{"id":1541,"date":"2021-07-03T19:29:02","date_gmt":"2021-07-03T19:29:02","guid":{"rendered":"https:\/\/labs.newsci.ai\/?p=1541"},"modified":"2021-07-03T19:31:41","modified_gmt":"2021-07-03T19:31:41","slug":"1541","status":"publish","type":"post","link":"https:\/\/labs.newsci.ai\/?p=1541","title":{"rendered":"Route optimization of a pub crawl with ORS and"},"content":{"rendered":"\n

<\/p>\n\n\n\n

\n
\n<\/div>
\n
\n

Route optimization of a pub crawl with ORS and ortools<\/code>\u00b6<\/a><\/h1>\n<\/div>\n<\/div>\n<\/div>\n
\n<\/div>
\n
\n

It's this of the year again (or will be in 6 months): the freshmen pour into the institute and as the diligent student council you are, you want to welcome them for their geo adventure with a stately pub crawl to prepare them for the challenges lying ahead.<\/p>\n

We want to give you the opportunity to route the pack of rookies in a fairly optimal way:<\/p>\n\n<\/div>\n<\/div>\n<\/div>\n

\n
\n
In\u00a0[6]:<\/div>\n
\n
\n
<\/span>mkdir<\/span> ors<\/span>-<\/span>pubcrawl<\/span>\nconda<\/span> create<\/span> -<\/span>n<\/span> ors<\/span>-<\/span>pubcrawl<\/span> python<\/span>=<\/span>3.6<\/span> folium<\/span> shapely<\/span>\ncd<\/span> ors<\/span>-<\/span>pubcrawl<\/span>\npip<\/span> install<\/span> openrouteservice<\/span> ortools<\/span>\n<\/pre><\/div>\n\n    <\/div>\n<\/div>\n<\/div>\n\n<\/div>\n
\n
\n
In\u00a0[1]:<\/div>\n
\n    
\n
<\/span>import<\/span> folium<\/span>\nfrom<\/span> shapely<\/span> import<\/span> wkt<\/span>,<\/span> geometry<\/span>\nimport<\/span> json<\/span>\nfrom<\/span> pprint<\/span> import<\/span> pprint<\/span>\n<\/pre><\/div>\n\n    <\/div>\n<\/div>\n<\/div>\n\n<\/div>\n
\n<\/div>
\n
\n
Now we're ready to start our most optimally planned pub crawl ever through hipster Kreuzberg! It will also be the most un-hipster pub crawl ever, as we'll cover ground with a taxi. At least it's safer than biking half-delirious.<\/p>\n
First the basic parameters: API key and the district polygon to limit our pub search. The Well Known Text was prepared in QGIS from Berlin authority's WFS<\/a> (QGIS field calculator has a geom_to_wkt<\/code> method). BTW, Berlin, hope you don't wonder why your feature services are so slow... Simplify is the magic word, simplify.<\/p>\n\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In\u00a0[2]:<\/div>\n
\n    
\n
<\/span>api_key<\/span> =<\/span> 'your_key'<\/span>\nwkt_str<\/span> =<\/span> 'Polygon ((13.43926404 52.48961046, 13.42040115 52.49586382, 13.42541101 52.48808523, 13.42368155 52.48635829, 13.40788599 52.48886084, 13.40852944 52.487142, 13.40745989 52.48614988, 13.40439187 52.48499746, 13.40154731 52.48500125, 13.40038591 52.48373202, 13.39423818 52.4838664, 13.39425346 52.48577149, 13.38629096 52.48582648, 13.38626853 52.48486362, 13.3715694 52.48495055, 13.37402099 52.4851697, 13.37416365 52.48771105, 13.37353615 52.48798191, 13.37539925 52.489432, 13.37643416 52.49167597, 13.36821531 52.49333093, 13.36952826 52.49886974, 13.37360623 52.50416333, 13.37497726 52.50337776, 13.37764916 52.5079675, 13.37893813 52.50693045, 13.39923153 52.50807711, 13.40022883 52.50938108, 13.40443425 52.50777471, 13.4052848 52.50821063, 13.40802944 52.50618019, 13.40997081 52.50692569, 13.41152096 52.50489127, 13.41407284 52.50403794, 13.41490921 52.50491634, 13.41760145 52.50417013, 13.41943091 52.50564912, 13.4230412 52.50498109, 13.42720031 52.50566607, 13.42940229 52.50857222, 13.45335235 52.49752496, 13.45090795 52.49710803, 13.44765912 52.49472124, 13.44497623 52.49442276, 13.43926404 52.48961046))'<\/span>\n\naoi_geom<\/span> =<\/span> wkt<\/span>.<\/span>loads<\/span>(<\/span>wkt_str<\/span>)<\/span> # load geometry from WKT string<\/span>\n\naoi_coords<\/span> =<\/span> list<\/span>(<\/span>aoi_geom<\/span>.<\/span>exterior<\/span>.<\/span>coords<\/span>)<\/span> # get coords from exterior ring<\/span>\naoi_coords<\/span> =<\/span> [(<\/span>y<\/span>,<\/span>x<\/span>)<\/span> for<\/span> x<\/span>,<\/span>y<\/span> in<\/span> aoi_coords<\/span>]<\/span> # swap (x,y) to (y,x). Really leaflet?!<\/span>\naoi_centroid<\/span> =<\/span> aoi_geom<\/span>.<\/span>centroid<\/span> # Kreuzberg center for map center<\/span>\n<\/pre><\/div>\n\n    <\/div>\n<\/div>\n<\/div>\n\n<\/div>\n
\n<\/div>
\n
\n
Next, add the Kreuzberg polygon as marker to the map, so we get a bit of orientation.<\/p>\n\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In\u00a0[4]:<\/div>\n
\n    
\n
<\/span>m<\/span> =<\/span> folium<\/span>.<\/span>Map<\/span>(<\/span>tiles<\/span>=<\/span>'Stamen Toner'<\/span>,<\/span>location<\/span>=<\/span>(<\/span>aoi_centroid<\/span>.<\/span>y<\/span>,<\/span> aoi_centroid<\/span>.<\/span>x<\/span>),<\/span> zoom_start<\/span>=<\/span>14<\/span>)<\/span>\nfolium<\/span>.<\/span>vector_layers<\/span>.<\/span>Polygon<\/span>(<\/span>aoi_coords<\/span>,<\/span>\n                                     color<\/span>=<\/span>'#ffd699'<\/span>,<\/span>\n                                     fill_color<\/span>=<\/span>'#ffd699'<\/span>,<\/span>\n                                    fill_opacity<\/span>=<\/span>0.2<\/span>,<\/span>\n                                     weight<\/span>=<\/span>3<\/span>)<\/span>.<\/span>add_to<\/span>(<\/span>m<\/span>)<\/span>\nm<\/span>\n<\/pre><\/div>\n\n    <\/div>\n<\/div>\n<\/div>\n\n
\n
\n\n\n
\n\n    
Out[4]:<\/div>\n\n\n\n
\n