Scribd
Upload
30 views

Travelling Salesman Problem

The document summarizes using a simulated annealing algorithm to solve a traveling salesman problem instance (gil26) within the Heuristic Lab optimization framework. The best solution found had a quality of 2507 using generalized exponential annealing over 1200 iterations with an inner iteration count of 250 and start temperature of 100. This solution was within 192 of the best known quality for this instance and found within 40 seconds. The simulated annealing parameters that led to the best found solution are described.

Uploaded by

Vishwas J
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
30 views

Travelling Salesman Problem

The document summarizes using a simulated annealing algorithm to solve a traveling salesman problem instance (gil26) within the Heuristic Lab optimization framework. The best solution found had a quality of 2507 using generalized exponential annealing over 1200 iterations with an inner iteration count of 250 and start temperature of 100. This solution was within 192 of the best known quality for this instance and found within 40 seconds. The simulated annealing parameters that led to the best found solution are described.

Uploaded by

Vishwas J
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Travelling Salesman Problem

Report Structure: Simulated Annealing Algorithm (Vishwas Jagadeeswarachar, 05/12/2019)

Description of the task to be solved: Travelling Salesman Problem with a Simulated


Annealing algorithm using Heuristic Lab Optimization framework: Instance=gil26

Description of parameters that were used until best found solution was found:
The various parameters used in the experiments are Annealing Operator, Inner
Iterations, Maximum Iteration, Move Generator, Start Temperature, Quality,
Execution Time and Best Known Quality.

 Table

The best found solution:


 Quality
1. Sequence of cities
55,241,37,121,222,42,156,248,61,236,176,68,189,7,2,117,122,134,145,83,38,175,232,10,14,168,
180,170,58,161,74,218,51,242,15,5,164,221,84,36,43,198,250,207,245,24,140,125,115,214,103,
73,171,139,163,6,159,27,155,143,191,259,34,238,183,217,118,93,150,85,148,246,87,49,90,76,2
11,106,239,116,97,13,142,173,47,127,44,186,92,160,178,45,215,62,179,151,199,21,119,22,235,
195,110,251,254,104,91,177,209,40,226,28,19,108,109,59,157,133,158,146,244,126,39,16,23,20
0,50,96,131,66,31,77,225,135,258,174,54,46,167,65,102,112,79,193,196,41,9,237,130,213,56,20
2,192,100,252,113,147,25,229,8,152,12,166,88,247,120,105,188,219,35,182,149,57,169,1,194,2
6,260,172,216,82,205,64,52,227,136,17,256,123,48,11,210,89,60,129,128,220,204,30,70,228,0,1
38,201,230,67,141,231,98,208,243,107,69,154,75,153,86,33,181,94,187,257,3,80,114,144,99,18
5,20,137,253,233,32,71,184,18,78,261,132,4,190,29,53,203,197,72,206,165,95,212,111,249,124,
63,162,223,81,234,10,240,224,255

2. Visualisation of solution:

 Parameters of Simulated Annealing used to find the best solution:


The parameters used in Simulating Annealing to find best solution are
a) Annealing operator : Generalized Exponential
b) Inner Iterations: 250
c) Maximum Iterations: 1200
d) Move Generator: Stochastic Inversion Multi Move Generator
e) Start Temperature: 100

Conclusions:
Using Heuristic Lab Optimization framework , the Travelling Salesman
Problem with a Simulated Annealing algorithm was solved to find the optimal
Solution to Best Quality of the Problem given( Instance =gil26).
In my experiments with several Annealing Operators used, I found
Generalized Exponential gives a BestQuality (2507) close to Best Known Quality
(2378), with execution time 00:00:40.63seconds.
So this solution is optimal or better for 1200 iterations, with 192 Absolute
Difference between Best Known Quality and Best Quality for the Instance gil26.

You might also like