jupyter/display-comparison-kpis.jupyter.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Comparing solutions to Deterministic, Robust and Stochastic formulations of Unit Commitment (UC) \n",
    "\n",
    "This notebook implements some charts that compare the different solution obtained from the three different formulations, over a set of 7 daily instances of UC.\n",
    "\n",
    "- [Comparing production costs between Robust and Stochastic formulations](#sav)\n",
    "- [Comparing renewable energy utilization between Robust and Stochastic formulations](#ren)\n",
    "- [Heatmap comparing solution performance of Robust, Stochastic and Deterministic formulations](#heat)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Comparing production costs between Robust and Stochastic formulations\n",
    "\n",
    "<a id=sav ></a>\n",
    "\n",
    "The following chart reports the daily cost savings (in Euros) that can be obtained by adopting the robust optimization formulation in place of the stochastic formulation. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Absolute Average Daily Cost Saving: 849.2K Euros\n",
      "Relative Average Daily Cost Saving: 3.7%\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkIAAAJmCAYAAACwmu8gAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwGTpCAEAk6UjBABMlo4QADBZ/w/GfuZaoRQBSwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f72cdc77c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import os, pandas as pd, numpy as np\n",
    "\n",
    "from IPython.display import display\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "import matplotlib.cm as cm\n",
    "\n",
    "\n",
    "df_dashboard = pd.read_csv(os.environ['DSX_PROJECT_DIR']+'/datasets/df_dashboard.csv')\n",
    "\n",
    "### comparing daily costs between the solutions to robust and stochastic formulations\n",
    "### the comparison returns the pontetial daily cost saving (it could be negative) if the robust formulation is adopted\n",
    "do_cost_comp = df_dashboard.copy()[['day', 'model', 'cost']]\n",
    "do_cost_comp = do_cost_comp.loc[do_cost_comp['model'] != 'D']\n",
    "do_cost_comp.drop_duplicates(inplace=True)\n",
    "do_cost_comp_R = do_cost_comp.copy().loc[do_cost_comp['model'] == 'R']\n",
    "do_cost_comp_R.rename(columns={'cost': 'cost_R'}, inplace=True)\n",
    "do_cost_comp_R.drop('model', axis=1, inplace=True)\n",
    "do_cost_comp = do_cost_comp.loc[do_cost_comp['model'] == 'S']\n",
    "do_cost_comp = do_cost_comp.merge(do_cost_comp_R, on='day', how='inner')\n",
    "do_cost_comp_X = do_cost_comp.copy()\n",
    "do_cost_comp['saving'] = do_cost_comp['cost'] - do_cost_comp['cost_R']\n",
    "\n",
    "do_cost_comp.to_csv(os.environ['DSX_PROJECT_DIR']+'/datasets/df_cost_comp.csv', index=False)\n",
    "\n",
    "relative_avg_saving = np.mean(do_cost_comp['saving'] / do_cost_comp['cost'])\n",
    "absolute_avg_saving = do_cost_comp['saving'].mean()\n",
    "\n",
    "do_cost_comp = do_cost_comp[['day', 'saving']]\n",
    "do_cost_comp.set_index('day', inplace=True)\n",
    "\n",
    "### potentail daily savings are displayed with bar chart\n",
    "fig = plt.figure(figsize=(7, 7), dpi= 100)\n",
    "cmap = mcolors.LinearSegmentedColormap.from_list(\"\", [\"AquaMarine\", \"DeepSkyBlue\", \"Navy\"])\n",
    "plt.bar(do_cost_comp.index, do_cost_comp['saving'], color=cmap((do_cost_comp['saving'].values - do_cost_comp['saving'].values.min()) / do_cost_comp['saving'].values.max()))\n",
    "plt.xticks(do_cost_comp.index, do_cost_comp.index.tolist())\n",
    "plt.yticks([])\n",
    "plt.xlabel('Days')\n",
    "plt.title('Daily Cost Savings (Euros)')\n",
    "normalize = mcolors.Normalize(vmin=do_cost_comp['saving'].values.min(), vmax=do_cost_comp['saving'].values.max())\n",
    "scalarmappaple = cm.ScalarMappable(norm=normalize, cmap=cmap)\n",
    "scalarmappaple.set_array(do_cost_comp['saving'])\n",
    "legend_ticks = [do_cost_comp['saving'].values.min(), do_cost_comp['saving'].values.mean(), do_cost_comp['saving'].values.max()]\n",
    "cbar = plt.colorbar(scalarmappaple, ticks=legend_ticks)\n",
    "cbar.ax.set_yticklabels([\"{:.1f}K\".format(float(tick) / 1000) for tick in legend_ticks])\n",
    "\n",
    "print(\"Absolute Average Daily Cost Saving: {:.1f}K Euros\".format(absolute_avg_saving / 1000))\n",
    "print(\"Relative Average Daily Cost Saving: {:.1f}%\".format(relative_avg_saving * 100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Comparing renewable energy utilization between Robust and Stochastic formulations\n",
    "\n",
    "<a id=ren ></a>\n",
    "\n",
    "The following chart reports the daily relative gap of the renewable utilization between the robust and the stochastic formulations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Daily Renewable Utilization Gain: 5.1%\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlUAAAJmCAYAAAB16HWxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f72cd911050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn import preprocessing\n",
    "\n",
    "### comparing renewable utilization rate between the solutions to robust and stochastic formulations\n",
    "### the comparison is based on the ratio between renewable and total energy production (these values are normalized in order to get a relative gap)\n",
    "### the potential daily gain in renewable utilization (it could be negative) is returned if the robust formulation is adopted\n",
    "do_ren_comp = df_dashboard.copy()[['day', 'model', 'thermal_prod', 'renewable_prod']]\n",
    "do_ren_comp = do_ren_comp.loc[do_ren_comp['model'] != 'D']\n",
    "do_ren_comp = do_ren_comp.groupby(['day', 'model']).sum()\n",
    "do_ren_comp['ren_rate'] = do_ren_comp['renewable_prod'] / (do_ren_comp['thermal_prod'] + do_ren_comp['renewable_prod'])\n",
    "\n",
    "x = do_ren_comp[['ren_rate']].values.astype(float)\n",
    "min_max_scaler = preprocessing.MinMaxScaler()\n",
    "x_scaled = min_max_scaler.fit_transform(x)\n",
    "do_ren_comp['ren_rate'] = x_scaled\n",
    "\n",
    "ref_df = do_ren_comp.copy()\n",
    "ref_df.reset_index(inplace=True)\n",
    "ref_df = ref_df[['day', 'model', 'ren_rate']]\n",
    "ref_df = ref_df.loc[ref_df['model'] == 'R']\n",
    "ref_df.rename(columns={'ren_rate': 'ref'}, inplace=True)\n",
    "ref_df.drop('model', axis=1, inplace=True)\n",
    "do_ren_comp.reset_index(inplace=True)\n",
    "do_ren_comp = do_ren_comp.merge(ref_df, on='day', how='inner')\n",
    "do_ren_comp = do_ren_comp.loc[do_ren_comp['model'] == 'S']\n",
    "do_ren_comp['ren_rate'] = do_ren_comp.apply(lambda x: x.ref - x.ren_rate, axis=1)\n",
    "do_ren_comp = do_ren_comp[['day', 'ren_rate']]\n",
    "do_ren_comp.set_index('day', inplace=True)\n",
    "\n",
    "do_ren_comp.to_csv(os.environ['DSX_PROJECT_DIR']+'/datasets/df_ren_comp.csv')\n",
    "\n",
    "avg_ren_util_gain = do_ren_comp['ren_rate'].mean()\n",
    "\n",
    "### potential daily gains in renewable utilization are displayed with a line chart\n",
    "fig, ax = plt.subplots(figsize=(7, 7), dpi= 100)\n",
    "ax.plot(do_ren_comp.index, do_ren_comp['ren_rate'], color='ForestGreen')\n",
    "plt.xticks(do_ren_comp.index, do_ren_comp.index.tolist())\n",
    "max_y_tick = 1 + int(np.ceil(do_ren_comp['ren_rate'].values.max() * 100))\n",
    "plt.yticks([float(x) / 100 for x in range(0, max_y_tick)], [\"{}%\".format(tick) if tick != 0 else 0 for tick in range(0, max_y_tick)])\n",
    "plt.fill_between(do_ren_comp.index, do_ren_comp['ren_rate'], alpha=0.4, color='ForestGreen')\n",
    "plt.xlabel('Days')\n",
    "plt.title('Daily Renewable Utilization Gain')\n",
    "\n",
    "print(\"Average Daily Renewable Utilization Gain: {:.1f}%\".format(avg_ren_util_gain * 100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Heatmap comparing solution performance of Robust, Stochastic and Deterministic formulations\n",
    "\n",
    "<a id=heat ></a>\n",
    "\n",
    "The heatmap shows how three models preform in each day, according to the following criteria: the comparison is based on an ex-post analysis that captures the performance of the optimal plan recommended by each model at the beginning of a day, against the actual energy demand registered at the end of the same day.  \n",
    "The color scale from blue to red illustrates the proximity of planned energy production and system reserve compared to the actual demand. In particular, purple/blue means that the total planned power is very close to the actual demand load registered, and lower and upper reserve levels are suitably centered respect to it. Red/dark red means that the total planned power and related reserve levels turn out to be unbalanced respect to the actual load registered: this requires suitable tactical adjustments to mitigate the deviation from the plan, which lead to higher operational costs at decreased sustainability. From the chart you can see that, for each day of the testbed, the robust formulation provided the best energy production plan against the actual demand, whereas the deterministic formulation’s solution led to the highest deviation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "### we use a heatmap to compare the solution quality of the three formulation\n",
    "### the kpi used measures the proximity of the recommended plan respect to the actual energy demand registered for each day (the higher value, the higher model performance)\n",
    "### kpi values have been normalized to achieve the comparison \n",
    "df_actual_demand = pd.read_csv(os.environ['DSX_PROJECT_DIR']+'/datasets/actual_demand.csv')\n",
    "df_dashboard = df_dashboard.merge(df_actual_demand, on='period', how='inner')\n",
    "df_dashboard['kpi'] = df_dashboard.apply(lambda x: 0.3 * abs(x.PROD_TOT - x.actual_demand) + \\\n",
    "                                         0.7 * (abs(2 * x.PROD_TOT - x.RES_DOWN_TOT - x.actual_demand) if 2 * x.PROD_TOT - x.RES_DOWN_TOT > x.actual_demand \\\n",
    "                                                    else (abs(2 * x.PROD_TOT + x.RES_UP_TOT - x.actual_demand) if 2 * x.PROD_TOT + x.RES_UP_TOT < x.actual_demand \\\n",
    "                                                          else 0)), axis=1)\n",
    "\n",
    "df_dashboard = df_dashboard[['model', 'day', 'period', 'Date', 'Hour', 'kpi', 'actual_demand']]\n",
    "df_kpi = df_dashboard[['model', 'day', 'kpi']].copy()\n",
    "df_kpi = df_kpi.groupby(['model', 'day']).mean()\n",
    "df_kpi.reset_index(inplace=True)\n",
    "\n",
    "x = df_kpi[['kpi']].values.astype(float)\n",
    "min_max_scaler = preprocessing.MinMaxScaler()\n",
    "x_scaled = min_max_scaler.fit_transform(x)\n",
    "df_kpi['kpi'] = x_scaled\n",
    "df_kpi['kpi'] = 1 - df_kpi['kpi']\n",
    "\n",
    "df_kpi.to_csv(os.environ['DSX_PROJECT_DIR']+'/datasets/df_heatmap_comp.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "### the implementation of the heatmap is based on this resource: https://matplotlib.org/gallery/images_contours_and_fields/image_annotated_heatmap.html\n",
    "from mpl_toolkits.axes_grid1 import make_axes_locatable\n",
    "\n",
    "def heatmap(data, row_labels, col_labels, ax=None,\n",
    "            cbar_kw={}, cbarlabel=\"\", **kwargs):\n",
    "    \"\"\"\n",
    "    Create a heatmap from a numpy array and two lists of labels.\n",
    "\n",
    "    Arguments:\n",
    "        data       : A 2D numpy array of shape (N,M)\n",
    "        row_labels : A list or array of length N with the labels\n",
    "                     for the rows\n",
    "        col_labels : A list or array of length M with the labels\n",
    "                     for the columns\n",
    "    Optional arguments:\n",
    "        ax         : A matplotlib.axes.Axes instance to which the heatmap\n",
    "                     is plotted. If not provided, use current axes or\n",
    "                     create a new one.\n",
    "        cbar_kw    : A dictionary with arguments to\n",
    "                     :meth:`matplotlib.Figure.colorbar`.\n",
    "        cbarlabel  : The label for the colorbar\n",
    "    All other arguments are directly passed on to the imshow call.\n",
    "    \"\"\"\n",
    "\n",
    "    if not ax:\n",
    "        ax = plt.gca()\n",
    "\n",
    "    # Plot the heatmap\n",
    "    im = ax.imshow(data, **kwargs)\n",
    "\n",
    "    # Create colorbar\n",
    "    divider = make_axes_locatable(ax)\n",
    "    cax = divider.append_axes(\"right\", size=\"1%\", pad=0.1)\n",
    "    cbar = ax.figure.colorbar(im, ax=ax, cax=cax, **cbar_kw)\n",
    "    #cbar.ax.set_yticklabels(['0', '', '', '',, '', '1'])\n",
    "    cbar.ax.set_ylabel(cbarlabel, rotation=0, labelpad=2, va=\"bottom\")\n",
    "\n",
    "    # We want to show all ticks...\n",
    "    ax.set_xticks(np.arange(data.shape[1]))\n",
    "    ax.set_yticks(np.arange(data.shape[0]))\n",
    "    # ... and label them with the respective list entries.\n",
    "    ax.set_xticklabels(col_labels)\n",
    "    ax.set_yticklabels(row_labels)\n",
    "\n",
    "    # Let the horizontal axes labeling appear on top.\n",
    "    ax.tick_params(top=False, bottom=True,\n",
    "                   labeltop=False, labelbottom=True)\n",
    "\n",
    "    # Rotate the tick labels and set their alignment.\n",
    "    plt.setp(ax.get_xticklabels(), rotation=30, ha=\"right\",\n",
    "             rotation_mode=\"anchor\")\n",
    "\n",
    "    # Turn spines off and create white grid.\n",
    "    for edge, spine in ax.spines.items():\n",
    "        spine.set_visible(False)\n",
    "\n",
    "    ax.set_xticks(np.arange(data.shape[1]+1)-.5, minor=True)\n",
    "    ax.set_yticks(np.arange(data.shape[0]+1)-.5, minor=True)\n",
    "    ax.grid(which=\"minor\", color=\"w\", linestyle='-', linewidth=3)\n",
    "    ax.tick_params(which=\"minor\", bottom=False, left=False)\n",
    "\n",
    "    return im, cbar\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Solution Quality Gain: 45.7%\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKYAAAH1CAYAAADMPPYYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzMzMzMzMzMysEv8Dc9DYo2B9PuwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7305994dd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_heatmap = pd.DataFrame(data={'model':['R', 'S', 'D']})\n",
    "df_heatmap.set_index('model', inplace=True)\n",
    "days = df_kpi['day'].unique().tolist()\n",
    "for day in days:\n",
    "    for model in df_heatmap.index.values.tolist():\n",
    "        df_heatmap.at[model, day] = df_kpi.loc[(df_kpi['model'] == model) & (df_kpi['day'] == day), 'kpi'].values[0]\n",
    "        \n",
    "fig, ax = plt.subplots(figsize=(12, 12), dpi= 100)\n",
    "im, cbar = heatmap(df_heatmap.values, ['Robust', 'Stochastic', 'Deterministic'], [\"day {}\".format(d) for d in df_kpi['day'].values], ax=ax,\n",
    "                   cmap=\"Spectral\", cbarlabel=\"\")\n",
    "\n",
    "fig.tight_layout()\n",
    "\n",
    "avg_sol_quality_gain = np.concatenate((df_heatmap.loc['R'] - df_heatmap.loc['S'], df_heatmap.loc['R'] - df_heatmap.loc['D'])).mean()\n",
    "print(\"Average Solution Quality Gain: {:.1f}%\".format(avg_sol_quality_gain * 100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Author\n",
    "\n",
    "- __Gianmaria Leo__ is a Senior Operations Research Engineer and Data Scientist with Data Science Elite team (IBM Analytics).\n",
    "\n",
    "Copyright © IBM Corp. 2018. Released as licensed Sample Materials."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python2.7 with DSX Spark 2.0.2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}