استفاده از روش بهينه سازي الگوريتم كلوني مورچگان در سيستم اطلاعات جغرافيايي

Application of ant colony optimization method in GIS

ﻣﻮﺿﻮع اﺳﺘﻔﺎده از روش ﻫﺎي ﻓﺮا-اﺑﺘﮑﺎري ﺑﺮاي ﮐﺎرﺑﺮد در ﻣﺴﺎﺋﻞ ﺑﻬﯿﻨﻪ ﺳﺎزي ﺗﺮﮐﯿﺒﯽ، زﻣﯿﻨﻪ ﺗﺤﻘﯿﻘﺎﺗﯽ اﺳﺖ ﮐﻪ ﺑﺎ ﺳﺮﻋﺖ در ﺣﺎل رﺷﺪ اﺳﺖ. اﯾﻦ اﻣﺮ ﺑﻪ دﻟﯿﻞ اﻫﻤﯿﺖ ﻣﺴﺎﺋﻞ ﺑﻬﯿﻨﻪ ﺳﺎزي ﺗﺮﮐﯿﺒﯽ در دﻧﯿﺎي ﺻﻨﻌﺖ و ﻋﻠﻢ اﺳﺖ. در ﺳﺎل ﻫﺎي اﺧﯿﺮ ﯾﮑﯽ از ﻣﻬﻤﺘﺮﯾﻦ و اﻣﯿﺪ ﺑﺨﺶ ﺗﺮﯾﻦ ﺗﺤﻘﯿﻘﺎت، »روش ﻫﺎي ﻓﺮا-اﺑﺘﮑﺎري ﺑﺮﮔﺮﻓﺘﻪ از ﻃﺒﯿﻌﺖ« ﺑﻮده اﺳﺖ ﮐﻪ در ﺣﻞ ﻣﺴﺎﺋﻞ ﻣﺸﮑﻞ ﺗﺮﮐﯿﺒﯽ 6ﻧﺘﺎﯾﺞ ﺑﺴﯿﺎر ﺧﻮﺑﯽ داﺷﺘﻪ اﺳﺖ. اﻟﮕﻮرﯾﺘﻢ ﻫﺎي ﻓﺮا-اﺑﺘﮑﺎري ﻫﻨﮕﺎﻣﯽ ﺑﺮاي ﺣﻞ ﯾﮏ ﻣﺴﺌﻠﻪ اﺳﺘﻔﺎده ﻣﯽ ﺷﻮﻧﺪ ﮐﻪ ﻫﻤﮕﺎم ﺑﺎ اﻓﺰاﯾﺶ اﺑﻌﺎد ﻣﺴﺌﻠﻪ ﻓﻀﺎي ﺷﺪﻧﯽ ﺑﻪ ﺻﻮرت ﭼﺸﻢﮔﯿﺮي اﻓﺰاﯾﺶ ﯾﺎﺑﺪ ﮐﻪ اﺻﻄﻼﺣﺎ اﯾﻨﮕﻮﻧﻪ ﻣﺴﺎﺋﻞ NP-hard ﻧﺎﻣﯿﺪه ﻣﯽ ﺷﻮﻧﺪ. ﯾﮑﯽ از روش ﻫﺎي ﻓﺮا-اﺑﺘﮑﺎري ﭘﺮﮐﺎرﺑﺮد در اﯾﻦ زﻣﯿﻨﻪ اﻟﮕﻮرﯾﺘﻢ ﺑﻬﯿﻨﻪ ﺳﺎزي ﮐﻠﻮﻧﯽ ﻣﻮرﭼﮕﺎن اﺳﺖ ﮐﻪ اﻣﺮوزه در ﺣﻞ ﻣﺴﺎﺋﻞ ﺗﺨﺼﯿﺺ ﻣﻨﺎﺑﻊ ﻣﮑﺎﻧﯽ، ﻣﺴﯿﺮﯾﺎﺑﯽ و ﻣﮑﺎن ﯾﺎﺑﯽ در ﻣﺤﯿﻂ ﻫﺎي ﺳﯿﺴﺘﻢ اﻃﻼﻋﺎت ﺟﻐﺮاﻓﯿﺎﯾﯽ ﮐﺎرﺑﺮد دارد. در اﯾﻦ ﺗﺤﻘﯿﻖ ﺿﻤﻦ ﺑﺮرﺳﯽ اﻟﮕﻮرﯾﺘﻢ ﮐﻠﻮﻧﯽ ﻣﻮرﭼﮕﺎن ﺑﻪ ﺑﯿﺎن و ﭘﺎراﻣﺘﺮﻫﺎي ﻣﻮرد ﻧﯿﺎز آن ﺑﺮاي اﺳﺘﻔﺎده در ﻣﺤﯿﻂ ﺳﯿﺴﺘﻢ اﻃﻼﻋﺎت ﺟﻐﺮاﻓﯿﺎﯾﯽ ﭘﺮداﺧﺘﻪ ﻣﯽ ﺷﻮد.

Swarm intelligence is one of the new growing methods that is considered in artificial intelligence as a function of the social interaction of components. The Basics of swarm intelligence are based on the study of the behavior of social organisms such as some insects (bees, ants, termites) or even humans. The issue of using meta-heuristic methods for application in hybrid optimization problems is a rapidly growing field of research. This is due to the importance of hybrid optimization issues in the world of industry and science. In recent years, one of the most important and promising researches has been "supra-innovative methods derived from nature", which has had very good results in solving problems of combined problems. Meta-heuristic algorithms are used to solve a problem when, as the size of the problem increases dramatically, so-called NP-hard problems. One of the most widely used meta-innovative methods in this field is the ant colony optimization algorithm, which is used today in solving the problems of spatial resource allocation, routing, and location in GIS environments. In this research, while examining the ant colony algorithm, its expression and parameters required for use in the GIS environment are discussed. The ability of algorithms based on food search in the ant colony algorithm is significantly dependent on the optimal determination of the parameters in these algorithms.