{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Analyse du risque de défaillance des joints toriques de la navette Challenger" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Le 27 Janvier 1986, veille du décollage de la navette *Challenger*, eu\n", "lieu une télé-conférence de trois heures entre les ingénieurs de la\n", "Morton Thiokol (constructeur d'un des moteurs) et de la NASA. La\n", "discussion portait principalement sur les conséquences de la\n", "température prévue au moment du décollage de 31°F (juste en dessous de\n", "0°C) sur le succès du vol et en particulier sur la performance des\n", "joints toriques utilisés dans les moteurs. En effet, aucun test\n", "n'avait été effectué à cette température.\n", "\n", "L'étude qui suit reprend donc une partie des analyses effectuées cette\n", "nuit là et dont l'objectif était d'évaluer l'influence potentielle de\n", "la température et de la pression à laquelle sont soumis les joints\n", "toriques sur leur probabilité de dysfonctionnement. Pour cela, nous\n", "disposons des résultats des expériences réalisées par les ingénieurs\n", "de la NASA durant les 6 années précédant le lancement de la navette\n", "Challenger.\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Chargement des données\n", "Nous commençons donc par charger ces données:" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateCountTemperaturePressureMalfunction
04/12/81666500
111/12/81670501
23/22/82669500
311/11/82668500
44/04/83667500
56/18/82672500
68/30/836731000
711/28/836701000
82/03/846572001
94/06/846632001
108/30/846702001
1110/05/846782000
1211/08/846672000
131/24/856532002
144/12/856672000
154/29/856752000
166/17/856702000
177/29/856812000
188/27/856762000
1910/03/856792000
2010/30/856752002
2111/26/856762000
221/12/866582001
\n", "
" ], "text/plain": [ " Date Count Temperature Pressure Malfunction\n", "0 4/12/81 6 66 50 0\n", "1 11/12/81 6 70 50 1\n", "2 3/22/82 6 69 50 0\n", "3 11/11/82 6 68 50 0\n", "4 4/04/83 6 67 50 0\n", "5 6/18/82 6 72 50 0\n", "6 8/30/83 6 73 100 0\n", "7 11/28/83 6 70 100 0\n", "8 2/03/84 6 57 200 1\n", "9 4/06/84 6 63 200 1\n", "10 8/30/84 6 70 200 1\n", "11 10/05/84 6 78 200 0\n", "12 11/08/84 6 67 200 0\n", "13 1/24/85 6 53 200 2\n", "14 4/12/85 6 67 200 0\n", "15 4/29/85 6 75 200 0\n", "16 6/17/85 6 70 200 0\n", "17 7/29/85 6 81 200 0\n", "18 8/27/85 6 76 200 0\n", "19 10/03/85 6 79 200 0\n", "20 10/30/85 6 75 200 2\n", "21 11/26/85 6 76 200 0\n", "22 1/12/86 6 58 200 1" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import numpy as np\n", "import pandas as pd\n", "import statsmodels as sm\n", "from statsmodels.tools import add_constant\n", "data = pd.read_csv(\"shuttle.csv\")\n", "data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Le jeu de données nous indique la date de l'essai, le nombre de joints\n", "toriques mesurés (il y en a 6 sur le lançeur principal), la\n", "température (en Farenheit) et la pression (en psi), et enfin le\n", "nombre de dysfonctionnements relevés. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Regression logistique" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data['problem'] = np.where(data['Malfunction']>0, 1, 0)\n", "data.plot(x=\"Temperature\",y=\"Malfunction\",kind=\"scatter\")\n", "data.plot(x=\"Temperature\",y=\"problem\",kind=\"scatter\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Optimization terminated successfully.\n", " Current function value: 0.441635\n", " Iterations 7\n", " Logit Regression Results \n", "==============================================================================\n", "Dep. Variable: problem No. Observations: 23\n", "Model: Logit Df Residuals: 21\n", "Method: MLE Df Model: 1\n", "Date: Sun, 29 Nov 2020 Pseudo R-squ.: 0.2813\n", "Time: 20:53:49 Log-Likelihood: -10.158\n", "converged: True LL-Null: -14.134\n", " LLR p-value: 0.004804\n", "===============================================================================\n", " coef std err z P>|z| [0.025 0.975]\n", "-------------------------------------------------------------------------------\n", "const 15.0429 7.379 2.039 0.041 0.581 29.505\n", "Temperature -0.2322 0.108 -2.145 0.032 -0.444 -0.020\n", "===============================================================================\n" ] } ], "source": [ "#data[\"Malfunction\"] = data[\"Malfunction\"].astype('category')\n", "\n", "y = data[\"problem\"]\n", "\n", "# on ne prend que les colonnes quantitatives\n", "x = data[\"Temperature\"]\n", "# on ajoute une colonne pour la constante\n", "x_stat = sm.tools.add_constant(x)\n", "# on ajuste le modèle\n", "model = sm.api.Logit(y, x_stat)\n", "result = model.fit()\n", "print(result.summary())" ] }, { "cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[9.99969558e-01 9.99951569e-01 9.99922951e-01 9.99877426e-01\n", " 9.99805006e-01 9.99689811e-01 9.99506599e-01 9.99215256e-01\n", " 9.98752098e-01 9.98016125e-01 9.96847468e-01 9.94993835e-01\n", " 9.92058981e-01 9.87425318e-01 9.80141989e-01 9.68773521e-01\n", " 9.51220648e-01 9.24569286e-01 8.85115205e-01 8.28844843e-01\n", " 7.52713482e-01 6.56742588e-01 5.45991136e-01 4.30493132e-01\n", " 3.22094054e-01 2.29968258e-01 1.58049102e-01 1.05538936e-01\n", " 6.90440720e-02 4.45405463e-02 2.84673270e-02 1.80846183e-02\n", " 1.14441228e-02 7.22401354e-03 4.55293511e-03 2.86663594e-03\n", " 1.80377025e-03 1.13453604e-03 7.13423407e-04 4.48547518e-04]\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "X1= range(20,100,2)\n", "X1_test = sm.tools.add_constant(X1)\n", "ypred = result.predict(X1_test)\n", "print(ypred)\n", "ax1 = data.plot(x=\"Temperature\",y=\"problem\",kind=\"scatter\")\n", "ax1 = plt.plot(X1, ypred ,'r--')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Inspection graphique des données\n", "Les vols où aucun incident n'est relevé n'apportant aucun information\n", "sur l'influence de la température ou de la pression sur les\n", "dysfonctionnements, nous nous concentrons sur les expériences où au\n", "moins un joint a été défectueux.\n" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DateCountTemperaturePressureMalfunctionproblem
111/12/816705011
82/03/8465720011
94/06/8466320011
108/30/8467020011
131/24/8565320021
2010/30/8567520021
221/12/8665820011
\n", "
" ], "text/plain": [ " Date Count Temperature Pressure Malfunction problem\n", "1 11/12/81 6 70 50 1 1\n", "8 2/03/84 6 57 200 1 1\n", "9 4/06/84 6 63 200 1 1\n", "10 8/30/84 6 70 200 1 1\n", "13 1/24/85 6 53 200 2 1\n", "20 10/30/85 6 75 200 2 1\n", "22 1/12/86 6 58 200 1 1" ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = data[data.Malfunction>0]\n", "data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Très bien, nous avons une variabilité de température importante mais\n", "la pression est quasiment toujours égale à 200, ce qui devrait\n", "simplifier l'analyse.\n", "\n", "Comment la fréquence d'échecs varie-t-elle avec la température ?\n" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "%matplotlib inline\n", "pd.set_option('mode.chained_assignment',None) # this removes a useless warning from pandas\n", "import matplotlib.pyplot as plt\n", "\n", "data[\"Frequency\"]=data.Malfunction/data.Count\n", "data.plot(x=\"Temperature\",y=\"Frequency\",kind=\"scatter\",ylim=[0,1])\n", "plt.grid(True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "À première vue, ce n'est pas flagrant mais bon, essayons quand même\n", "d'estimer l'impact de la température $t$ sur la probabilité de\n", "dysfonctionnements d'un joint. \n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Estimation de l'influence de la température\n", "\n", "Supposons que chacun des 6 joints toriques est endommagé avec la même\n", "probabilité et indépendamment des autres et que cette probabilité ne\n", "dépend que de la température. Si on note $p(t)$ cette probabilité, le\n", "nombre de joints $D$ dysfonctionnant lorsque l'on effectue le vol à\n", "température $t$ suit une loi binomiale de paramètre $n=6$ et\n", "$p=p(t)$. Pour relier $p(t)$ à $t$, on va donc effectuer une\n", "régression logistique." ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
Generalized Linear Model Regression Results
Dep. Variable: Frequency No. Observations: 7
Model: GLM Df Residuals: 5
Model Family: Binomial Df Model: 1
Link Function: logit Scale: 1.0000
Method: IRLS Log-Likelihood: -2.5250
Date: Sun, 29 Nov 2020 Deviance: 0.22231
Time: 21:12:59 Pearson chi2: 0.236
No. Iterations: 4 Covariance Type: nonrobust
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err z P>|z| [0.025 0.975]
Intercept -1.3895 7.828 -0.178 0.859 -16.732 13.953
Temperature 0.0014 0.122 0.012 0.991 -0.238 0.240
" ], "text/plain": [ "\n", "\"\"\"\n", " Generalized Linear Model Regression Results \n", "==============================================================================\n", "Dep. Variable: Frequency No. Observations: 7\n", "Model: GLM Df Residuals: 5\n", "Model Family: Binomial Df Model: 1\n", "Link Function: logit Scale: 1.0000\n", "Method: IRLS Log-Likelihood: -2.5250\n", "Date: Sun, 29 Nov 2020 Deviance: 0.22231\n", "Time: 21:12:59 Pearson chi2: 0.236\n", "No. Iterations: 4 Covariance Type: nonrobust\n", "===============================================================================\n", " coef std err z P>|z| [0.025 0.975]\n", "-------------------------------------------------------------------------------\n", "Intercept -1.3895 7.828 -0.178 0.859 -16.732 13.953\n", "Temperature 0.0014 0.122 0.012 0.991 -0.238 0.240\n", "===============================================================================\n", "\"\"\"" ] }, "execution_count": 46, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import statsmodels.api as sm\n", "\n", "data[\"Success\"]=data.Count-data.Malfunction\n", "data[\"Intercept\"]=1\n", "\n", "logmodel=sm.GLM(data['Frequency'], data[['Intercept','Temperature']], family=sm.families.Binomial(sm.families.links.logit)).fit()\n", "\n", "logmodel.summary()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "L'estimateur le plus probable du paramètre de température est 0.0014\n", "et l'erreur standard de cet estimateur est de 0.122, autrement dit on\n", "ne peut pas distinguer d'impact particulier et il faut prendre nos\n", "estimations avec des pincettes.\n" ] }, { "cell_type": "code", "execution_count": 47, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1 5\n", "8 5\n", "9 5\n", "10 5\n", "13 4\n", "20 4\n", "22 5\n", "Name: Success, dtype: int64" ] }, "execution_count": 47, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data[\"Success\"]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Estimation de la probabilité de dysfonctionnant des joints toriques\n", "La température prévue le jour du décollage est de 31°F. Essayons\n", "d'estimer la probabilité de dysfonctionnement des joints toriques à\n", "cette température à partir du modèle que nous venons de construire:\n" ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEKCAYAAADpfBXhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAGzdJREFUeJzt3X+UVOWd5/H3t6tBGhohoGGAJoHM4cA6UX41jUriNkYBc+KvWQ2io4k7LHEnJJPdIxs5J7OaWT1n57S7h0zWiIwyTOLR1nEVNcsG1E3HiauxQRAEhh9riDadBDGj0Noo3f3dP+6t6qrqbrq6qO6qevy8zulD3VvPfe7z7aI+dfupW7fM3RERkbBUFHsAIiJSeAp3EZEAKdxFRAKkcBcRCZDCXUQkQAp3EZEA9RvuZrbBzI6a2Rt93G9m9rdmdsjMdpnZ3MIPU0REBiKXI/eNwNLT3H8FMD3+WQncf+bDEhGRM9FvuLv7i8AfTtPkauDHHnkFGGtmEws1QBERGbjKAvQxGXg7bbklXvfb7IZmtpLo6J6qqqp5U6ZMyWuHXV1dVFSE8XaBailNodQSSh2gWpIOHDhwzN3P7a9dIcLdelnX6zUN3H09sB6gtrbWt23bltcOm5qaqK+vz2vbUqNaSlMotYRSB6iWJDP7TS7tCvEy2AKkH4LXAK0F6FdERPJUiHB/BrglPmvmQuB9d+8xJSMiIkOn32kZM3sUqAfOMbMW4E5gGIC7rwM2A18GDgEfArcO1mBFRCQ3/Ya7uy/v534HvlmwEYlIWTh16hQtLS2cPHlySPY3ZswY9u3bNyT7Gmy51DJixAhqamoYNmxYXvsoxBuqIvIJ1NLSwujRo5k6dSpmvZ1XUVgnTpxg9OjRg76fodBfLe7Ou+++S0tLC9OmTctrH2GcVyQiQ+7kyZOMHz9+SIL9k8bMGD9+/Bn9VaRwF5G8KdgHz5n+bhXuIiIB0py7iJStRCLB+eefn1retGkTU6dOLd6ASojCXUTKVlVVFTt37uzz/o6ODiorP5kxp2kZEQnKxo0buf7667nyyitZvHgxAA0NDcyfP58LLriAO++8M9X2nnvuYcaMGVx22WUsX76ce++9F4D6+nqSl0c5duxY6q+Bzs5OVq9enerrgQceALovJ3Ddddcxc+ZMbrrpJqKzxKG5uZmLL76YWbNmUVdXx4kTJ1iyZEnGi9LChQvZtWtXQX8Pn8yXNBEpqO8/u4e9rccL2ud5k87mziv/5LRt2tvbmT17NgDTpk3jqaeeAuDll19m165djBs3jq1bt3Lw4EFeffVV3J2rrrqKF198kVGjRtHY2MiOHTvo6Ohg7ty5zJs377T7e+ihhxgzZgzNzc189NFHLFy4MPUCsmPHDvbs2cOkSZNYuHAhL730EnV1dSxbtozHHnuM+fPnc/z4caqqqrjlllvYuHEja9eu5cCBA3z00UdccMEFBfitdVO4i0jZ6mta5vLLL2fcuHEAbN26la1btzJnzhwA2traOHjwICdOnODaa69l5MiRAFx11VX97m/r1q3s2rWLJ554AoD333+fgwcPMnz4cOrq6qipqQFg9uzZHD58mDFjxjBx4kTmz58PwNlnnw3Atddey8KFC2loaGDDhg18/etfP7NfRC8U7iJyxvo7wh5qo0aNSt12d9asWcM3vvGNjDZr167t83TDyspKurq6ADLONXd3fvjDH7JkyZKM9k1NTZx11lmp5UQiQUdHB+7e6z5GjhzJ5ZdfztNPP83jjz9OvlfIPR3NuYtI0JYsWcKGDRtoa2sD4MiRIxw9epRLLrmEp556ivb2dk6cOMGzzz6b2mbq1Kls374dIHWUnuzr/vvv59SpUwAcOHCADz74oM99z5w5k9bWVpqbm4Hok6kdHR0ArFixgm9/+9vMnz8/9VdGIenIXUSCtnjxYvbt28dFF10EQHV1NQ8//DBz585l2bJlzJ49m89+9rN88YtfTG1z++2389WvfpWf/OQnXHrppan1K1as4PDhw8ydOxd359xzz2XTpk197nv48OE89thjfOtb36K9vZ2qqiqef/55AObNm8fZZ5/NrbcO0rUW3b0oP/PmzfN8/fznP89721KjWkpTKLUMZh179+4dtL57c/z48UHt/8477/SGhoZB3UfS8ePH/ciRIz59+nTv7Ozss11vv2Ngm+eQsZqWEREZYo888ggLFizgnnvuGbSvDtS0jIgIcNdddw3Zvm688cYeb/AWmo7cRSRv7r1+XbIUwJn+bhXuIpKXESNG8O677yrgB4HH13MfMWJE3n1oWkZE8lJTU0NLSwvvvPPOkOzv5MmTZxR2pSSXWpLfxJQvhbuI5GXYsGF5f0tQPpqamlKfMi13Q1GLpmVERAKkcBcRCZDCXUQkQAp3EZEAKdxFRAKkcBcRCZDCXUQkQAp3EZEAKdxFRAKkcBcRCZDCXUQkQAp3EZEAKdxFRAKkcBcRCZDCXUQkQAp3EZEA5RTuZrbUzPab2SEzu6OX+8eY2bNm9rqZ7TGzWws/VBERyVW/4W5mCeA+4ArgPGC5mZ2X1eybwF53nwXUA//NzIYXeKwiIpKjXI7c64BD7v6mu38MNAJXZ7VxYLSZGVAN/AHoKOhIRUQkZ9bfN5eb2XXAUndfES/fDCxw91VpbUYDzwAzgdHAMnf/X730tRJYCTBhwoR5jY2NeQ26ra2N6urqvLYtNaqlNIVSSyh1gGpJWrRo0XZ3r+2vXS5fkG29rMt+RVgC7AQuBf4YeM7M/sndj2ds5L4eWA9QW1vr9fX1Oey+p6amJvLdttSoltIUSi2h1AGqZaBymZZpAaakLdcArVltbgWe9Mgh4NdER/EiIlIEuYR7MzDdzKbFb5LeQDQFk+4t4EsAZjYBmAG8WciBiohI7vqdlnH3DjNbBWwBEsAGd99jZrfF968D/guw0cx2E03jfNfdjw3iuEVE5DRymXPH3TcDm7PWrUu73QosLuzQREQkX/qEqohIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIByinczWypme03s0NmdkcfberNbKeZ7TGzXxR2mCIiMhCV/TUwswRwH3A50AI0m9kz7r43rc1Y4EfAUnd/y8w+PVgDFhGR/uVy5F4HHHL3N939Y6ARuDqrzY3Ak+7+FoC7Hy3sMEVEZCDM3U/fwOw6oiPyFfHyzcACd1+V1mYtMAz4E2A08AN3/3Evfa0EVgJMmDBhXmNjY16Dbmtro7q6Oq9tS41qKU2h1BJKHaBakhYtWrTd3Wv7a9fvtAxgvazLfkWoBOYBXwKqgJfN7BV3P5Cxkft6YD1AbW2t19fX57D7npqamsh321KjWkpTKLWEUgeoloHKJdxbgClpyzVAay9tjrn7B8AHZvYiMAs4gIiIDLlc5tybgelmNs3MhgM3AM9ktXka+KKZVZrZSGABsK+wQxURkVz1e+Tu7h1mtgrYAiSADe6+x8xui+9f5+77zOxnwC6gC3jQ3d8YzIGLiEjfcpmWwd03A5uz1q3LWm4AGgo3NBERyZc+oSoiEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIByinczWypme03s0Nmdsdp2s03s04zu65wQxQRkYHqN9zNLAHcB1wBnAcsN7Pz+mj3N8CWQg9SREQGJpcj9zrgkLu/6e4fA43A1b20+xbwP4GjBRyfiIjkwdz99A2iKZal7r4iXr4ZWODuq9LaTAYeAS4FHgJ+6u5P9NLXSmAlwIQJE+Y1NjbmNei2tjaqq6vz2rbUqJbSFEotodQBqiVp0aJF2929tr92lTn0Zb2sy35FWAt81907zXprHm/kvh5YD1BbW+v19fU57L6npqYm8t221KiW0hRKLaHUAaploHIJ9xZgStpyDdCa1aYWaIyD/Rzgy2bW4e6bCjJKEREZkFzCvRmYbmbTgCPADcCN6Q3cfVrytpltJJqWUbCLiBRJv+Hu7h1mtoroLJgEsMHd95jZbfH96wZ5jCIiMkC5HLnj7puBzVnreg11d//6mQ9LRETOhD6hKiISIIW7iEiAFO4iIgFSuIuIBEjhLiISoJzOlhEZLJt2HKFhy35a32tn0tgqVi+ZwTVzJhd7WJIjPX6lS+EuRbNpxxHWPLmb9lOdABx5r501T+4GUECUAT1+pU3TMlI0DVv2p4Ihqf1UJw1b9hdpRDIQevxKm8Jdiqb1vfYBrZfSosevtCncpWgmja0a0HopLXr8SpvCXYpm9ZIZVA1LZKyrGpZg9ZIZRRqRDIQev9KmN1SlaJJvuulsi/Kkx6+0KdylqK6ZM1lhUMb0+JUuTcuIiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgBTuIiIBUriLiARI4S4iEiCFu4hIgPQdqiKSoavL6XSns8txhy6Plru6onVdyXVdTpdHbTq7utt0xcvpbaKf7vVdXaS173ubzi5Sbfa9dYq3X/lNarm7PXGfcT/ueHx/97hJG0e0nN0muW16ne6Zv4vM9k6nkzEeT+/D02qNl5P7rJ8E9fWD+zgq3KVkZD8hkqHicRBET55kG9KehJlP8vQnWXKb04ZH6knXHQxvHDnFse0tWWPKDLfsfWYHTPc2PYOlOxy6t+m+Py18soIlo413B1SX9+yz050PPzzJWS+/EPeT2Wdv++3yYv8v6MfeN3JqVmFQYUZFhZEwi5YrjES8bPG6RIVRYfH6CsMMEvFytH3cT2pdtDy8siJubyTifsBIVGT22b0tGcuf+uh3g/t7IsdwN7OlwA+ABPCgu//XrPtvAr4bL7YB/97dXy/kQAdD8kmf+YSnO0RSr7ZpbdJDp7cnd3/bpx21uDu7ftfB8ddb+36SZh0Z9NwHWUcSWUHQ1d1PakwZYdMzJLt6CZLMcfUMlU532to+ZETzz+Px9AzAjPBLO5JJLpec3fn9F06GQPTETwuF7Cd7HDLpwVIRbxMFUWawGFEflRUVnFVpcX+k2ifDKn2/R3//eyZNPKc76LL67G2/6X2mB1oqsNLWRTWQ6is7BLtDMnus3WNJtkn1k94mDthEhfHKyy/zhYULM+usyBpDfNvMCvt/ocCamo4N+j76DXczSwD3AZcDLUCzmT3j7nvTmv0a+Nfu/i9mdgWwHlgwGAP+xYF3+N4vP6TqtV9kBFnGEVp6qPU4ius+qvJSyZOdO864i+wndnqYdB8xdD+Zsp8MmUcZWdvHbXo7WrG0J+CxinYm/tHYVD+ZT0LSjpjiJ3bqqCrtyCrtiZ1ZU+YTurfwSMT9ZuwjOzxS9VtakKSHV7T9tldf5eKLLszpd5PcPjn2UtLU1ER9/axiD6MgPjWignNHn1XsYZSNXI7c64BD7v4mgJk1AlcDqXB39/+b1v4VoKaQg0xXfVaCCaMqmPDp6h5BlgqKrCd1X0FD9vanOUrp7jvz6Cc9ELqDI3P/GUdyWcG0fds2Llwwv8f22WPv3nd32Gb3XWxRkMwp9jAK4u1RFUwZN7LYwxDJm3k/h69mdh2w1N1XxMs3AwvcfVUf7W8HZibbZ923ElgJMGHChHmNjY15DbqtrY3q6uq8ti01qqU0hVJLKHWAaklatGjRdnev7a9dLkfuvR0S9vqKYGaLgD8HvtDb/e6+nmjKhtraWq/P8+3i6Agxv21LjWopTaHUEkodoFoGKpdwbwGmpC3XAK3ZjczsAuBB4Ap3f7cwwxMRkXzk8iGmZmC6mU0zs+HADcAz6Q3M7DPAk8DN7n6g8MMUEZGB6PfI3d07zGwVsIXoVMgN7r7HzG6L718H/GdgPPCj+I29jlzmhEREZHDkdJ67u28GNmetW5d2ewXQ4w1UkaG2accRGrbsp/W9diaNrWL1khkAPdZdM2fykOx7MPaTi+9t2s2jv3qb73z+FH++ZjPLF0zh7mvOL8pYpDj0CVUJxqYdR1jz5G7aT3UCcOS9dlb/4+tgcKrTU+vWPLkboKDB29u+B2M/ufjept08/MpbqeVO99SyAv6TQxcOk2A0bNmfCtekU12eCvak9lOdNGzZP+j7Hoz95OLRX709oPUSJoW7BKP1vfZBaXsm/RV6P7no7OOzK32tlzAp3CUYk8ZWDUrbM+mv0PvJRaKPTyv3tV7CpHCXYKxeMoOqYYmMdcMqjGGJzFCrGpZIvdE6mPsejP3kYvmCKQNaL2HSG6oSjOQbl8U4W6avfRfjbJnkm6bJOfaEmc6W+QRSuEtQrpkzuddAHYqQ7WvfxXD3Nedz9zXn09TUxP+7qb7Yw5Ei0LSMiEiAFO4iIgFSuIuIBEjhLiISIIW7iEiAFO4iIgFSuIuIBEjhLiISIIW7iEiAFO4iIgFSuIuIBEjhLiISIIW7iEiAFO4iIgFSuIuIBEjhLiISIIW7iEiAFO4iIgFSuIuIBEjhLiISIIW7iEiAFO4iIgFSuIuIBEjhLiISIIW7iEiAFO4iIgFSuIuIBEjhLiISoJzC3cyWmtl+MztkZnf0cr+Z2d/G9+8ys7mFH6qIiOSq33A3swRwH3AFcB6w3MzOy2p2BTA9/lkJ3F/gcYqIyADkcuReBxxy9zfd/WOgEbg6q83VwI898gow1swmFnisIiKSo8oc2kwG3k5bbgEW5NBmMvDb9EZmtpLoyB6gzcz2D2i03c4BjuW5balRLaUplFpCqQNUS9Jnc2mUS7hbL+s8jza4+3pgfQ77PP2AzLa5e+2Z9lMKVEtpCqWWUOoA1TJQuUzLtABT0pZrgNY82oiIyBDJJdybgelmNs3MhgM3AM9ktXkGuCU+a+ZC4H13/212RyIiMjT6nZZx9w4zWwVsARLABnffY2a3xfevAzYDXwYOAR8Ctw7ekIECTO2UENVSmkKpJZQ6QLUMiLn3mBoXEZEyp0+oiogESOEuIhKgkg93MxthZq+a2etmtsfMvh+vH2dmz5nZwfjfTxV7rLkws4SZ7TCzn8bL5VrHYTPbbWY7zWxbvK5caxlrZk+Y2T+b2T4zu6gcazGzGfHjkfw5bmbfKdNa/kP8fH/DzB6Nc6Ds6gAws7+M69hjZt+J1w16LSUf7sBHwKXuPguYDSyNz8i5A3jB3acDL8TL5eAvgX1py+VaB8Aid5+ddr5uudbyA+Bn7j4TmEX0+JRdLe6+P348ZgPziE5ueIoyq8XMJgPfBmrd/fNEJ3LcQJnVAWBmnwf+HdEn/WcBXzGz6QxFLe5eNj/ASOA1ok/I7gcmxusnAvuLPb4cxl8TP5CXAj+N15VdHfFYDwPnZK0ru1qAs4FfE59cUM61ZI1/MfBSOdZC9yfexxGd0ffTuJ6yqiMe5/XAg2nLfwX8p6GopRyO3JNTGTuBo8Bz7v4rYILH59LH/366mGPM0VqiB7YrbV051gHRJ5C3mtn2+LISUJ61fA54B/j7eLrsQTMbRXnWku4G4NH4dlnV4u5HgHuBt4guYfK+u2+lzOqIvQFcYmbjzWwk0SnjUxiCWsoi3N2906M/NWuAuvhPnbJiZl8Bjrr79mKPpUAWuvtcoiuCftPMLin2gPJUCcwF7nf3OcAHlMGf+6cTf9jwKuAfiz2WfMTzz1cD04BJwCgz+7Pijio/7r4P+BvgOeBnwOtAx1DsuyzCPcnd3wOagKXA75NXnoz/PVrEoeViIXCVmR0murLmpWb2MOVXBwDu3hr/e5RoXreO8qylBWiJ/xoEeIIo7MuxlqQrgNfc/ffxcrnVchnwa3d/x91PAU8CF1N+dQDg7g+5+1x3vwT4A3CQIail5MPdzM41s7Hx7SqiB/6fiS558LW42deAp4szwty4+xp3r3H3qUR/Mv8fd/8zyqwOADMbZWajk7eJ5kPfoAxrcfffAW+b2Yx41ZeAvZRhLWmW0z0lA+VXy1vAhWY20syM6DHZR/nVAYCZfTr+9zPAnxI9NoNeS8l/QtXMLgD+gegd8wrgcXf/azMbDzwOfIboP8P17v6H4o00d2ZWD9zu7l8pxzrM7HNER+sQTWs84u73lGMtAGY2G3gQGA68SXT5jArKs5aRRG9Gfs7d34/Xld3jEp/yvIxoCmMHsAKopszqADCzfwLGA6eA/+juLwzFY1Ly4S4iIgNX8tMyIiIycAp3EZEAKdxFRAKkcBcRCZDCXUQkQLl8QbbIkIpPE3shXvwjoJPoEgEAde7+cVEGdhpm9m+BzfF58yJFp1MhpaSZ2V1Am7vfWwJjSbh7Zx/3/RJY5e47B9BfpbsPyUfR5ZNH0zJSVszsaxZd33+nmf3IzCrMrNLM3jOzBjN7zcy2mNkCM/uFmb1pZl+Ot11hZk/F9+83s+/l2O/dZvYq0XWNvm9mzfH1uddZZBnR5agfi7cfbmYtaZ+svtDMno9v321mD5jZc0QXK6s0s/8e73uXma0Y+t+qhEjhLmUjvmDctcDF8YXkKoku5QAwBtgaX8zsY+Auoo+tXw/8dVo3dfE2c4EbzWx2Dv2+5u517v4y8AN3nw+cH9+31N0fA3YCyzy6nnp/00ZzgCvd/WZgJdEF5eqA+UQXYftMPr8fkXSac5dychlRAG6LLjlCFdFH7QHa3f25+PZuosvEdpjZbmBqWh9b3P1fAMxsE/AFoudBX/1+TPelFgC+ZGargRHAOcB24H8PsI6n3f1kfHsx8K/MLP3FZDrRR9JF8qZwl3JiwAZ3/6uMlWaVRCGc1EX0DV7J2+n/z7PfZPJ++m33+I2p+Lot/wOY6+5HzOxuopDvTQfdfxlnt/kgq6a/cPcXECkgTctIOXke+KqZnQPRWTV5TGEstug7U0cSXTP8pQH0W0X0YnEsvirmv0m77wQwOm35MNFX3ZHVLtsW4C/iF5Lk96BWDbAmkR505C5lw913x1cLfN7MKoiusncb0DqAbn4JPAL8MfCT5NktufTr7u+a2T8QXd74N8Cv0u7+e+BBM2snmte/C/g7M/sd8OppxvMA0ZUBd8ZTQkeJXnREzohOhZRPjPhMlM+7+3eKPRaRwaZpGRGRAOnIXUQkQDpyFxEJkMJdRCRACncRkQAp3EVEAqRwFxEJ0P8HfLcy7/zjy3oAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "%matplotlib inline\n", "data_pred = pd.DataFrame({'Temperature': np.linspace(start=30, stop=90, num=121), 'Intercept': 1})\n", "data_pred['Frequency'] = logmodel.predict(data_pred[['Intercept','Temperature']])\n", "data_pred.plot(x=\"Temperature\",y=\"Frequency\",kind=\"line\",ylim=[0,1])\n", "plt.scatter(x=data[\"Temperature\"],y=data[\"Frequency\"])\n", "plt.grid(True)" ] }, { "cell_type": "markdown", "metadata": { "hideCode": false, "hidePrompt": false, "scrolled": true }, "source": [ "Comme on pouvait s'attendre au vu des données initiales, la\n", "température n'a pas d'impact notable sur la probabilité d'échec des\n", "joints toriques. Elle sera d'environ 0.2, comme dans les essais\n", "précédents où nous il y a eu défaillance d'au moins un joint. Revenons\n", "à l'ensemble des données initiales pour estimer la probabilité de\n", "défaillance d'un joint:\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "data = pd.read_csv(\"shuttle.csv\")\n", "print(np.sum(data.Malfunction)/np.sum(data.Count))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Cette probabilité est donc d'environ $p=0.065$, sachant qu'il existe\n", "un joint primaire un joint secondaire sur chacune des trois parties du\n", "lançeur, la probabilité de défaillance des deux joints d'un lançeur\n", "est de $p^2 \\approx 0.00425$. La probabilité de défaillance d'un des\n", "lançeur est donc de $1-(1-p^2)^3 \\approx 1.2%$. Ça serait vraiment\n", "pas de chance... Tout est sous contrôle, le décollage peut donc avoir\n", "lieu demain comme prévu.\n", "\n", "Seulement, le lendemain, la navette Challenger explosera et emportera\n", "avec elle ses sept membres d'équipages. L'opinion publique est\n", "fortement touchée et lors de l'enquête qui suivra, la fiabilité des\n", "joints toriques sera directement mise en cause. Au delà des problèmes\n", "de communication interne à la NASA qui sont pour beaucoup dans ce\n", "fiasco, l'analyse précédente comporte (au moins) un petit\n", "problème... Saurez-vous le trouver ? Vous êtes libre de modifier cette\n", "analyse et de regarder ce jeu de données sous tous les angles afin\n", "d'expliquer ce qui ne va pas." ] } ], "metadata": { "celltoolbar": "Hide code", "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": "3.6.4" } }, "nbformat": 4, "nbformat_minor": 2 }