    "    plt.hist(data[num_col[i]], color='#1DA1F2', edgecolor = 'Black')\n",
    "    plt.axvline(x=data[num_col[i]].mean(),ls='--',color='red',alpha=0.4,label='Mean')\n",
    "    plt.axvline(x=data[num_col[i]].median(),ls='--',color='yellow',alpha=0.4,label='Median')\n",
    "    plt.xlabel(num_col[i], size=12)\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ecc2c424",
   "metadata": {},
   "source": [
    "- `Delivery_person_Age` and `Time_taken(min)` column have relatively normal data distribution\n",
    "- `Delivery_person_Ratings` has negative skew data distribution, meanwhile `distance` column has positive skew data distribution"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47d79478",
   "metadata": {},
   "source": [
    "### Bivariate Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "012032bf",
   "metadata": {},
   "source": [
    "In bivariate analysis, relationships between data features and target (time taken) will be explored."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "1227436c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Relationship between distance with time taken\n",
    "\n",
    "sns.lmplot(x='Distance', y=\"Time_taken(min)\",\n",
    "           line_kws={'color': 'red'}, data=data)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f6c8f6e",
   "metadata": {},
   "source": [
    "There is consistent relationship between the time taken and the distance travelled to deliver the food. It looks like majority food delivered within 25-27 minutes regardless of distance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "033cbcfd",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Relationship between age of the delivery partner with time taken\n",
    "\n",
    "sns.lmplot(x='Delivery_person_Age', y=\"Time_taken(min)\",\n",
    "           line_kws={'color': 'red'}, data=data)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c89a368d",
   "metadata": {},
   "source": [
    "There is a linear relationship between the time taken to deliver the food and the age of the person who delivering the food. It looks like person with the young age able to take less time than person with old age to deliver the food to customers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d40849a4",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Relationship between rating with time taken\n",
    "\n",
    "sns.lmplot(x='Delivery_person_Ratings', y=\"Time_taken(min)\",\n",
    "           line_kws={'color': 'red'}, data=data)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b7e30eb",
   "metadata": {},
   "source": [
    "There is an inverse linear relationship between the time taken to deliver the food and the delivery person ratings. It looks like person with the higher ratings take a less time to deliver the food than person with low ratings."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a36f1504",
   "metadata": {},
   "source": [
    "Check if type of food ordered by the customer and the type of vehicle used by the delivery partner affects the delivery time or not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "dc52211e",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Boxplot for type of food ordered, type of vehicle, and time taken\n",
    "\n",
    "plt.figure(figsize=(14,6))\n",
    "sns.boxplot(x=\"Type_of_vehicle\", \n",
    "            y=\"Time_taken(min)\", hue=\"Type_of_order\",data=data)\n",
    "plt.legend(loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f40fa59",
   "metadata": {},
   "source": [
    "It looks like there is not much difference between the time taken depending on the vehicle they are driving and the type of food they are delivering."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2538bae3",
   "metadata": {},
   "source": [
    "### Multivariate Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71c5bdf1",
   "metadata": {},
   "source": [
    "In multivariate analysis, Numerical features relationship strengths will be checked by heatmap correlation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "733eb5c1",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Heatmap for the correlation data\n",
    "\n",
    "correlation = data.corr()\n",
    "\n",
    "plt.figure(figsize = (8, 8))\n",
    "sns.heatmap(correlation, annot=True, cmap='BuGn')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88cb63c8",
   "metadata": {},
   "source": [
    "- There are no strong correlation between features (correlation < 0.8)\n",
    "- The highest correlation with `Time_taken(min)` column is `Delivery_person_Age` with -0.33 (negative means inverse relationship)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12bc3304",
   "metadata": {},
   "source": [
    "### Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f3772e1",
   "metadata": {},
   "source": [
    "From the results of the exploratory data analysis that was carried out, it was concluded that the factors that most influence the time of delivering food is ratings of the delivery person with inverse relationship."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86a990df",
   "metadata": {},
   "source": [
    "## Feature Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db8c432b",
   "metadata": {},
   "source": [
    "`ID` and `Delivery_person_ID` columns are both identities and will not be used for modeling (both columns will be dropped)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "d8b58363",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 45593 entries, 0 to 45592\n",
      "Data columns (total 6 columns):\n",
      " #   Column                   Non-Null Count  Dtype  \n",
      "---  ------                   --------------  -----  \n",
      " 0   Delivery_person_Age      45593 non-null  int64  \n",
      " 1   Delivery_person_Ratings  45593 non-null  float64\n",
      " 2   Distance                 45593 non-null  float64\n",
      " 3   Type_of_order            45593 non-null  object \n",
      " 4   Type_of_vehicle          45593 non-null  object \n",
      " 5   Time_taken(min)          45593 non-null  int64  \n",
      "dtypes: float64(2), int64(2), object(2)\n",
      "memory usage: 2.1+ MB\n"
     ]
    }
   ],
   "source": [
    "# Drop unused columns\n",
    "\n",
    "data = data.drop(['ID', 'Delivery_person_ID'], axis=1)\n",
    "\n",
    "# Check info\n",
    "\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07eeb69c",
   "metadata": {},
   "source": [
    "### Outlier Handling"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63e84d68",
   "metadata": {},
   "source": [
    "Based on outlier check, outliers in `Delivery_person_Ratings` and `distance` needs to removed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "7f3f076d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove outliers based on max value\n",
    "\n",
    "def max_value(df1, variable, top):\n",
    "    return np.where(df1[variable]>top, top, df1[variable])\n",
    "\n",
    "for df1 in [data]:\n",
    "  df1['Distance'] = max_value(df1, 'Distance', 31.964945393444165)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "041c9f90",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove outliers based on min value\n",
    "\n",
    "def min_value(df2, variable, low):\n",
    "    return np.where(df2[variable]<low, low, df2[variable])\n",
    "\n",
    "for df2 in [data]:\n",
    "  df2['Delivery_person_Ratings'] = min_value(df2, 'Delivery_person_Ratings', 3.999999999999999)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "540453c1",
   "metadata": {},
   "source": [
    "### One Hot Encoding"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88a6a753",
   "metadata": {},
   "source": [
    "`Type_of_order` and `Type_of_vehicle` columns are categorical data. Both columns have to transform into numerical data so that it can be used for the modeling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "1b44f14d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Delivery_person_Age</th>\n",
       "      <th>Delivery_person_Ratings</th>\n",
       "      <th>Distance</th>\n",
       "      <th>Time_taken(min)</th>\n",
       "      <th>Type_of_order_Buffet</th>\n",
       "      <th>Type_of_order_Drinks</th>\n",
       "      <th>Type_of_order_Meal</th>\n",
       "      <th>Type_of_order_Snack</th>\n",
       "      <th>Type_of_vehicle_bicycle</th>\n",
       "      <th>Type_of_vehicle_electric_scooter</th>\n",
       "      <th>Type_of_vehicle_motorcycle</th>\n",
       "      <th>Type_of_vehicle_scooter</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>37</td>\n",
       "      <td>4.9</td>\n",
       "      <td>3.025149</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>34</td>\n",
       "      <td>4.5</td>\n",
       "      <td>20.183530</td>\n",
       "      <td>33</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>23</td>\n",
       "      <td>4.4</td>\n",
       "      <td>1.552758</td>\n",
       "      <td>26</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>38</td>\n",
       "      <td>4.7</td>\n",
       "      <td>7.790401</td>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>32</td>\n",
       "      <td>4.6</td>\n",
       "      <td>6.210138</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Delivery_person_Age  Delivery_person_Ratings   Distance  Time_taken(min)  \\\n",
       "0                   37                      4.9   3.025149               24   \n",
       "1                   34                      4.5  20.183530               33   \n",
       "2                   23                      4.4   1.552758               26   \n",
       "3                   38                      4.7   7.790401               21   \n",
       "4                   32                      4.6   6.210138               30   \n",
       "\n",
       "   Type_of_order_Buffet   Type_of_order_Drinks   Type_of_order_Meal   \\\n",
       "0                      0                      0                    0   \n",
       "1                      0                      0                    0   \n",
       "2                      0                      1                    0   \n",
       "3                      1                      0                    0   \n",
       "4                      0                      0                    0   \n",
       "\n",
       "   Type_of_order_Snack   Type_of_vehicle_bicycle   \\\n",
       "0                     1                         0   \n",
       "1                     1                         0   \n",
       "2                     0                         0   \n",
       "3                     0                         0   \n",
       "4                     1                         0   \n",
       "\n",
       "   Type_of_vehicle_electric_scooter   Type_of_vehicle_motorcycle   \\\n",
       "0                                  0                            1   \n",
       "1                                  0                            0   \n",
       "2                                  0                            1   \n",
       "3                                  0                            1   \n",
       "4                                  0                            0   \n",
       "\n",
       "   Type_of_vehicle_scooter   \n",
       "0                         0  \n",
       "1                         1  \n",
       "2                         0  \n",
       "3                         0  \n",
       "4                         1  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# One Hot Encoding for type of order and type of vehicle\n",
    "\n",
    "data = pd.get_dummies(data)\n",
    "\n",
    "# Take a look at data\n",
    "\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "6e156830",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 45593 entries, 0 to 45592\n",
      "Data columns (total 12 columns):\n",
      " #   Column                             Non-Null Count  Dtype  \n",
      "---  ------                             --------------  -----  \n",
      " 0   Delivery_person_Age                45593 non-null  int64  \n",
      " 1   Delivery_person_Ratings            45593 non-null  float64\n",
      " 2   Distance                           45593 non-null  float64\n",
      " 3   Time_taken(min)                    45593 non-null  int64  \n",
      " 4   Type_of_order_Buffet               45593 non-null  uint8  \n",
      " 5   Type_of_order_Drinks               45593 non-null  uint8  \n",
      " 6   Type_of_order_Meal                 45593 non-null  uint8  \n",
      " 7   Type_of_order_Snack                45593 non-null  uint8  \n",
      " 8   Type_of_vehicle_bicycle            45593 non-null  uint8  \n",
      " 9   Type_of_vehicle_electric_scooter   45593 non-null  uint8  \n",
      " 10  Type_of_vehicle_motorcycle         45593 non-null  uint8  \n",
      " 11  Type_of_vehicle_scooter            45593 non-null  uint8  \n",
      "dtypes: float64(2), int64(2), uint8(8)\n",
      "memory usage: 1.7 MB\n"
     ]
    }
   ],
   "source": [
    "# General info\n",
    "\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2af75a93",
   "metadata": {},
   "source": [
    "### Split Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "afe6c346",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split train-test data into proportion 80:20\n",
    "\n",
    "X = data.drop(columns='Time_taken(min)')\n",
    "y = data['Time_taken(min)']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, \n",
    "                                                    test_size=0.20, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a522e6b",
   "metadata": {},
   "source": [
    "### Multicollinearity Check"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "f00b7d26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>feature</th>\n",
       "      <th>vif score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Delivery_person_Age</td>\n",
       "      <td>1.010354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Delivery_person_Ratings</td>\n",
       "      <td>1.016234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Distance</td>\n",
       "      <td>1.005852</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   feature  vif score\n",
       "1      Delivery_person_Age   1.010354\n",
       "2  Delivery_person_Ratings   1.016234\n",
       "3                 Distance   1.005852"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Multicollinearity check using vif score\n",
    "\n",
    "x = add_constant(X_train[['Delivery_person_Age', 'Delivery_person_Ratings', 'Distance']])\n",
    "df_vif = pd.DataFrame([vif(x.values, i) for i in range(x.shape[1])],\n",
    "                      index = x.columns).reset_index()\n",
    "df_vif.columns = ['feature', 'vif score']\n",
    "df_vif = df_vif.loc[df_vif.feature != 'const']\n",
    "df_vif"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96dbd161",
   "metadata": {},
   "source": [
    "There are no multicollinearity since the vif score is below than 10."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa2449fe",
   "metadata": {},
   "source": [
    "## Modeling and Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9f25841",
   "metadata": {},
   "source": [
    "Since this is a regression problem, regression algorithms will be used for model training. 5 algorithms will be used for model training. The [RMSE](https://en.wikipedia.org/wiki/Root-mean-square_deviation) score is a reference for how well the model performs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "e08ed50f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create rmse calculation\n",
    "\n",
    "def rmse(targets,predictions):\n",
    "    differences = predictions - targets\n",
    "    differences_squared = differences ** 2\n",
    "    mean_of_differences_squared = differences_squared.mean()\n",
    "    rmse_val = np.sqrt(mean_of_differences_squared)\n",
    "    return rmse_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "f2858034",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[LinearRegression(),\n",
       " Ridge(),\n",
       " Lasso(),\n",
       " DecisionTreeRegressor(),\n",
       " RandomForestRegressor()]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load regression algorithms\n",
    "\n",
    "regressors = [LinearRegression(), Ridge(), Lasso(), \n",
    "              DecisionTreeRegressor(), RandomForestRegressor()]\n",
    "\n",
    "regressors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "36937900",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create dataframe for model performance evaluation\n",
    "\n",
    "df = pd.DataFrame(columns = ['Algorithm', 'RMSE'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "625884e7",
   "metadata": {},
   "source": [
    "### Algorithm Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "43efd331",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.\n",
      "The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.\n",
      "The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.\n",
      "The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.\n",
      "The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.\n"
     ]
    }
   ],
   "source": [
    "# Model training\n",
    "\n",
    "for regressor in regressors:\n",
    "    regressor.fit(X_train, y_train)\n",
    "    y_pred = regressor.predict(X_test)\n",
    "    model = str(type(regressor)).split('.')[-1][:-2]\n",
    "    \n",
    "    # RMSE value\n",
    "    \n",
    "    rmse_value = rmse(y_test, y_pred)\n",
    "    \n",
    "    # Input calculation result into a dataframe\n",
    "    \n",
    "    df = df.append({'Algorithm': model,\n",
    "                    'RMSE': rmse_value},\n",
    "                     ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "7053fa69",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Algorithm</th>\n",
       "      <th>RMSE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LinearRegression</td>\n",
       "      <td>7.927093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ridge</td>\n",
       "      <td>7.927084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Lasso</td>\n",
       "      <td>8.518059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DecisionTreeRegressor</td>\n",
       "      <td>10.291782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>RandomForestRegressor</td>\n",
       "      <td>7.775025</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Algorithm       RMSE\n",
       "0       LinearRegression   7.927093\n",
       "1                  Ridge   7.927084\n",
       "2                  Lasso   8.518059\n",
       "3  DecisionTreeRegressor  10.291782\n",
       "4  RandomForestRegressor   7.775025"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Models evaluation\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0694e3b",
   "metadata": {},
   "source": [
    "- Based on each RMSE (Root Squared Mean Error) result from 5 trained algorithms, best algorithm is random forest regressor because have the smallest error than other algorithms\n",
    "- Performance of the model will try to be improved with hyperparameter tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fef1c94c",
   "metadata": {},
   "source": [
    "### Hyperparameter Tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52144dba",
   "metadata": {},
   "source": [
    "Hyperparameter tuning using GridSearchCV library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "9acacb50",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, estimator=RandomForestRegressor(),\n",
       "             param_grid={'bootstrap': ('True', 'False'),\n",
       "                         'max_depth': (10, 20, 30, 40, 50),\n",
       "                         'n_estimators': (100, 200, 300, 400, 500)},\n",
       "             scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Set parameter for tuning\n",
    "\n",
    "parameters = {'n_estimators': (100, 200, 300, 400, 500),\n",
    "              'max_depth':(10, 20, 30, 40, 50),\n",
    "              'bootstrap':('True', 'False')\n",
    "             }\n",
    "\n",
    "rf = RandomForestRegressor()\n",
    "\n",
    "grid_search_forest = GridSearchCV(rf, parameters, cv=5, scoring='neg_mean_squared_error')\n",
    "grid_search_forest.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "b21d5dfc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters: {'bootstrap': 'True', 'max_depth': 10, 'n_estimators': 300}\n",
      "\n",
      "\n",
      "Best estimator: RandomForestRegressor(bootstrap='True', max_depth=10, n_estimators=300)\n"
     ]
    }
   ],
   "source": [
    "# Get the best parameters and estimator\n",
    "\n",
    "print(\"Best parameters:\", grid_search_forest.best_params_)\n",
    "print('\\n')\n",
    "print(\"Best estimator:\", grid_search_forest.best_estimator_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "9c56fd82",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE:  7.281613193317689\n"
     ]
    }
   ],
   "source": [
    "# Trained model using best parameter\n",
    "\n",
    "model = grid_search_forest.best_estimator_\n",
    "model.fit(X_train, y_train)\n",
    "y_pred = model.predict(X_test)\n",
    "\n",
    "# Evaluate tuned model\n",
    "\n",
    "rmse_value = rmse(y_test, y_pred)\n",
    "print('RMSE: ', rmse_value)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c857ddd0",
   "metadata": {},
   "source": [
    "### Feature Importances"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb6f3a88",
   "metadata": {},
   "source": [
    "All features will be calculated to check the importance. Higher score means that the specific feature will have a larger effect on the model that is being used to predict a certain variable. To calculate feature importances, [SHAP](https://towardsdatascience.com/using-shap-values-to-explain-how-your-machine-learning-model-works-732b3f40e137) method will be used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "b3ff11e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create function\n",
    "\n",
    "def print_feature_importances_random_forest(random_forest_model):\n",
    "    \n",
    "    '''\n",
    "    Prints the feature importances of a Random Forest model in an ordered way.\n",
    "    '''\n",
    "    \n",
    "    # Fetch the feature importances and feature names\n",
    "    \n",
    "    importances = random_forest_model.feature_importances_\n",
    "    features = random_forest_model.feature_names_in_\n",
    "    \n",
    "    # Organize them in a dictionary\n",
    "    \n",
    "    feature_importances = {fea: imp for imp, fea in zip(importances, features)}\n",
    "    \n",
    "    # Sorts the dictionary\n",
    "    \n",
    "    feature_importances = {k: v for k, v in sorted(feature_importances.items(), key=lambda item: item[1], reverse = True)}\n",
    "    \n",
    "    # Prints the feature importances\n",
    "    \n",
    "    for k, v in feature_importances.items():\n",
    "        print(f\"{k} -> {v:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "38d8437c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Delivery_person_Ratings -> 0.4235\n",
      "Distance -> 0.3111\n",
      "Delivery_person_Age -> 0.1838\n",
      "Type_of_vehicle_motorcycle  -> 0.0603\n",
      "Type_of_order_Drinks  -> 0.0046\n",
      "Type_of_order_Meal  -> 0.0046\n",
      "Type_of_order_Buffet  -> 0.0045\n",
      "Type_of_order_Snack  -> 0.0043\n",
      "Type_of_vehicle_scooter  -> 0.0018\n",
      "Type_of_vehicle_electric_scooter  -> 0.0009\n",
      "Type_of_vehicle_bicycle  -> 0.0008\n"
     ]
    }
   ],
   "source": [
    "# Calculate feature importance score\n",
    "\n",
    "print_feature_importances_random_forest(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "9f37c6bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Permutation explainer: 9120it [4:05:16,  1.61s/it]                                                                     \n"
     ]
    }
   ],
   "source": [
    "# Calculate SHAP values\n",
    "\n",
    "explainer = shap.Explainer(model.predict, X_test)\n",
    "shap_values = explainer(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "a1c315ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x424.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Feature importance visualization\n",
    "\n",
    "shap.summary_plot(shap_values)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dcac6d4",
   "metadata": {},
   "source": [
    "`Delivery_person_Ratings` variable become the most importance feature in this model, followed by `Distance` and `Delivery_person_age` variables. Then, we can see `Delivery_person_Ratings` variable has a negative contribution when its values are high, and a positive contribution on low values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "1c840f54",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# Save Model\n",
    "\n",
    "with open('Deliver_Predict','wb') as r:\n",
    "    pickle.dump(model,r)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd4bb5f2",
   "metadata": {},
   "source": [
    "## Conclusion and Recommendation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82e11a5a",
   "metadata": {},
   "source": [
    "### Conclusion\n",
    "\n",
    "1. Rating of person in previous delivers become is the most influential factor on the delivery time of food to the destination location. Person with the higher ratings take a less time to deliver the food than person with low ratings\n",
    "2. Model has RMSE score 7.28 and that means error between delivery time prediction and delivery time actual is 7.28 minutes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "97d9737c",
   "metadata": {},
   "source": [
    "### Recommendation\n",
    "\n",
    "The rating obtained by the deliveryman is a representation of the deliveryman's performance in delivering food to the intended location in terms of delivery time. Of course this is a potential loss of customers if this continues to happen. Delivery time performance needs to be maintained so that the rating obtained is high and customer trust can still be maintained.\n",
    "The RMSE value of the model can be used as a guarantee of delivery time performance which can be given to the customer so that as much as possible the delivery time is not more than the existing RMSE score (delay in delivery time of not more than 7.28 minutes)."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}