kpi-calculation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction to MindSphere Analytics APIs - KPI Calculation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This demo will use the mindconnect-nodejs CLI to call the mindsphere APIs. \n",
    "The information about the installation and usage of CLI:\n",
    "\n",
    "https://opensource.mindsphere.io/docs/mindconnect-nodejs/cli/index.html\n",
    "\n",
    "\n",
    "You will need to get and configure the service credentials to use this notebook. \n",
    "Here is a description how to acquire and setup the service credentials:\n",
    "\n",
    "https://opensource.mindsphere.io/docs/mindconnect-nodejs/li/setting-up-the-cli.html\n",
    "\n",
    "This is only a **demo** for the MindSphere KPI calculation API \n",
    "\n",
    "https://developer.mindsphere.io/apis/analytics-kpicalculation/api-kpicalculation-overview.html\n",
    "\n",
    "\n",
    "\n",
    "## Setup the mindconnect-nodejs CLI\n",
    "\n",
    "please change the line below to the passkey you were using when setting up the service credentials:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "passkey = \"passkey.4.unit.test\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## KPI Calculation with CLI\n",
    "\n",
    "```mc kpi-calculation``` CLI command:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Usage: mc kpi-calculation|kp [options]\n",
      "\n",
      "calculate kpi states or compute kpis @\n",
      "\n",
      "Options:\n",
      "  -f, --file <timeseries>            timeseries file (default:\n",
      "                                     \"timeseries-sample.json\")\n",
      "  -c, --calendar <calendar>          timeseries with planned outages\n",
      "  -e, --events <events>              timeseries with control system events\n",
      "  -m, --mode [states|kpis]           mode see @ Additional Documentation\n",
      "  -t, --target <target>              target variable\n",
      "  -n, --initialstate <initialstate>  Initial state [RSH, SH, POH, FOH]\n",
      "                                     (default: \"RSH\")\n",
      "  -d, --defaultstate <defaultstate>  Default state [RSH, FOH] (default: \"FOH\")\n",
      "  -h, --threshold <threshold>        threshold parameter (default: 1)\n",
      "  -s, --shutdown <shutdown>          shutdown threshold parameter in\n",
      "                                     milliseconds (default: 5000)\n",
      "  -y, --retry <number>               retry attempts before giving up (default:\n",
      "                                     3)\n",
      "  -p, --passkey <passkey>            passkey\n",
      "  -v, --verbose                      verbose output\n",
      "  -h, --help                         output usage information\n",
      "\n",
      "  Examples:\n",
      "\n",
      "    mc kpi-calculation --mode states --file timeseries.json --calendar calendar.json  --target Temperature --threshold 30 \t calculate kpi states based on temperature\n",
      "    mc kpi-calculation --mode kpis --file blubb.json --target state  \t calculate kpis for state timeseries\n",
      "\n",
      "  State KPIs:\n",
      "\n",
      "    No Data Hours (NoData), Period Hours (PH), Available Hours (AH) Service Hours (SH) \n",
      "    Reserve Shutdown Hours (RSH), Unavailble Hours (UH), Planned Outage Hours (POH) Forced Outage Hours (FOH) \n",
      "\n",
      "  Additional Documentation:\n",
      "\n",
      "    https://developer.mindsphere.io/apis/analytics-kpicalculation/api-kpicalculation-basics-kpi.html\n",
      "\n",
      "  Important: \n",
      "\n",
      "    you need to supply the service credentials for this operation and provide the passkey \n",
      "\n",
      "    how to get service credentials: \n",
      "    https://developer.mindsphere.io/howto/howto-selfhosted-api-access.html#creating-service-credentials\n",
      "\n",
      "  More Information: \n",
      "\n",
      "    https://opensource.mindsphere.io\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!mc kpi-calculation --help"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup the jupyter notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import json\n",
    "from pandas import json_normalize\n",
    "plt.rcParams['figure.figsize'] = [6, 6]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculating KPIs\n",
    "\n",
    "This uses the example request from the [documentation](https://developer.mindsphere.io/apis/analytics-kpicalculation/api-kpicalculation-samples.html) for calculation\n",
    "\n",
    "![](https://developer.mindsphere.io/apis/analytics-kpicalculation/images/Use-case-1-Example.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>_time</th>\n",
       "      <th>kpistatus</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-05-04 00:00:00+00:00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-05-04 00:30:00+00:00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-05-04 01:00:00+00:00</td>\n",
       "      <td>UNKNOWN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-05-04 01:30:00+00:00</td>\n",
       "      <td>UNKNOWN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-05-04 02:00:00+00:00</td>\n",
       "      <td>UNKNOWN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2017-05-04 02:30:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2017-05-04 03:00:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2017-05-04 03:30:00+00:00</td>\n",
       "      <td>POH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2017-05-04 04:00:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2017-05-04 04:30:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2017-05-04 05:00:00+00:00</td>\n",
       "      <td>POH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2017-05-04 05:30:00+00:00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2017-05-04 06:00:00+00:00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2017-05-04 06:30:00+00:00</td>\n",
       "      <td>UNKNOWN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2017-05-04 07:00:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2017-05-04 07:30:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2017-05-04 08:00:00+00:00</td>\n",
       "      <td>POH</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       _time kpistatus\n",
       "0  2017-05-04 00:00:00+00:00       NaN\n",
       "1  2017-05-04 00:30:00+00:00       NaN\n",
       "2  2017-05-04 01:00:00+00:00   UNKNOWN\n",
       "3  2017-05-04 01:30:00+00:00   UNKNOWN\n",
       "4  2017-05-04 02:00:00+00:00   UNKNOWN\n",
       "5  2017-05-04 02:30:00+00:00        SH\n",
       "6  2017-05-04 03:00:00+00:00        SH\n",
       "7  2017-05-04 03:30:00+00:00       POH\n",
       "8  2017-05-04 04:00:00+00:00        SH\n",
       "9  2017-05-04 04:30:00+00:00        SH\n",
       "10 2017-05-04 05:00:00+00:00       POH\n",
       "11 2017-05-04 05:30:00+00:00       NaN\n",
       "12 2017-05-04 06:00:00+00:00       NaN\n",
       "13 2017-05-04 06:30:00+00:00   UNKNOWN\n",
       "14 2017-05-04 07:00:00+00:00       FOH\n",
       "15 2017-05-04 07:30:00+00:00       FOH\n",
       "16 2017-05-04 08:00:00+00:00       POH"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_json(\"./data/kpis/sample-states.json\")\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Example KPIs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23cd69829b0>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "!mc kpi-calculation \\\n",
    "--mode kpis \\\n",
    "--file \"data/kpis/sample-states.json\" \\\n",
    "--target kpistatus \\\n",
    "--initialstate RSH \\\n",
    "--passkey $passkey \\\n",
    "> temp/kpi-sample.json\n",
    "\n",
    "with open('temp/kpi-sample.json') as f:\n",
    "  data = json.load(f)\n",
    "\n",
    "df = json_normalize (data).transpose()\n",
    "df[0:8].plot(kind=\"bar\", title=\"availability\")\n",
    "df[9:].plot(kind=\"bar\", title=\"factor\")\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Determining Unit States from Sensor Control Unit and Calendar Data¶\n",
    "\n",
    "This uses the example request from the [documentation](https://developer.mindsphere.io/apis/analytics-kpicalculation/api-kpicalculation-samples.html#determining-unit-states-from-sensor-control-unit-and-calendar-data) for calculation\n",
    "\n",
    "This sample provides a generic procedure for determining the state of a compressor with a pressure sensor, events from the compressor control unit, and a maintenance calendar.\n",
    "                                                       \n",
    "![](https://developer.mindsphere.io/apis/analytics-kpicalculation/images/Use-case-2-Example.png)\n",
    "\n",
    "## Calculate the states\n",
    "\n",
    "In this section the [KPI state calculations](https://developer.mindsphere.io/apis/analytics-kpicalculation/api-kpicalculation-basics-kpi-states.html) are calculated with measurement data taken from correctly formatted JSON files.\n",
    "* **Sensor Readings** are taken from `sample-timeseries.json` file with the `sensor` column used as the target variable. Sensor readings must be in numeric form.\n",
    "* **Calendar Entries** are taken from `calendar.json` file.\n",
    "* **Control System Events** are taken from `sample-events.json` file."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Timeseries data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>_time</th>\n",
       "      <th>sensor</th>\n",
       "      <th>sensor1</th>\n",
       "      <th>sensor2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-01-01 00:10:00+00:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01-01 00:50:00+00:00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-01-01 01:00:00+00:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-01-01 01:30:00+00:00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-01-01 02:00:00+00:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2017-01-01 03:00:00+00:00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2017-01-01 03:30:00+00:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2017-01-01 04:30:00+00:00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2017-01-01 05:10:00+00:00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2017-01-01 05:30:00+00:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2017-01-01 05:50:00+00:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2017-01-01 06:30:00+00:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2017-01-01 07:10:00+00:00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2017-01-01 07:20:00+00:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       _time  sensor  sensor1  sensor2\n",
       "0  2017-01-01 00:10:00+00:00     2.0      2.0      NaN\n",
       "1  2017-01-01 00:50:00+00:00     1.0      NaN      NaN\n",
       "2  2017-01-01 01:00:00+00:00     2.0      NaN      NaN\n",
       "3  2017-01-01 01:30:00+00:00     1.0      NaN      NaN\n",
       "4  2017-01-01 02:00:00+00:00     2.0      NaN      NaN\n",
       "5  2017-01-01 03:00:00+00:00     1.0      NaN      NaN\n",
       "6  2017-01-01 03:30:00+00:00     2.0      NaN      NaN\n",
       "7  2017-01-01 04:30:00+00:00     1.0      NaN      NaN\n",
       "8  2017-01-01 05:10:00+00:00     1.0      NaN      NaN\n",
       "9  2017-01-01 05:30:00+00:00     NaN      1.0      NaN\n",
       "10 2017-01-01 05:50:00+00:00     2.0      NaN      NaN\n",
       "11 2017-01-01 06:30:00+00:00     NaN      NaN      3.0\n",
       "12 2017-01-01 07:10:00+00:00     1.0      NaN      NaN\n",
       "13 2017-01-01 07:20:00+00:00     2.0      NaN      NaN"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_json(\"./data/kpis/sample-timeseries.json\")\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "calendar data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>PlannedOutage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{'from': '2017-01-01T10:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>{'from': '2017-01-01T06:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>{'from': '2017-01-01T00:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>{'from': '2017-01-01T04:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>{'from': '2017-01-01T02:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>{'from': '2017-01-01T08:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>{'from': '2016-12-31T22:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>{'from': '2017-01-01T03:00:00.000Z', 'to': '20...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       PlannedOutage\n",
       "0  {'from': '2017-01-01T10:00:00.000Z', 'to': '20...\n",
       "1  {'from': '2017-01-01T06:00:00.000Z', 'to': '20...\n",
       "2  {'from': '2017-01-01T00:00:00.000Z', 'to': '20...\n",
       "3  {'from': '2017-01-01T04:00:00.000Z', 'to': '20...\n",
       "4  {'from': '2017-01-01T02:00:00.000Z', 'to': '20...\n",
       "5  {'from': '2017-01-01T08:00:00.000Z', 'to': '20...\n",
       "6  {'from': '2016-12-31T22:00:00.000Z', 'to': '20...\n",
       "7  {'from': '2017-01-01T03:00:00.000Z', 'to': '20..."
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "calendar = pd.read_json(\"./data/kpis/calendar.json\")\n",
    "calendar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>_time</th>\n",
       "      <th>type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-01-01 01:29:57+00:00</td>\n",
       "      <td>NORMAL_STOP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01-01 01:30:00+00:00</td>\n",
       "      <td>SHUTDOWN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-01-01 04:30:01+00:00</td>\n",
       "      <td>SHUTDOWN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      _time         type\n",
       "0 2017-01-01 01:29:57+00:00  NORMAL_STOP\n",
       "1 2017-01-01 01:30:00+00:00     SHUTDOWN\n",
       "2 2017-01-01 04:30:01+00:00     SHUTDOWN"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "events = pd.read_json(\"./data/kpis/sample-events.json\")\n",
    "events"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>_time</th>\n",
       "      <th>state</th>\n",
       "      <th>source</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-01-01 00:10:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "      <td>SIGNAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01-01 00:50:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-01-01 01:00:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "      <td>SIGNAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-01-01 01:30:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-01-01 02:00:00+00:00</td>\n",
       "      <td>POH</td>\n",
       "      <td>CALENDAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2017-01-01 03:00:00+00:00</td>\n",
       "      <td>POH</td>\n",
       "      <td>CALENDAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2017-01-01 04:00:00+00:00</td>\n",
       "      <td>POH</td>\n",
       "      <td>CALENDAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2017-01-01 05:00:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2017-01-01 05:10:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2017-01-01 05:30:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2017-01-01 05:50:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "      <td>SIGNAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2017-01-01 06:00:00+00:00</td>\n",
       "      <td>POH</td>\n",
       "      <td>CALENDAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2017-01-01 07:00:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "      <td>SIGNAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2017-01-01 07:10:00+00:00</td>\n",
       "      <td>FOH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2017-01-01 07:20:00+00:00</td>\n",
       "      <td>SH</td>\n",
       "      <td>SIGNAL</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       _time state    source\n",
       "0  2017-01-01 00:10:00+00:00    SH    SIGNAL\n",
       "1  2017-01-01 00:50:00+00:00   FOH   DEFAULT\n",
       "2  2017-01-01 01:00:00+00:00    SH    SIGNAL\n",
       "3  2017-01-01 01:30:00+00:00   FOH   DEFAULT\n",
       "4  2017-01-01 02:00:00+00:00   POH  CALENDAR\n",
       "5  2017-01-01 03:00:00+00:00   POH  CALENDAR\n",
       "6  2017-01-01 04:00:00+00:00   POH  CALENDAR\n",
       "7  2017-01-01 05:00:00+00:00   FOH   DEFAULT\n",
       "8  2017-01-01 05:10:00+00:00   FOH   DEFAULT\n",
       "9  2017-01-01 05:30:00+00:00   FOH   DEFAULT\n",
       "10 2017-01-01 05:50:00+00:00    SH    SIGNAL\n",
       "11 2017-01-01 06:00:00+00:00   POH  CALENDAR\n",
       "12 2017-01-01 07:00:00+00:00    SH    SIGNAL\n",
       "13 2017-01-01 07:10:00+00:00   FOH   DEFAULT\n",
       "14 2017-01-01 07:20:00+00:00    SH    SIGNAL"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "!mc kpi-calculation \\\n",
    "--mode states \\\n",
    "--file \"./data/kpis/sample-timeseries.json\" \\\n",
    "--calendar \"./data/kpis/calendar.json\" \\\n",
    "--events \"./data/kpis/sample-events.json\" \\\n",
    "--target sensor \\\n",
    "--threshold 1.1 \\\n",
    "--shutdown 5000 \\\n",
    "--passkey $passkey \\\n",
    "> temp/compressor-states.json\n",
    "\n",
    "compressor = pd.read_json(\"./temp/compressor-states.json\")\n",
    "compressor\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate kpis for compressor\n",
    "\n",
    "In this section the [KPIs](https://developer.mindsphere.io/apis/analytics-kpicalculation/api-kpicalculation-basics-kpi.html) are computed with the state data calculated within the previous section.\n",
    "* **State KPIs** are taken from `compressor-states.json` file with the `state` column used as the target variable.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23cd48e0eb8>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "!mc kpi-calculation \\\n",
    "--mode kpis \\\n",
    "--file \"temp/compressor-states.json\" \\\n",
    "--target state \\\n",
    "--initialstate RSH \\\n",
    "--passkey $passkey \\\n",
    "> temp/kpi-compressor.json\n",
    "\n",
    "with open('temp/kpi-compressor.json') as f:\n",
    "  data = json.load(f)\n",
    "\n",
    "df = json_normalize (data).transpose()\n",
    "df[0:8].plot(kind=\"bar\", title=\"availability\")\n",
    "df[9:].plot(kind=\"bar\", title=\"factor\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Industrial Pump - TimeSeries Data\n",
    "\n",
    "We will use the downloaded industrial pump data from MindSphere to calculate the KPIs of the malfunctioning pump.\n",
    "The mindsphere demo application is available at\n",
    "\n",
    "https://dreamforce.mindconnect.rocks\n",
    "\n",
    "This application can be used without mindsphere credentials. \n",
    "* username: guest@mindsphere.io\n",
    "* password: Siemens123!\n",
    "\n",
    "Following data was recorded over the time:\n",
    "\n",
    "- **Flow** - Water flow through the pump\n",
    "- **MotorCurrent** Motor Current\n",
    "- **PressureIn** Inlet pressure\n",
    "- **PressureOut** Outlet pressure\n",
    "- **StuffingBoxTemperature** The temperature of the pump\n",
    "\n",
    "![pumpschema](./images/Pump-Station-Scheme.jpg)\n",
    "\n",
    "The historic data was downloaded from MindSphere and  is available in the **data** directory of this notebook.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## First look at the data\n",
    "\n",
    "The first file already has a typical behavior of the malfunctioning pump. The stuffing box temperature is increased and the flow is dropping."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23cd6b0d4e0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = pd.read_json(\"./data/PUMP-011/PumpData/PumpData_0.json\")\n",
    "data.plot(x=\"_time\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Determining the KPI States\n",
    "\n",
    "We will use the PUMP-011 data which is showing problems with flow availability.\n",
    "\n",
    "See also : https://developer.mindsphere.io/apis/analytics-kpicalculation/api-kpicalculation-overview.html\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>MotorCurrent</th>\n",
       "      <th>PressureOut</th>\n",
       "      <th>StuffingBoxTemperature</th>\n",
       "      <th>PressureIn</th>\n",
       "      <th>_time</th>\n",
       "      <th>Flow</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>50.0</td>\n",
       "      <td>2.92081</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:00:00.278000+00:00</td>\n",
       "      <td>240.996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50.0</td>\n",
       "      <td>3.02930</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:00:10.332000+00:00</td>\n",
       "      <td>248.293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>51.0</td>\n",
       "      <td>3.03644</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:00:20.385000+00:00</td>\n",
       "      <td>249.549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>51.0</td>\n",
       "      <td>2.97076</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:00:30.441000+00:00</td>\n",
       "      <td>245.991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>49.0</td>\n",
       "      <td>3.09670</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:00:40.496000+00:00</td>\n",
       "      <td>240.101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>51.0</td>\n",
       "      <td>2.98840</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:00:50.560000+00:00</td>\n",
       "      <td>244.602</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>49.0</td>\n",
       "      <td>2.94345</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:01:00.615000+00:00</td>\n",
       "      <td>242.534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>51.0</td>\n",
       "      <td>3.04051</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:01:10.668000+00:00</td>\n",
       "      <td>249.716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>51.0</td>\n",
       "      <td>2.95434</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:01:20.729000+00:00</td>\n",
       "      <td>246.405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>51.0</td>\n",
       "      <td>2.94290</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2020-02-05 23:01:30.786000+00:00</td>\n",
       "      <td>242.101</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   MotorCurrent  PressureOut  StuffingBoxTemperature  PressureIn  \\\n",
       "0          50.0      2.92081                    25.0         1.0   \n",
       "1          50.0      3.02930                    25.0         1.0   \n",
       "2          51.0      3.03644                    25.0         1.0   \n",
       "3          51.0      2.97076                    25.0         1.0   \n",
       "4          49.0      3.09670                    25.0         1.0   \n",
       "5          51.0      2.98840                    25.0         1.0   \n",
       "6          49.0      2.94345                    25.0         1.0   \n",
       "7          51.0      3.04051                    25.0         1.0   \n",
       "8          51.0      2.95434                    25.0         1.0   \n",
       "9          51.0      2.94290                    25.0         1.0   \n",
       "\n",
       "                             _time     Flow  \n",
       "0 2020-02-05 23:00:00.278000+00:00  240.996  \n",
       "1 2020-02-05 23:00:10.332000+00:00  248.293  \n",
       "2 2020-02-05 23:00:20.385000+00:00  249.549  \n",
       "3 2020-02-05 23:00:30.441000+00:00  245.991  \n",
       "4 2020-02-05 23:00:40.496000+00:00  240.101  \n",
       "5 2020-02-05 23:00:50.560000+00:00  244.602  \n",
       "6 2020-02-05 23:01:00.615000+00:00  242.534  \n",
       "7 2020-02-05 23:01:10.668000+00:00  249.716  \n",
       "8 2020-02-05 23:01:20.729000+00:00  246.405  \n",
       "9 2020-02-05 23:01:30.786000+00:00  242.101  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## KPI State Calculation \n",
    "\n",
    "First step is to calculate the corresponding KPI states from the Flow signal. We will assume that the flow must be > 200 for the machine to be running. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>_time</th>\n",
       "      <th>state</th>\n",
       "      <th>source</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-02-05T23:00:00.278Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-02-05T23:00:10.332Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-02-05T23:00:20.385Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-02-05T23:00:30.441Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020-02-05T23:00:40.496Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>2020-02-05T23:32:41.097Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>2020-02-05T23:32:51.150Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>2020-02-05T23:33:01.206Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>2020-02-05T23:33:11.261Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>199</th>\n",
       "      <td>2020-02-05T23:33:21.314Z</td>\n",
       "      <td>RSH</td>\n",
       "      <td>DEFAULT</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>200 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        _time state   source\n",
       "0    2020-02-05T23:00:00.278Z   RSH  DEFAULT\n",
       "1    2020-02-05T23:00:10.332Z   RSH  DEFAULT\n",
       "2    2020-02-05T23:00:20.385Z   RSH  DEFAULT\n",
       "3    2020-02-05T23:00:30.441Z   RSH  DEFAULT\n",
       "4    2020-02-05T23:00:40.496Z   RSH  DEFAULT\n",
       "..                        ...   ...      ...\n",
       "195  2020-02-05T23:32:41.097Z   RSH  DEFAULT\n",
       "196  2020-02-05T23:32:51.150Z   RSH  DEFAULT\n",
       "197  2020-02-05T23:33:01.206Z   RSH  DEFAULT\n",
       "198  2020-02-05T23:33:11.261Z   RSH  DEFAULT\n",
       "199  2020-02-05T23:33:21.314Z   RSH  DEFAULT\n",
       "\n",
       "[200 rows x 3 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "!mc kpi-calculation \\\n",
    "--mode states \\\n",
    "--file \"./data/PUMP-011/PumpData/PumpData_0.json\" \\\n",
    "--target State \\\n",
    "--threshold 200 \\\n",
    "--passkey $passkey \\\n",
    "> temp/states.json\n",
    "\n",
    "with open('temp/states.json') as f:\n",
    "  data = json.load(f)\n",
    "\n",
    "df = json_normalize (data)\n",
    "df\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculating KPIs\n",
    "\n",
    "We will now calculate the KPIs for the selected time period (Note this is only a bit more than a half an hour of data, the availibilities are calculated for that)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x23cd6c342e8>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "result = !mc kpi-calculation \\\n",
    "--mode kpis \\\n",
    "--file \"temp/states.json\" \\\n",
    "--target State \\\n",
    "--passkey $passkey \\\n",
    "> temp/kpi.json\n",
    "\n",
    "with open('temp/kpi.json') as f:\n",
    "  data = json.load(f)\n",
    "\n",
    "df = json_normalize (data).transpose()\n",
    "df[0:8].plot(kind=\"bar\", title=\"availability\")\n",
    "df[9:].plot(kind=\"bar\", title=\"factor\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Siemens API Notice\n",
    "\n",
    "This project has been released under an Open Source license. The release may include and/or use APIs to Siemens’ or third parties’ products or services. In no event shall the project’s Open Source license grant any rights in or to these APIs, products or services that would alter, expand, be inconsistent with, or supersede any terms of separate license agreements applicable to those APIs. “API” means application programming interfaces and their specifications and implementing code that allows other software to communicate with or call on Siemens’ or third parties’ products or services and may be made available through Siemens’ or third parties’ products, documentations or otherwise."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}