pgr_TSPeuclidean

_images/boost-inside.jpeg

Boost Graph Inside

Availability:

Description

Problem Definition

The travelling salesperson problem (TSP) asks the following question:

Given a list of cities and the distances between each pair of cities, which is the shortest possible route that visits each city exactly once and returns to the origin city?

General Characteristics

  • This problem is an NP-hard optimization problem.
  • Metric Algorithm is used
  • Implementation generates solutions that are twice as long as the optimal tour in the worst case when:
    • Graph is undirected
    • Graph is fully connected
    • Graph where traveling costs on edges obey the triangle inequality.
  • On an undirected graph:
    • The traveling costs are symmetric:
    • Traveling costs from u to v are just as much as traveling from v to u

Characteristics

  • Duplicated identifiers with different coordinates are not allowed

    • The coordinates are quite the same for the same identifier, for example

      1, 3.5, 1
      1, 3.499999999999 0.9999999
      
    • The coordinates are quite different for the same identifier, for example

      2 , 3.5, 1.0
      2 , 3.6, 1.1
      
    • Any duplicated identifier will be ignored. The coordinates that will be kept is arbitrarly.

Signatures

Summary

pgr_TSPeuclidean(Coordinates SQL, [start_id], [end_id])
RETURNS SETOF (seq, node, cost, agg_cost)
Example:With default values
SELECT * FROM pgr_TSPeuclidean(
    $$
    SELECT id, st_X(the_geom) AS x, st_Y(the_geom)AS y  FROM edge_table_vertices_pgr
    $$);
 seq | node |      cost      |   agg_cost
-----+------+----------------+---------------
   1 |    1 |              0 |             0
   2 |    2 |              1 |             1
   3 |    8 |  1.41421356237 | 2.41421356237
   4 |    7 |              1 | 3.41421356237
   5 |   14 |  1.58113883008 | 4.99535239246
   6 |   15 |            1.5 | 6.49535239246
   7 |   13 |            0.5 | 6.99535239246
   8 |   17 |            1.5 | 8.49535239246
   9 |   12 |  1.11803398875 | 9.61338638121
  10 |    9 |              1 | 10.6133863812
  11 |   16 | 0.583095189485 | 11.1964815707
  12 |    6 | 0.583095189485 | 11.7795767602
  13 |   11 |              1 | 12.7795767602
  14 |   10 |              1 | 13.7795767602
  15 |    5 |              1 | 14.7795767602
  16 |    4 |   2.2360679775 | 17.0156447377
  17 |    3 |              1 | 18.0156447377
  18 |    1 |  1.41421356237 |    19.4298583
(18 rows)

Parameters

Parameter Type Default Description
Coordinates SQL TEXT   An SQL query, described in the Coordinates SQL section
start_vid BIGINT 0

The first visiting vertex

  • When 0 any vertex can become the first visiting vertex.
end_vid BIGINT 0

Last visiting vertex before returning to start_vid.

  • When 0 any vertex can become the last visiting vertex before returning to start_vid.
  • When NOT 0 and start_vid = 0 then it is the first and last vertex

Inner query

Coordinates SQL

Coordinates SQL: an SQL query, which should return a set of rows with the following columns:

Column Type Description
id ANY-INTEGER Identifier of the starting vertex.
x ANY-NUMERICAL X value of the coordinate.
y ANY-NUMERICAL Y value of the coordinate.

Result Columns

Returns SET OF (seq, node, cost, agg_cost)

Column Type Description
seq INTEGER Row sequence.
node BIGINT Identifier of the node/coordinate/point.
cost FLOAT

Cost to traverse from the current node to the next node in the path sequence.

  • 0 for the last row in the tour sequence.
agg_cost FLOAT

Aggregate cost from the node at seq = 1 to the current node.

  • 0 for the first row in the tour sequence.

Additional Examples

Example:Test 29 cities of Western Sahara

This example shows how to make performance tests using University of Waterloo’s example data using the 29 cities of Western Sahara dataset

Creating a table for the data and storing the data

CREATE TABLE wi29 (id BIGINT, x FLOAT, y FLOAT, geom geometry);
INSERT INTO wi29 (id, x, y) VALUES
(1,20833.3333,17100.0000),
(2,20900.0000,17066.6667),
(3,21300.0000,13016.6667),
(4,21600.0000,14150.0000),
(5,21600.0000,14966.6667),
(6,21600.0000,16500.0000),
(7,22183.3333,13133.3333),
(8,22583.3333,14300.0000),
(9,22683.3333,12716.6667),
(10,23616.6667,15866.6667),
(11,23700.0000,15933.3333),
(12,23883.3333,14533.3333),
(13,24166.6667,13250.0000),
(14,25149.1667,12365.8333),
(15,26133.3333,14500.0000),
(16,26150.0000,10550.0000),
(17,26283.3333,12766.6667),
(18,26433.3333,13433.3333),
(19,26550.0000,13850.0000),
(20,26733.3333,11683.3333),
(21,27026.1111,13051.9444),
(22,27096.1111,13415.8333),
(23,27153.6111,13203.3333),
(24,27166.6667,9833.3333),
(25,27233.3333,10450.0000),
(26,27233.3333,11783.3333),
(27,27266.6667,10383.3333),
(28,27433.3333,12400.0000),
(29,27462.5000,12992.2222);

Adding a geometry (for visual purposes)

UPDATE wi29 SET geom = ST_makePoint(x,y);

Getting a total cost of the tour, compare the value with the length of an optimal tour is 27603, given on the dataset

SELECT *
FROM pgr_TSPeuclidean($$SELECT * FROM wi29$$)
WHERE seq = 30;
 seq | node |     cost      |   agg_cost
-----+------+---------------+---------------
  30 |    1 | 2266.91173136 | 28777.4854127
(1 row)

Getting a geometry of the tour

WITH
tsp_results AS (SELECT seq, geom FROM pgr_TSPeuclidean($$SELECT * FROM wi29$$) JOIN wi29 ON (node = id))
SELECT ST_MakeLine(ARRAY(SELECT geom FROM tsp_results ORDER BY seq));
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    st_makeline
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
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
(1 row)

Visualy, The first image is the optimal solution and the second image is the solution obtained with pgr_TSPeuclidean.

_images/wi29optimal.png _images/wi29Solution.png