osm-and-gtfs-generation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:33:43.444185Z",
     "start_time": "2020-04-02T15:33:42.973737Z"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# OSM & GTFS Generation\n",
    "\n",
    "Here you can build a simple **road network** and **transit route** for a fake city. We use these outputs for testing network combination tooling.\n",
    "\n",
    "**Outputs:**\n",
    "- **OSM Road Network**\n",
    "- **GTFS Bus Transit**\n",
    "\n",
    "These outputs are based on a sequential generation of the following city features:\n",
    "\n",
    "**Facilities:** Generate activity locations (including households) clustered around centres of density.\n",
    "\n",
    "**Road Networks:** Generate a road network based on facility density.\n",
    "\n",
    "**Road Transit Routes:** Generate road transit routes using the road network.\n",
    "\n",
    "We additionally apply spatial some jitter to add noise."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:33:45.681657Z",
     "start_time": "2020-04-02T15:33:43.601773Z"
    }
   },
   "outputs": [],
   "source": [
    "from citychef import spatial\n",
    "from citychef import graph\n",
    "from citychef import household as hh\n",
    "from citychef import tree\n",
    "from citychef import osm\n",
    "from citychef import gtfs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:33:53.700507Z",
     "start_time": "2020-04-02T15:33:53.633323Z"
    }
   },
   "outputs": [],
   "source": [
    "# uncomment if you are making changes to core modules\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "%reload_ext autoreload"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:33:56.163535Z",
     "start_time": "2020-04-02T15:33:56.134246Z"
    }
   },
   "outputs": [],
   "source": [
    "city_dir = \"./osm_gtfs_example\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:33:57.330188Z",
     "start_time": "2020-04-02T15:33:57.305222Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING - potentially overwritting previous results\n"
     ]
    }
   ],
   "source": [
    "if not os.path.exists(city_dir):\n",
    "    os.mkdir(city_dir)\n",
    "else:\n",
    "    print(\"WARNING - potentially overwritting previous results\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Facilities\n",
    "First we generate the location of various facilities, including households where the populations will live and activity locations such work places, leisure, education, shopping and health facilities.\n",
    "\n",
    "#### Method\n",
    "1. define seed bounding box (DEGREES) - TODO add reprojection\n",
    "2. uniformly sample 'centres' within bounding box\n",
    "3. poisson sample number of facilities of each type around each centre\n",
    "4. gaussian sample locations of facilities around each centre"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:33:58.938345Z",
     "start_time": "2020-04-02T15:33:58.904751Z"
    }
   },
   "outputs": [],
   "source": [
    "bbox = np.array([[0,0],[.01,.01]])  # min x, min y, max x, max y (DEGREES)\n",
    "centres = spatial.Centres(bbox, number=1)\n",
    "facilities = {\n",
    "    'households': spatial.Clusters(centres, size=1000, sigma=.005),\n",
    "}\n",
    "bbox = spatial.collect_bbox(facilities) # adjust the bbox for max extends"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Road Network\n",
    "\n",
    "We build a road network for car travel. We use a quad tree to define the density of the network so that accessibility is likely better where household density is higher. The connectivity of the resulting networks is based on parent - child relationships within the quad tree. This results in poor connectivity between neighbouring quad grids if they are removed by several generations, ie their shared parent is distant.\n",
    "\n",
    "Edges are assigned a distance, a freespeed (based on distance) and a freeflow traversal time.\n",
    "\n",
    "#### Method\n",
    "\n",
    "1. Build a grid tree with maximum household density of N.\n",
    "2. Build a network of roads from the grid tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:34:01.300089Z",
     "start_time": "2020-04-02T15:34:00.886110Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/networkx/drawing/nx_pylab.py:579: MatplotlibDeprecationWarning: \n",
      "The iterable function was deprecated in Matplotlib 3.1 and will be removed in 3.3. Use np.iterable instead.\n",
      "  if not cb.iterable(width):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(-0.012840569945441029,\n",
       " 0.021910712861664156,\n",
       " -0.014253964598006267,\n",
       " 0.018251237111683753)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "car_network = graph.TreeNetwork(\n",
    "    bbox, facilities['households'], grid='regular', max_points=200\n",
    ")\n",
    "car_network.plot()\n",
    "plt.axis('equal')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bus Network\n",
    "\n",
    "We build a bus network for public transit travel on the road network. A route generation algorithm attempts to build viable routes by making weighted random walks across the road network from seed locations.\n",
    "\n",
    "The random walks are seeded based on a weighted sample from the road network vertices. Where the vertices are weighted by household density at each vertices. From each seed location the weighted random walk traverses the possible network based on weighted random choice of each possible step. Where weights are based on the household density of the possible choices, the straightness of the resulting graph and the number of times the edges have already been traversed by a route. Routes cannot repeat the previous edge so that the traversal ends at dead ends in the road network or when a maximum number of stops is reached.\n",
    "\n",
    "Edges are assigned a distance, a freespeed (based on distance) and a freeflow traversal time as per the underlying road network. For simplicity, no consideration is made of a schedule or interchange times.\n",
    "\n",
    "#### Method\n",
    "\n",
    "1. Build an array of houshold density for each vertices of the road network graph\n",
    "2. Sample number of required routes based on population and a poisson distribution\n",
    "3. Generate route\n",
    "4. Add generated routes that meet minimum requirements for stops and length\n",
    "5. Combine routes to form total bus network graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:34:03.853581Z",
     "start_time": "2020-04-02T15:34:03.811846Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<citychef.graph.PTRoute at 0x123cb8750>]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "buses = graph.Transit(car_network, facilities['households'], density_radius=.0001)\n",
    "buses.build_routes(num_routes=1, max_length=11, min_length=11, straightness=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Apply Noise to the Transit Network\n",
    "\n",
    "To add noise to our test data we:\n",
    "\n",
    "i) add interpolate some new transit stops so that the stop locations do not have corresponding road network nodes.\n",
    "\n",
    "ii) add some spatial jitter to transit stop locations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:34:05.818615Z",
     "start_time": "2020-04-02T15:34:05.791172Z"
    }
   },
   "outputs": [],
   "source": [
    "buses.interpolate_routes()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:34:06.302698Z",
     "start_time": "2020-04-02T15:34:06.274631Z"
    }
   },
   "outputs": [],
   "source": [
    "buses.jitter_locations(30/111139)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:34:07.754334Z",
     "start_time": "2020-04-02T15:34:06.832882Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/networkx/drawing/nx_pylab.py:579: MatplotlibDeprecationWarning: \n",
      "The iterable function was deprecated in Matplotlib 3.1 and will be removed in 3.3. Use np.iterable instead.\n",
      "  if not cb.iterable(width):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(-0.012259762404143818,\n",
       " 0.021329905320366972,\n",
       " -0.013657771426476029,\n",
       " 0.017655043940153513)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x1728 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1, figsize=(24, 24))\n",
    "car_network.plot(ax=ax)\n",
    "buses.plot(ax=ax)\n",
    "plt.axis('equal')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:35:18.731548Z",
     "start_time": "2020-04-02T15:35:18.700309Z"
    }
   },
   "outputs": [],
   "source": [
    "# fig.savefig('image.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:35:18.992851Z",
     "start_time": "2020-04-02T15:35:18.952287Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "108.68682991072858\n",
      "685.4283796357361\n"
     ]
    }
   ],
   "source": [
    "# check min and max distances\n",
    "print(buses.min_link_length * 111139)\n",
    "print(buses.max_link_length * 111139) # approx conv to m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-31T18:00:28.881332Z",
     "start_time": "2020-03-31T18:00:28.735018Z"
    }
   },
   "source": [
    "### Outputs\n",
    "\n",
    "Convert the road network to OSM format xml and the transit to GTFS."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:35:20.350057Z",
     "start_time": "2020-04-02T15:35:20.057246Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6eaa75a515a14b3bba53713a82a6992b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a1cba1ea93a4445b9a58461b64fd48b6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "osm.nx_to_osm(\n",
    "    g=car_network.g,\n",
    "    path=os.path.join(city_dir, \"osm.xml\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-04-02T15:35:21.848413Z",
     "start_time": "2020-04-02T15:35:21.425756Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e5633dfe91b4e7f8b4af65db5b6cb4e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gtfs.build_gtfs(\n",
    "    transit=buses,\n",
    "    out_dir=city_dir,\n",
    "    agency_id=0,\n",
    "    agency_name='test_bus_inc'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}