‫‪569‬‬ ‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪ ،1385‬ﺍﺯ ﺻﻔﺤﻪ ‪ 569‬ﺗﺎ ‪583‬‬

‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ ﺗﺼﻤﯿﻢﯾﺎﺭ ﺑﺮﺍی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺑﺎ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ‬

‫ﺭوﺵ ﺍﺳﺘﺪﻻﻝ ﻣﺒﺘﻨﯽ ﺑﺮ ﻣﻮﺭﺩ ﺩﺭ ﻣﺤﯿﻂ ﻓﺎﺯی‬

‫ﻓﺮﻫﺎﺩ ﻓﺎﺋﺰ‬
‫ﺩﺍﻧﺸﺠﻮی ﺩﮐﺘﺮی ﻣﻬﻨﺪﺳﯽ ﺻﻨﺎﯾﻊ ‪ -‬ﺩﺍﻧﺸﮕﺎﻩ ﺻﻨﻌﺘﯽ ﺍﻣﯿﺮ ﮐﺒﯿﺮ‬
‫ﺳﯿﺪﺣﺴﻦ ﻗﺪﺳﯽﭘﻮﺭ‬
‫ﺩﺍﻧﺸﯿﺎﺭ ﺩﺍﻧﺸﮑﺪۀ ﻣﻬﻨﺪﺳﯽ ﺻﻨﺎﯾﻊ ‪ -‬ﺩﺍﻧﺸﮕﺎﻩ ﺻﻨﻌﺘﯽ ﺍﻣﯿﺮ ﮐﺒﯿﺮ‬
‫‪Ghodsypo@aut.ac.ir‬‬
‫ﻣﻬﺪی ﻏﻀﻨﻔﺮی‬
‫ﺩﺍﻧﺸﯿﺎﺭ ﺩﺍﻧﺸﮑﺪۀ ﻣﻬﻨﺪﺳﯽ ﺻﻨﺎﯾﻊ – ﺩﺍﻧﺸﮕﺎﻩ ﻋﻠﻢ و ﺻﻨﻌﺖ ﺍﯾﺮﺍﻥ‬
‫)ﺗﺎﺭﯾﺦ ﺩﺭﯾﺎﻓﺖ ‪ ،84/8/30‬ﺗﺎﺭﯾﺦ ﺩﺭﯾﺎﻓﺖ ﺭوﺍﯾﺖ ﺍﺻﻼﺡ ﺷﺪﻩ ‪ ، 85/2/2‬ﺗﺎﺭﯾﺦ ﺗﺼﻮﯾﺐ ‪(85/3/6‬‬
‫ﭼﮑﯿﺪﻩ‬
‫ﻣﻘﺎﻟﻪ ﺣﺎﺿﺮ ﮐﺎﺭﺑﺮﺩ ﺭوﺵ "ﺍﺳﺘﺪﻻﻝ ﻣﺒﺘﻨﯽ ﺑﺮ ﻣﻮﺭﺩ" ﯾﺎ )‪ Case-Based Reasoning (CBR‬ﺭﺍ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﻣﻮﺭﺩ‬
‫ﺗﻮﺟﻪ ﻗﺮﺍﺭ ﻣﯽﺩﻫﺪ ﮐﻪ ﺑﺮ ﻣﺒﻨﺎی ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺗﺠﺮﺑﯿﺎﺕ ﮔﺬﺷﺘﻪ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﺋﻞ ﻣﺸﺎﺑﻪ ﺟﺪﯾﺪ ﺍﺳﺘﻮﺍﺭ ﺍﺳﺖ و ﺍﺧﯿﺮﺍً ﻧﯿﺰ ﺑﺮﺍی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﮑﺎﺭ‬
‫ﮔﺮﻓﺘﻪ ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ ﺑﻪ ﺑﻤﻨﻈﻮﺭ ﭘﻮﺷﺶ ﺩﺍﺩﻥ ﻣﺸﺨﺼﻪﻫﺎﺋﯽ ﮐﻪ ﻋﻤﻠﮑﺮﺩ ﺁﻧﻬﺎ ﺩﺭ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﺼﻮﺭﺕ ﻏﯿﺮﺩﻗﯿﻖ ﺍﺭﺯﯾﺎﺑﯽ ﻣﯽﺷﻮﻧﺪ‪،‬‬
‫ﺍﺯ ﻗﺎﺑﻠﯿﺘﻬﺎی ﻣﻨﻄﻖ ﻓﺎﺯی ﺑﻬﺮﻩﺑﺮﺩﺍﺭی ﺷﺪﻩ و ﭘﺎﺭﺍﻣﺘﺮﻫﺎی ﻧﺎﺩﻗﯿﻖ ﺑﮑﻤﮏ ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﺑﺎ ﺗﻮﺍﺑﻊ ﻋﻀﻮﯾﺖ ﺧﻄﯽ ﻣﺸﺨﺺ ﺷﺪﻩﺍﻧﺪ‪ .‬ﻋﻼوﻩ ﺑﺮ ﺁﻥ ﺑﻤﻨﻈﻮﺭ‬
‫ﺑﺎﻻ ﺑﺮﺩﻥ ﺩﻗﺖ و ﮐﺎﺭﺍﺋﯽ ﻓﺮﺁﯾﻨﺪ ﺑﺎﺯﯾﺎﺑﯽ ﻓﺮوﺷﻨﺪۀ ﻣﻨﺎﺳﺐ ﺍﺯ ﻣﯿﺎﻥ ﮔﺰﯾﻨﻪﻫﺎی ﻣﻮﺟﻮﺩ ﺩﺭ ﺑﺎﻧﮏ ﺍﻃﻼﻋﺎﺗﯽ ﻓﺮوﺷﻨﺪﮔﺎﻥ‪ ،‬ﺭوﺵ ‪ TOPSIS‬ﻓﺎﺯی ﺑﺮﺍی‬
‫ﺍﺭﺯﯾﺎﺑﯽ و ﺍوﻟﻮﯾﺖﺑﻨﺪی ﻧﻬﺎﺋﯽ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﭘﯿﺸﻨﻬﺎﺩ و ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬

‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ‪ ،‬ﺍﺳﺘﺪﻻﻝ ﻣﺒﺘﻨﯽ ﺑﺮ ﻣﻮﺭﺩ‪ ،‬ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی ‪ ،‬ﺍﻋﺪﺍﺩ ﻓﺎﺯی‪TOPSIS ،‬‬ ‫وﺍژﻩ ﻫﺎی ﮐﻠﯿﺪی ‪:‬‬
‫ﻓﺎﺯی‬

‫ﺗﮑﺮﺍﺭی ﻣﻄﺮﺡ ﺍﺳﺖ‪ ،‬ﻣﺪﯾﺮﺍﻥ ﺧﺮﯾﺪ ﺑﻄﻮﺭ ﻃﺒﯿﻌﯽ ﺍﺯ ﺗﺠﺮﺑﮥ‬
‫ﺧﺮﯾﺪﻫﺎی ﻗﺒﻠﯽ ﺑﺮﺍی ﺗﺼﻤﯿﻢﮔﯿﺮی ﭘﯿﺮﺍﻣﻮﻥ ﺍﻧﺘﺨﺎﺏ‬
‫ﻓﺮوﺷﻨﺪﻩ ﺑﺮﺍی ﺍﻧﺠﺎﻡ ﺧﺮﯾﺪﻫﺎی ﻣﺸﺎﺑﻪ ﺟﺪﯾﺪ‪ ،‬ﺍﺳﺘﻔﺎﺩﻩ‬
‫ﻣﯽﻧﻤﺎﯾﻨﺪ‪ .‬ﺑﻨﺎﺑﺮﺍﯾﻦ ﺍﺭﺍﺋﮥ ﺭوﺵ ﻧﻈﺎﻡﻣﻨﺪی ﮐﻪ ﺑﺘﻮﺍﻧﺪ ﺍﺯ ﺍﯾﻦ‬
‫ﺭﻓﺘﺎﺭ ﻃﺒﯿﻌﯽ ﭘﺸﺘﯿﺒﺎﻧﯽ ﻧﻤﺎﯾﺪ‪ ،‬ﺑﻪ ﺍﺗﺨﺎﺫ ﺗﺼﻤﯿﻢ ﻣﻨﺎﺳﺐ‬
‫ﮐﻤﮏ ﻗﺎﺑﻞ ﺗﻮﺟﻬﯽ ﺧﻮﺍﻫﺪ ﻧﻤﻮﺩ‪ .‬ﺭوﺵ "ﺍﺳﺘﺪﻻﻝ ﻣﺒﺘﻨﯽ ﺑﺮ‬
‫ﻣﻮﺭﺩ" ﮐﻪ ﺑﺮ ﺍﺳﺎﺱ ﺩﯾﺪﮔﺎﻩ "ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭﺍﻩﺣﻞﻫﺎی ﺍﺭﺍﺋﻪ‬
‫ﺷﺪﻩ ﺑﺮﺍی ﻣﺴﺎﺋﻞ ﺣﻞ ﺷﺪۀ ﻗﺒﻠﯽ ﺑﻤﻨﻈﻮﺭ ﺣﻞ ﻣﺴﺎﺋﻞ ﻣﺸﺎﺑﻪ‬
‫ﺟﺪﯾﺪ" ﺍﯾﺠﺎﺩ ﺷﺪﻩ ﺍﺳﺖ‪ ،‬ﻣﯽﺗﻮﺍﻧﺪ ﺍﺑﺰﺍﺭ ﻣﻨﺎﺳﺒﯽ ﺭﺍ ﺑﺮﺍی‬
‫ﺗﺼﻤﯿﻢ ﮔﯿﺮی ﺩﺭ ﺍﯾﻦ ﺭﺍﺑﻄﻪ ﻓﺮﺍﻫﻢ ﺁوﺭﺩ‪.‬‬
‫ﺩﺭ ﻣﺮوﺭ ﺗﺤﻘﯿﻘﺎﺕ ﭘﯿﺸﯿﻦ ﮐﻪ ﺩﺭ ﺑﺨﺶ ﺑﻌﺪ ﺑﻪ ﺁﻥ‬
‫ﺧﻮﺍﻫﯿﻢ ﭘﺮﺩﺍﺧﺖ‪ ،‬ﺑﻪ ﻣﻮﺍﺭﺩی ﺍﺯ ﮐﺎﺭﺑﺮﺩ ﺍﯾﻦ ﺭوﺵ ‪ -‬ﮐﻪ ﺩﺭ‬
‫ﺍﺩﺍﻣﮥ ﻣﻄﺎﻟﺐ ﺑﻄﻮﺭ ﻣﺨﺘﺼﺮ ‪ CBR‬ﻧﺎﻣﯿﺪﻩ ﺧﻮﺍﻫﺪ ﺷﺪ ‪ -‬ﺩﺭ‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﺮﻣﯽﺧﻮﺭﯾﻢ‪ .‬ﺩﺭ ﺍﯾﻦ ﻧﻮﺷﺘﺎﺭ‪ ،‬ﻣﺪﻟﯽ ﺑﺮ‬
‫ﻣﻘﺪﻣﻪ‬
‫ﻫﺰﯾﻨﮥ ﺗﺎﻣﯿﻦ ﻣﻮﺍﺩ ﺍوﻟﯿﻪ و ﻗﻄﻌﺎﺕ ﺗﺮﮐﯿﺒﯽ‪ 1‬ﺍﺯ ﻃﺮﯾﻖ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ‪ ،‬ﺑﺨﺶ ﻗﺎﺑﻞ ﺗﻮﺟﻬﯽ ﺍﺯ ﻫﺰﯾﻨﮥ ﺗﻤﺎﻡ ﺷﺪۀ ﮐﺎﻻﻫﺎ‬
‫ﺭﺍ ﺗﺸﮑﯿﻞ ﻣﯽﺩﻫﺪ‪ .‬ﺑﻄﻮﺭ ﻣﺘﻮﺳﻂ ‪ 70‬ﺩﺭﺻﺪ ﺍﺭﺯﺵ ﻣﺤﺼﻮﻝ‬
‫ﻧﻬﺎﺋﯽ ﮐﺎﺭﺧﺎﻧﺠﺎﺕ ﺭﺍ ﻫﺰﯾﻨﮥ ﺧﺮﯾﺪ ﻣﻮﺍﺩ ﺧﺎﻡ و ﺧﺪﻣﺎﺕ‬
‫ﺩﺭﯾﺎﻓﺘﯽ ﺍﺯ ﺑﯿﺮوﻥ ﺗﺸﮑﯿﻞ ﻣﯽﺩﻫﺪ]‪ .[1‬ﺍﯾﻦ ﻧﺴﺒﺖ ﺩﺭ‬
‫ﺷﺮﮐﺘﻬﺎی ﺑﺎ ﺗﮑﻨﻮﻟﻮژی ﺑﺎﻻ‪ ،‬ﺣﺘﯽ ﺑﻪ ‪ 80‬ﺩﺭﺻﺪ ﻧﯿﺰ ﺑﺎﻟﻎ‬
‫ﻣﯽﮔﺮﺩﺩ]‪ .[2‬ﺍﺯ ﺍﯾﻦ ﺭو ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﺮﺍی ﺷﺮﮐﺘﻬﺎ ﺍﺯ‬
‫ﺍﻫﻤﯿﺖ ﻓﻮﻕﺍﻟﻌﺎﺩﻩﺍی ﺑﺮﺧﻮﺭﺩﺍﺭ ﺍﺳﺖ‪ .‬ﺑﻪ ﻫﻤﺎﻥ ﺍﻧﺪﺍﺯﻩ ﮐﻪ‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﻣﻨﺎﺳﺐ ﺩﺭ ﺗﻘﻠﯿﻞ ﻫﺰﯾﻨﻪﻫﺎ ﻣﻮﺛﺮ ﺍﺳﺖ و‬
‫ﺑﺎﻋﺚ ﺍﻓﺰﺍﯾﺶ ﻗﺪﺭﺕ ﺭﻗﺎﺑﺖ ﺷﺮﮐﺘﻬﺎ ﻣﯽﺷﻮﺩ‪ ،‬ﺍﻧﺘﺨﺎﺏ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﻧﺎﻣﻨﺎﺳﺐ ﻧﯿﺰ ﻣﯽﺗﻮﺍﻧﺪ ﺑﺎﻋﺚ ﺗﻨﺰﻝ ﻣﻮﻗﻌﯿﺖ ﻣﺎﻟﯽ‬
‫و ﻋﻤﻠﯿﺎﺗﯽ ﺷﺮﮐﺘﻬﺎ ﺷﻮﺩ‪.‬‬
‫ﺩﺭ ﺷﺮﮐﺘﻬﺎﺋﯽ ﮐﻪ ﺗﺎﻣﯿﻦ ﻣﻮﺍﺩ ﺍوﻟﯿﻪ و ﺑﻪ ﺗﺒﻊ ﺁﻥ ﻓﺮﺁﯾﻨﺪ‬
‫ﺧﺮﯾﺪ و ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺑﻌﻨﻮﺍﻥ ﯾﮏ ﻓﻌﺎﻟﯿﺖ ﺭوﺯﻣﺮﻩ و‬
 ‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪1385‬‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ وﺯﻥ ﺍﺧﺘﺼﺎﺹ ﺩﺍﺩﻩ ﻣﯽﺷﻮﺩ‪ ،‬ﺳﭙﺲ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺩﺭ ﺭﺍﺑﻄﻪ ﺑﺎ ﻫﺮ ﮐﺪﺍﻡ ﺍﺯ ﺍﯾﻦ ﻣﻌﯿﺎﺭﻫﺎ ﻣﻮﺭﺩ‬
‫ﺍﺭﺯﯾﺎﺑﯽ ﻗﺮﺍﺭ ﮔﺮﻓﺘﻪ و ﺍﻣﺘﯿﺎﺯ ﻫﺮ ﮐﺪﺍﻡ ﻣﺸﺨﺺ ﻣﯽﮔﺮﺩﺩ‪ .‬ﺩﺭ‬
‫ﻧﻬﺎﯾﺖ ﺍﻣﺘﯿﺎﺯ ﻣﻮﺯوﻥ ﻫﺮ ﻓﺮوﺷﻨﺪﻩ‪ ،‬ﮐﻪ ﺍﺯ ﻣﺠﻤﻮﻉ ﺣﺎﺻﻠﻀﺮﺏ‬
‫وﺯﻥ ﻫﺮ ﻣﻌﯿﺎﺭ ﺩﺭ ﺍﻣﺘﯿﺎﺯ ﮐﺴﺐ ﺷﺪﻩ ﺗﻮﺳﻂ ﻓﺮوﺷﻨﺪﻩ ﺩﺭ‬
‫ﺭﺍﺑﻄﻪ ﺑﺎ ﺁﻥ ﻣﻌﯿﺎﺭ ﺑﺪﺳﺖ ﻣﯽﺁﯾﺪ‪ ،‬ﺑﺮﺍی ﺍوﻟﻮﯾﺖﺑﻨﺪی و‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪۀ ﻣﻨﺎﺳﺐ ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﺷﻮﺩ]‪.[10،9،8‬‬
‫ﺭوﺷﻬﺎی ﺗﺼﻤﯿﻢﮔﯿﺮی ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ ﻧﯿﺰ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﻟﮥ‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﻣﻮﺭﺩ ﺍﺳﺘﻔﺎﺩﻩ ﻗﺮﺍﺭ ﮔﺮﻓﺘﻪﺍﻧﺪ ﮐﻪ ﺩﺭ ﺍﯾﻦ‬
‫ﺯﻣﯿﻨﻪ ﻣﯽﺗﻮﺍﻥ ﺑﻪ ﮐﺎﺭﺑﺮﺩ ﺗﺌﻮﺭی ﻣﻄﻠﻮﺑﯿﺖ ]‪[11‬و ﺭوﺵ‬
‫ﻓﺮﺁﯾﻨﺪ ﺗﺤﻠﯿﻞ ﺳﻠﺴﻠﻪ ﻣﺮﺍﺗﺒﯽ‪ 3‬ﺍﺷﺎﺭﻩ ﻧﻤﻮﺩ]‪.[13،12‬‬
‫ﺭوﺷﻬﺎی ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﺭﯾﺎﺿﯽ ﺍﺯ ﺩﯾﺮﺑﺎﺯ ﺑﺮﺍی ﺍﻧﺘﺨﺎﺏ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﻣﻮﺭﺩ ﺗﻮﺟﻪ ﻣﺤﻘﻘﯿﻦ ﺑﻮﺩﻩﺍﻧﺪ‪ .‬ﯾﮑﯽ ﺍﺯ ﺍوﻟﯿﻦ‬
‫ﺗﺤﻘﯿﻘﺎﺕ ﺩﺭ ﺍﯾﻦ ﺯﻣﯿﻨﻪ ﺩﺭ ﺳﺎﻝ ‪ 1974‬ﺍﺭﺍﺋﻪ ﺷﺪﻩ ﮐﻪ ﺩﺭ ﺁﻥ‬
‫ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﻋﺪﺩ ﺻﺤﯿﺢ ﺗﻠﻔﯿﻘﯽ‪ 4‬ﺑﺮﺍی ﺑﻬﺮﻩﺑﺮﺩﺍﺭی ﺍﺯ‬
‫ﺣﺪﺍﮐﺜﺮ ﺗﺨﻔﯿﻒ ﭘﯿﺸﻨﻬﺎﺩی ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﺷﺪﻩ‬
‫ﺍﺳﺖ ]‪ .[14‬ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﺧﻄﯽ ﻧﯿﺰ ﺍﺯ ﺟﻤﻠﻪ ﺭوﺷﻬﺎﺋﯽ ﺑﻮﺩﻩ‬
‫ﺍﺳﺖ ﮐﻪ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺁﻥ ﺑﺮﺍی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﮔﺰﺍﺭﺵ‬
‫ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﺭوﺷﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﺧﻄﯽ ﺑﺎ ﮐﻤﯿﻨﻪ‬
‫ﮐﺮﺩﻥ ﻫﺰﯾﻨﮥ ﺧﺮﯾﺪ و ﻣﻮﺟﻮﺩی ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﻣﺤﺪوﺩﯾﺘﻬﺎﺋﯽ ﺍﺯ‬
‫ﻗﺒﯿﻞ ﻇﺮﻓﯿﺖ ﻓﺮوﺷﻨﺪﮔﺎﻥ‪ ،‬ﻣﯿﺰﺍﻥ ﺗﻘﺎﺿﺎی ﺧﺮﯾﺪﺍﺭ‪ ،‬ﮐﯿﻔﯿﺖ و‬
‫ﺳﺮوﯾﺲﺩﻫﯽ ﻓﺮوﺷﻨﺪﻩ ﻣﺒﺎﺩﺭﺕ ﺑﻪ ﺷﻨﺎﺳﺎﺋﯽ ﻓﺮوﺷﻨﺪﮔﺎﻥ‬
‫ﻣﻨﺎﺳﺐ و ﺗﺨﺼﯿﺺ ﺳﻔﺎﺭﺵ ﺑﻪ ﻫﺮ ﮐﺪﺍﻡ ﺍﺯ ﺁﻧﻬﺎ‬
‫ﻣﯽﮐﻨﻨﺪ]‪ .[16،15‬ﻫﻤﭽﻨﯿﻦ ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﺁﺭﻣﺎﻧﯽ ﺍﺯ ﻧﻮﻉ ﺧﻄﯽ‬
‫و ﻏﯿﺮ ﺧﻄﯽ ﻧﯿﺰ ﺩﺭ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬
‫ﺩﺭ ﺗﺤﻘﯿﻘﺎﺗﯽ ﮐﻪ ﺍﯾﻦ ﺭوﺵ ﺭﺍ ﻣﻮﺭﺩ ﺍﺳﺘﻔﺎﺩﻩ ﻗﺮﺍﺭ ﺩﺍﺩﻩﺍﻧﺪ‪،‬‬
‫ﺍﻫﺪﺍﻑ ﻗﯿﻤﺖ‪ ،‬ﺳﻄﺢ ﺳﺮوﯾﺲ‪ ،‬ﺗﺤﻮﯾﻞ و ﮐﯿﻔﯿﺖ ﺑﻌﻨﻮﺍﻥ ﺗﻮﺍﺑﻊ‬
‫ﻫﺪﻑ ﻣﺪ ﻧﻈﺮ ﻣﺤﻘﻘﯿﻦ ﺑﻮﺩﻩ ﺍﺳﺖ]‪ .[18،17‬ﻧﯿﺰ ﺑﺮﺍی‬
‫ﺍﻧﺪﺍﺯﻩﮔﯿﺮی ﻋﻤﻠﮑﺮﺩ ﻓﺮوﺷﻨﺪﮔﺎﻥ‪ ،‬ﺭوﺵ ﺁﻧﺎﻟﯿﺰ ﭘﻮﺷﺸﯽ‬
‫ﺩﺍﺩﻩﻫﺎ‪ 5‬ﭘﯿﺸﻨﻬﺎﺩ و ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﺷﺪﻩ ﺍﺳﺖ]‪.[20،19‬‬
‫ﺍﺯ ﺭوﺷﻬﺎی ﺗﻠﻔﯿﻘﯽ ﻫﻢ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺍﺳﺘﻔﺎﺩﻩ ﺷﺪﻩ ﮐﻪ ﺍﺯ ﺁﻥ ﺟﻤﻠﻪ ﻣﯽﺗﻮﺍﻥ ﺑﻪ ﻣﺪﻟﯽ‬
‫ﺍﺷﺎﺭﻩ ﮐﺮﺩ ﮐﻪ ﺑﺎ ﺗﻠﻔﯿﻖ ﻧﺘﺎﯾﺞ ﺣﺎﺻﻞ ﺍﺯ ﺭوﺵ ‪ AHP‬ﺩﺭ ﯾﮏ‬
‫ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﺧﻄﯽ‪ ،‬ﺑﻪ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ و ﺗﺨﺼﯿﺺ‬
‫ﺳﻔﺎﺭﺵ ﻣﯽﭘﺮﺩﺍﺯﺩ]‪.[21‬‬
‫ﺩﺭ ﻣﺮوﺭ ﻣﻄﺎﻟﻌﺎﺗﯽ‪ ،‬ﺑﻪ ﺗﺤﻘﯿﻘﺎﺗﯽ ﺑﺮﻣﯽﺧﻮﺭﯾﻢ ﮐﻪ ﻣﻮﺿﻮﻉ‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺭﺍ ﺩﺭ ﻣﺤﯿﻄﻬﺎی ﻏﯿﺮ ﺩﻗﯿﻖ ﺑﺮﺭﺳﯽ‬
‫ﻧﻤﻮﺩﻩﺍﻧﺪ‪ .‬ﺩﺭ ﯾﮑﯽ ﺍﺯ ﺍﯾﻦ ﺗﺤﻘﯿﻘﺎﺕ ﺑﺎ ﺑﮑﺎﺭﮔﯿﺮی ﻗﺎﺑﻠﯿﺖﻫﺎی‬
‫ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی ﺩﺭ ﺭوﺵ ﻓﺮﺁﯾﻨﺪ ﺗﺤﻠﯿﻞ ﺳﻠﺴﻠﻪ‬
‫‪570‬‬
‫ﻣﺒﻨﺎی ‪ CBR‬ﺗﻮﺳﻌﻪ ﺩﺍﺩﻩ ﻣﯽ ﺷﻮﺩ ﮐﻪ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ‬
‫ﺭﺍ ﺩﺭ ﻣﺤﯿﻄﻬﺎی ﻏﯿﺮ ﺩﻗﯿﻖ و ﻓﺎﺯی ﻣﺪ ﻧﻈﺮ ﻗﺮﺍﺭ ﻣﯽﺩﻫﺪ‪.‬‬
‫ﺩﺭ ﺍﺩﺍﻣﻪ و ﺩﺭ ﺑﺨﺶ ﻣﺮوﺭ ﻣﻄﺎﻟﻌﺎﺗﯽ ﭘﯿﺮﺍﻣﻮﻥ ﻣﻮﺿﻮﻉ‬
‫ﺗﺤﻘﯿﻖ ﻣﯽﭘﺮﺩﺍﺯﯾﻢ‪ .‬ﺑﺨﺶ ﻣﻌﺮﻓﯽ ﺍﺟﻤﺎﻟﯽ ﺑﻪ ﺭوﺵ ‪CBR‬‬
‫ﻣﯽﭘﺮﺩﺍﺯﺩ‪ .‬ﺩﺭ ﺑﺨﺶ ﺑﻤﻨﻈﻮﺭ ﮐﺴﺐ ﺁﻣﺎﺩﮔﯽ ﻻﺯﻡ ﺑﺮﺍی‬
‫ﺑﻬﺮﻩﺑﺮﺩﺍﺭی ﺍﺯ ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی ﺩﺭ ﻣﺪﻝ ﺍﺭﺍﺋﻪ ﺷﺪﻩ‪،‬‬
‫ﻣﻔﺎﻫﯿﻢ ﻣﺮﺗﺒﻂ ﺑﻪ ﺍﺟﻤﺎﻝ ﺗﻮﺿﯿﺢ ﺩﺍﺩﻩ ﻣﯽﺷﻮﺩ‪ .‬ﺩﺭ ﺑﺨﺶ‬
‫ﻣﺪﻝ ﻃﺮﺍﺣﯽ ﺷﺪﻩ ﺑﻪ ﺗﻔﺼﯿﻞ ﺗﺸﺮﯾﺢ ﻣﯽﮔﺮﺩﺩ و ﺩﺭ ﻗﺴﻤﺖ‬
‫ﺑﻌﺪی ﻣﺪﻝ ﻃﺮﺍﺣﯽ ﺷﺪﻩ ﺩﺭ ﯾﮏ ﻣﺜﺎﻝ ﭘﯿﺎﺩﻩﺳﺎﺯی ﻣﯽﺷﻮﺩ‪.‬‬
‫ﺑﺨﺶ ﻧﺘﯿﺠﻪﮔﯿﺮی و ﺍﺭﺍﺋﮥ ﭘﯿﺸﻨﻬﺎﺩ ﺑﺮﺍی ﺗﻮﺳﻌﮥ ﺗﺤﻘﯿﻖ‬
‫ﺣﺎﺿﺮ ﺍﺧﺘﺼﺎﺹ ﺩﺍﺭﺩ‪.‬‬
‫ﻣﺮوﺭ ﻣﻄﺎﻟﻌﺎﺗﯽ‬
‫ﻣﺮوﺭ ﻣﻄﺎﻟﻌﺎﺗﯽ ﺍﯾﻦ ﻧﻮﺷﺘﺎﺭ ﺑﺎ ﺗﺎﮐﯿﺪ ﺑﺮ ﺍﯾﻦ ﻣﻄﻠﺐ ﺁﻏﺎﺯ‬
‫ﻣﯽﺷﻮﺩ ﮐﻪ ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ )ﺗﺎﻣﯿﻦﮐﻨﻨﺪﻩ( ﻣﺎﻫﯿﺘﺎً‬
‫ﯾﮏ ﻣﺴﺎﻟﮥ ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ ﺍﺳﺖ‪ .‬ﺑﺮ ﺍﺳﺎﺱ ﯾﮏ ﺗﺤﻘﯿﻖ ﺑﻨﯿﺎﺩی‪،‬‬
‫‪ 23‬ﻣﻌﯿﺎﺭ ﻣﻬﻢ و ﭘﺮﺍﺳﺘﻔﺎﺩﻩ ﺩﺭ ﺍﻧﺘﺨﺎﺏ ﺗﺎﻣﯿﻦ ﮐﻨﻨﺪﮔﺎﻥ ﺍﺯ‬
‫ﻣﯿﺎﻥ ‪ 50‬ﻓﺎﮐﺘﻮﺭ ﻣﺠﺰﺍ ﺷﻨﺎﺳﺎﺋﯽ ﮔﺮﺩﯾﺪ‪ .‬ﺑﺮ ﺍﯾﻦ ﺍﺳﺎﺱ‬
‫ﮐﯿﻔﯿﺖ ﮐﺎﻻ‪ ،‬ﺗﺤﻮﯾﻞ ﺑﻪ ﻣﻮﻗﻊ‪ ،‬ﻋﻤﻠﮑﺮﺩ ﻣﻨﺎﺳﺐ ﻣﺤﺼﻮﻝ‪،‬‬
‫ﮔﺎﺭﺍﻧﺘﯽ ﮐﺎﻻ‪ ،‬ﻗﯿﻤﺖ ﮐﺎﻻ و ﻗﺎﺑﻠﯿﺘﻬﺎی ﻓﻨﯽ ﺁﻥ و ﺍﻣﮑﺎﻧﺎﺕ و‬
‫ﻇﺮﻓﯿﺖ ﺗﻮﻟﯿﺪ ﻓﺮوﺷﻨﺪﻩ ﺑﻌﻨﻮﺍﻥ ﻣﻬﻤﺘﺮﯾﻦ ﻋﻮﺍﻣﻞ ﺷﻨﺎﺳﺎﺋﯽ‬
‫ﮔﺮﺩﯾﺪﻧﺪ]‪.[3‬‬
‫وِﺑﺮ و ﻫﻤﮑﺎﺭﺍﻧﺶ ﺑﻪ ﺍﺳﺘﻨﺎﺩ ﺑﺮﺭﺳﯽ ‪ 74‬ﻣﻘﺎﻟﻪ ﮐﻪ ﺩﺭ‬
‫ﺭﺍﺑﻄﻪ ﺑﺎ ﺣﻞ ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ ﺗﺎﻣﯿﻦ ﮐﻨﻨﺪﻩ ﺗﻬﯿﻪ ﺷﺪﻩ ﺑﻮﺩﻧﺪ‪،‬‬
‫ﺩﺭﯾﺎﻓﺘﻨﺪ ﮐﻪ ﺑﯿﺶ ﺍﺯ ‪ 63‬ﺩﺭﺻﺪ ﻣﻘﺎﻻﺕ ﻣﻮﺿﻮﻉ ﺍﻧﺘﺨﺎﺏ‬
‫ﻓﺮوﺷﻨﺪﻩ ﺭﺍ ﺩﺭ ﻣﺤﯿﻂ ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ ﻣﺪ ﻧﻈﺮ ﻗﺮﺍﺭ ﺩﺍﺩﻩﺍﻧﺪ]‪.[4‬‬
‫ﺩﺭ ﻣﺮﺟﻊ ]‪ [5‬ﺷﺶ ﻓﺎﮐﺘﻮﺭ ﻣﻬﻢ ﺗﺤﻮﯾﻞ ﺑﻤﻮﻗﻊ‪ ،‬ﮐﯿﻔﯿﺖ‪،‬‬
‫ﻗﯿﻤﺖ‪ ،‬ﺣﺮﻓﻪﺍی ﺑﻮﺩﻥ‪ ،‬ﻣﺴﻮﻟﯿﺖ ﭘﺬﯾﺮی ﺩﺭ ﻗﺒﺎﻝ ﻧﯿﺎﺯﻫﺎی‬
‫ﻣﺸﺘﺮی و ﺭوﺍﺑﻂ ﺑﻠﻨﺪ ﻣﺪﺕ ﺑﺎ ﻓﺮوﺷﻨﺪﻩ ﻋﻮﺍﻣﻞ ﻣﻬﻢ ﺍﻧﺘﺨﺎﺏ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺩﺭ ﮐﺸﻮﺭ ﭼﯿﻦ ﺍﺭﺯﯾﺎﺑﯽ ﺷﺪﻩ ﺍﺳﺖ و ﺑﺎﻻﺧﺮﻩ ﺩﺭ‬
‫ﻣﺮﺟﻊ ]‪ [6‬ﻓﺎﮐﺘﻮﺭﻫﺎی ﻫﺰﯾﻨﻪ‪ ،‬ﮐﯿﻔﯿﺖ‪ ،‬ﺗﺤﻮﯾﻞ و ﻇﺮﻓﯿﺖ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﻣﻌﯿﺎﺭﻫﺎﺋﯽ ﻣﻌﺮﻓﯽ ﺷﺪﻩﺍﻧﺪ ﮐﻪ ﺩﺭ ﻋﻤﻞ ﺑﺮﺍی‬
‫ﺍﺭﺯﯾﺎﺑﯽ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﮐﺎﺭﺑﺮﺩ ﺑﯿﺸﺘﺮی ﺩﺍﺭﻧﺪ‪ .‬ﺑﻬﺮ ﺣﺎﻝ ﺍﻧﺘﺨﺎﺏ‬
‫ﻣﻌﯿﺎﺭﻫﺎی ﺍﺭﺯﯾﺎﺑﯽ ﻓﺮوﺷﻨﺪﮔﺎﻥ وﺍﺑﺴﺘﻪ ﺑﻪ ﻧﻮﻉ ﻣﺤﺼﻮﻝ و‬
‫ﺷﺮﺍﯾﻂ ﻣﺴﺎﻟﻪ ﺍﺳﺖ]‪.[7‬‬
‫ﺍﯾﻨﮏ ﺑﻪ ﻣﺮوﺭ ﺭوﺷﻬﺎﺋﯽ ﻣﯽﭘﺮﺩﺍﺯﯾﻢ ﮐﻪ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﻟﮥ‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺍﺭﺍﺋﻪ ﺷﺪﻩﺍﻧﺪ‪ .‬ﺭوﺷﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ وﺯﻥ ﺩﻫﯽ‬
‫ﺧﻄﯽ‪ 2‬ﺍﺯ ﻣﺮﺳﻮﻡ ﺗﺮﯾﻦ ﺍﺑﺰﺍﺭﻫﺎی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﻪ‬
‫ﺣﺴﺎﺏ ﻣﯽﺁﯾﻨﺪ‪ .‬ﺩﺭ ﺍﯾﻦ ﺭوﺵ ﺍﺑﺘﺪﺍ ﺑﻪ ﻫﺮ ﯾﮏ ﺍﺯ ﻣﻌﯿﺎﺭﻫﺎی‬
 ‫‪571‬‬ ‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ ‪.....‬‬

‫ﻧﯿﺎﺯﻫﺎی ﺧﺮﯾﺪﺍﺭ ﻣﺒﻨﺎی ﺑﺎﺯﯾﺎﺑﯽ و ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﻮﺩﻩ و‬
‫ﺑﻪ ﻣﻄﻠﻮﺑﯿﺖ ﻣﺸﺨﺼﻪﻫﺎی ﺑﺎﺯﯾﺎﺑﯽ ﺗﻮﺟﻬﯽ ﻧﺸﺪﻩ ﺍﺳﺖ‪ .‬ﺩﺭ‬
‫ﺗﺤﻘﯿﻖ ﺣﺎﺿﺮ ﺑﺎ ﺑﮑﺎﺭﮔﯿﺮی ﺭوﺵ ‪ ،TOPSIS8‬ﮐﺎﺭﺍﺋﯽ و ﺩﻗﺖ‬
‫ﻣﺮﺣﻠﮥ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺗﻘﻮﯾﺖ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬
‫]‪ .[24،23‬ﺍﺯ ﺭوﺷﻬﺎی ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﺭﯾﺎﺿﯽ ﺩﺭ ﻣﺤﯿﻂﻫﺎی‬
‫ﺭوﺵ ﺍﺳﺘﺪﻻﻝ ﻣﺒﺘﻨﯽ ﺑﺮ ﻣﻮﺭﺩ )‪(CBR‬‬
‫ﺭوﺵ ﺍﺳﺘﺪﻻﻝ ﻣﺒﺘﻨﯽ ﺑﺮ ﻣﻮﺭﺩ ﺑﺮ ﺍﺳﺎﺱ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ‬
‫ﭘﺎﺳﺦ ﻣﺴﺎﺋﻞ ﻗﺒﻠﯽ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﺋﻞ ﻣﺸﺎﺑﻪ ﺟﺪﯾﺪ ﺷﮑﻞ‬
‫ﮔﺮﻓﺘﻪ ﺍﺳﺖ‪ CBR .‬ﺑﻌﻨﻮﺍﻥ ﺭوﺷﯽ ﺷﻨﺎﺧﺘﻪ ﻣﯽﺷﻮﺩ ﮐﻪ ﺍﺯ‬
‫ﻧﺤﻮۀ ﺭﻓﺘﺎﺭ ﺍﻧﺴﺎﻧﻬﺎ ﺩﺭ ﺑﺮﺧﻮﺭﺩ ﺑﺎ ﻣﺴﺎﺋﻞ ﺟﺪﯾﺪ ﺍﻟﮕﻮﺑﺮﺩﺍﺭی‬
‫ﮐﺮﺩﻩ ﺍﺳﺖ؛ ﺑﻪ ﺍﯾﻦ ﺗﺮﺗﯿﺐ ﮐﻪ ﺍﺯ ﺗﺠﺮﺑﯿﺎﺕ ﮐﺴﺐ ﺷﺪﻩ ﺩﺭ‬
‫ﺣﻞ ﻣﺴﺎﺋﻞ ﮔﺬﺷﺘﻪ ﺑﻌﻨﻮﺍﻥ ﺭﺍﻫﻨﻤﺎﺋﯽ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﺋﻞ ﺟﺪﯾﺪ‬
‫ﺑﻬﺮﻩ ﻣﯽﺑﺮﺩ‪.‬‬
‫ﺣﻞ ﻣﺴﺎﻟﻪ ﺑﺮوﺵ ‪ CBR‬ﺩﺭ ﯾﮏ ﭼﺮﺧﻪ ﺍﻧﺠﺎﻡ ﻣﯽﮔﯿﺮﺩ و‬
‫ﺩﺭ ﺑﺮﮔﯿﺮﻧﺪۀ ﭼﻬﺎﺭ ﻋﻤﻞ ﻋﻤﺪﻩ ﺑﻪ ﺷﺮﺡ ﺯﯾﺮ ﺍﺳﺖ‪:‬‬
‫‪ – 1‬ﺑﺎﺯﯾﺎﺑﯽ‪" 9‬ﻣﻮﺭﺩ"‪ 10‬ﻣﺸﺎﺑﻪ ﺑﺎ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ‬
‫‪ – 2‬ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﭘﺎﺳﺦ ﻣﺴﺎﻟﮥ ﻣﺸﺎﺑﻪ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ ﺑﺮﺍی‬
‫ﺗﻬﯿﻪ ﭘﺎﺳﺦ ﭘﯿﺸﻨﻬﺎﺩی ﺑﺮﺍی ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ‬
‫‪ – 3‬ﺑﺎﺯﺑﯿﻨﯽ ﺩﺭ ﭘﺎﺳﺦ ﭘﯿﺸﻨﻬﺎﺩی ﺩﺭ ﺻﻮﺭﺕ وﺟﻮﺩ‬
‫ﻣﻐﺎﯾﺮﺕ ﺩﺭ ﺷﺮﺍﯾﻂ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ و ﻣﺴﺎﻟﮥ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ‬
‫‪ – 4‬ﻧﮕﻬﺪﺍﺭی ﻣﻮﺭﺩ ﺟﺪﯾﺪ )ﻣﺴﺎﻟﻪ ﺟﺪﯾﺪ و ﭘﺎﺳﺦ ﺁﻥ(‬
‫ﺑﺮﺍی ﺍﺳﺘﻔﺎﺩﻩ ﺩﺭ ﺁﯾﻨﺪﻩ‬
‫ﺷﮑﻞ )‪ (1‬ﭼﺮﺧﮥ ﺭوﺵ ‪ CBR‬ﺭﺍ ﺑﻪ ﺗﺼﻮﯾﺮ ﮐﺸﯿﺪﻩ‬
‫ﺍﺳﺖ]‪.[30‬‬
‫ﻣﺴﺎﻟﮥ‬
‫ﺟﺪﯾﺪ‬
‫ﻣﺮﺍﺗﺒﯽ‪ ،‬ﻣﺘﺪ ﺟﺪﯾﺪی ﺑﺮﺍی ﺍﺭﺯﯾﺎﺑﯽ ﺗﺎﻣﯿﻦ ﮐﻨﻨﺪﮔﺎﻥ ﺍﺭﺍﺋﻪ‬
‫ﺷﺪﻩ ﺍﺳﺖ ]‪ .[22‬ﻫﻤﭽﻨﯿﻦ ﺩﺭ ﺍﯾﻦ ﺯﻣﯿﻨﻪ ﺗﺤﻘﯿﻘﺎﺗﯽ ﺍﻧﺠﺎﻡ‬
‫ﺷﺪﻩ ﺍﺳﺖ ﮐﻪ ﮐﺎﺭﺑﺮﺩ ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی ﺭﺍ ﺩﺭ‬
‫ﻣﺤﺎﺳﺒﮥ ﺍﻣﺘﯿﺎﺯ ﻧﻬﺎﺋﯽ ﺗﺎﻣﯿﻦﮐﻨﻨﺪﮔﺎﻥ ﻣﻨﻌﮑﺲ ﻣﯽﮐﻨﺪ‬
‫ﻓﺎﺯی ﻧﯿﺰ ﺑﺮﺍی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺍﺳﺘﻔﺎﺩﻩ ﺷﺪﻩ ﺍﺳﺖ ﮐﻪ ﺑﻪ‬
‫ﻋﻨﻮﺍﻥ ﻧﻤﻮﻧﻪ ﻣﯽﺗﻮﺍﻥ ﺑﻪ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭوﺵ ﺑﺮﻧﺎﻣﻪﺭﯾﺰی ﺁﺭﻣﺎﻧﯽ‬
‫ﻓﺎﺯی ﺍﺷﺎﺭﻩ ﻧﻤﻮﺩ]‪.[25‬‬
‫ﺍﺧﯿﺮﺍً ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭوﺵ ‪ CBR‬ﻧﯿﺰ ﺩﺭ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﮔﺎﻥ‬
‫ﮔﺰﺍﺭﺵ ﮔﺮﺩﯾﺪﻩ ﺍﺳﺖ‪ .‬ﺩﺭ ﯾﮑﯽ ﺍﺯ ﻧﺨﺴﺘﯿﻦ ﺗﺤﻘﯿﻘﺎﺗﯽ ﮐﻪ ﺑﻪ‬
‫ﺍﯾﻦ ﻣﻘﻮﻟﻪ ﭘﺮﺩﺍﺧﺘﻪ ﺍﺳﺖ‪ ،‬ﺭوﺵ ‪ CBR‬ﺑﻌﻨﻮﺍﻥ ﺳﯿﺴﺘﻢ‬
‫ﭘﺸﺘﯿﺒﺎﻧﯽ ﺍﺯ ﺗﺼﻤﯿﻤﺎﺕ ﻣﺮﺑﻮﻁ ﺑﻪ ﺧﺮﯾﺪ ﭘﯿﺸﻨﻬﺎﺩ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬
‫ﻣﺤﻘﻖ ﺗﺎﮐﯿﺪ ﻣﯽﮐﻨﺪ ﮐﻪ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﭼﻨﯿﻦ ﺍﺑﺰﺍﺭی ﺑﻪ‬
‫ﺗﺼﻤﯿﻤﺎﺕ ﺳﺮﯾﻌﺘﺮ‪ ،‬ﺩﻗﯿﻘﺘﺮ‪ ،‬ﺍﺭﺯﺍﻧﺘﺮ و ﺑﺎ ﮐﯿﻔﯿﻔﺖﺗﺮ‬
‫ﻣﯽﺍﻧﺠﺎﻣﺪ]‪ .[26‬ﻫﻤﭽﻨﯿﻦ ﺑﻤﻨﻈﻮﺭ ﻣﺪﯾﺮﯾﺖ ﮐﺎﺭﺳﭙﺎﺭی‪ 6‬ﺑﻪ‬
‫ﺗﺎﻣﯿﻦﮐﻨﻨﺪﮔﺎﻥ و ﺍﺗﻮﻣﺎﺳﯿﻮﻥ ﺗﺼﻤﯿﻤﺎﺕ ﻣﺮﺗﺒﻂ‪ ،‬ﻣﺪﻟﯽ ﺑﺮ‬
‫ﻣﺒﻨﺎی ﺭوﺵ ‪ CBR‬ﺍﺭﺍﺋﻪ ﮔﺮﺩﯾﺪﻩ ﮐﻪ ﺍﺑﺰﺍﺭ ﻣﻨﺎﺳﺒﯽ ﺭﺍ ﺑﺮﺍی‬
‫ﻣﺪﯾﺮﯾﺖ ﺗﺎﻣﯿﻦ ﮐﻨﻨﺪﮔﺎﻥ ﻓﺮﺍﻫﻢ ﻧﻤﻮﺩﻩ ﺍﺳﺖ]‪.[27‬‬
‫ﺍﺯ ﺭوﺵ ‪ CBR‬ﺩﺭ ﺗﺼﻤﯿﻤﺎﺕ ﻣﺮﺗﺒﻂ ﺑﺎ ﺷﻨﺎﺳﺎﺋﯽ ﻋﻨﺎﺻﺮ‬
‫ﺗﺸﮑﯿﻞ ﺩﻫﻨﺪۀ ﺯﻧﺠﯿﺮۀ ﺗﺎﻣﯿﻦ و ﺍﺭﺯﯾﺎﺑﯽ ﺗﺎﻣﯿﻦﮐﻨﻨﺪﮔﺎﻥ‬
‫ﺑﺎﻟﻘﻮﻩ ﺑﻤﻨﻈﻮﺭ ﮐﺎﺭﺳﭙﺎﺭی ﻓﻌﺎﻟﯿﺘﻬﺎی ﺧﺎﺹ ﻧﯿﺰ ﺍﺳﺘﻔﺎﺩﻩ‬
‫ﮔﺮﺩﯾﺪﻩ ﺍﺳﺖ ]‪[28‬و ﻧﻬﺎﯾﺘﺎً ﺩﺭ ﯾﮏ ﻣﻘﻮﻟﮥ ﻣﺸﺎﺑﻪ ﺑﺎ ﻣﻮﺿﻮﻉ‬
‫ﺍﯾﻦ ﺗﺤﻘﯿﻖ‪ ،‬ﺭوﺵ ‪ CBR‬ﺑﻌﻨﻮﺍﻥ ﺳﯿﺴﺘﻢ ﭘﺸﺘﯿﺒﺎﻧﯽ ﺍﺯ‬
‫ﺗﺼﻤﯿﻤﺎﺕ ﺳﺎﺧﺖ ﯾﺎ ﺧﺮﯾﺪ‪ 7‬ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﺷﺪﻩ ﺍﺳﺖ]‪.[29‬‬
‫ﻣﺮوﺭ ﻣﻄﺎﻟﻌﺎﺗﯽ ﻓﻮﻕ ﻧﺸﺎﻥ ﻣﯽﺩﻫﺪ ﮐﻪ ﺭوﺵ ‪CBR‬‬

‫ﺑﺎﺯﯾﺎﺑﯽ ﻣﻮﺍﺭﺩ‬
‫ﺑﻌﻨﻮﺍﻥ ﯾﮑﯽ ﺍﺯ ﺍﺑﺰﺍﺭﻫﺎی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﻣﻮﺭﺩ ﺗﻮﺟﻪ‬
‫ﻣﺸﺎﺑﻪ‬ ‫ﻣﺤﻘﻘﯿﻦ ﺑﻮﺩﻩ ﺍﺳﺖ‪ .‬ﺗﺤﻘﯿﻘﺎﺕ ﻣﻨﺘﺸﺮ ﺷﺪﻩ ﺩﺭ ﺍﯾﻦ ﺯﻣﯿﻨﻪ‪،‬‬
‫ﻣﻮﺿﻮﻉ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ و ﺳﺎﯾﺮ ﻣﻘﻮﻻﺕ ﻣﺸﺎﺑﻪ ﻓﻮﻕ ﺭﺍ ﺩﺭ‬
‫ﺍﻧﺒﺎﺭۀ‬
‫ﺍﺳﺘﻔﺎﺩﻩ‬
‫ﻣﺤﯿﻂﻫﺎی ﻗﻄﻌﯽ و ﻣﻌﯿﻦ ﻣﻮﺭﺩ ﺑﺮﺭﺳﯽ ﻗﺮﺍﺭ ﺩﺍﺩﻩﺍﻧﺪ ﮐﻪ ﺩﺭ‬
‫ﻣﻮﺭﺩ‬
‫ﻣﺠﺪﺩ‬ ‫ﺁﻥ ﻫﻤﮥ ﺍﻃﻼﻋﺎﺕ ﻣﻮﺭﺩ ﻧﯿﺎﺯ ﺑﺮﺍی ﺗﺼﻤﯿﻢﮔﯿﺮی ﺑﺼﻮﺭﺕ‬
‫ﻧ‬ ‫)‪(Case-Base‬‬
‫ﮕﻬﺪ‬
‫ﺍ ﺭی‬
‫ﺩﻗﯿﻖ و ﻣﺸﺨﺺ وﺟﻮﺩ ﺩﺍﺭﻧﺪ و ﻫﯿﭻ ﮐﺪﺍﻡ ﺍﺯ ﺁﻧﻬﺎ ﻣﺴﺎﻟﻪ ﺭﺍ‬
‫ﺩﺭ ﺷﺮﺍﯾﻄﯽ ﮐﻪ ﺍﻃﻼﻋﺎﺕ و ﭘﺎﺭﺍﻣﺘﺮﻫﺎی ﻏﯿﺮ ﺩﻗﯿﻖ ﺩﺭ ﻓﺮﺁﯾﻨﺪ‬
‫ﺍﺻﻼﺡ‬
‫ﭘﺎﺳﺦ‬ ‫ﺭﺍﻩ ﺣﻞ‬ ‫ﺗﺼﻤﯿﻢﮔﯿﺮی وﺟﻮﺩ ﺩﺍﺷﺘﻪ ﺑﺎﺷﺪ‪ ،‬ﺑﺮﺭﺳﯽ ﻧﮑﺮﺩﻩﺍﻧﺪ‪ .‬ﺩﺭ‬
‫ﺗﺎﺋﯿﺪ ﺷﺪﻩ‬ ‫ﭘﯿﺸﻨﻬﺎﺩی‬
‫ﺗﺤﻘﯿﻖ ﺣﺎﺿﺮ ﺑﻤﻨﻈﻮﺭ ﭘﻮﺷﺶ ﺩﺍﺩﻥ ﺍﯾﻦ ﺧﻼء ﻣﻄﺎﻟﻌﺎﺗﯽ‪ ،‬ﺍﺯ‬
‫ﺷﮑﻞ ‪ :1‬ﭼﺮﺧﮥ ﺭوﺵ ﺍﺳﺘﺪﻻﻝ ﺑﺮ ﻣﺒﻨﺎی ﻣﻮﺭﺩ )‪.(CBR‬‬ ‫ﻗﺎﺑﻠﯿﺖﻫﺎی ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی ﺩﺭ ﺭوﺵ ‪CBR‬‬
‫ﺍﺳﺘﻔﺎﺩﻩ ﺷﺪﻩ و ﺩﺭ وﺍﻗﻊ ‪ CBR‬ﻓﺎﺯی ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ‬
‫ﻫﺮ "ﻣﻮﺭﺩ" )‪ (Case‬ﺍﺯ ﺩو ﻗﺴﻤﺖ ﺗﺸﮑﯿﻞ ﻣﯽﮔﺮﺩﺩ‪.‬‬ ‫ﻓﺮوﺷﻨﺪﻩ ﺩﺭ ﻣﺤﯿﻄﻬﺎی ﻏﯿﺮ ﺩﻗﯿﻖ ﭘﯿﺸﻨﻬﺎﺩ و ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ‬
‫ﻗﺴﻤﺖ ﺍوﻝ ﻫﺮ "ﻣﻮﺭﺩ" ﺑﻪ ﺑﯿﺎﻥ ﻣﺸﺨﺼﺎﺕ ﻣﺴﺎﻟﮥ ﻣﺮﺗﺒﻂ ﺑﺎ‬ ‫ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﻋﻼوﻩ ﺑﺮ ﺁﻥ‪ ،‬ﺩﺭ ﺗﺤﻘﯿﻘﺎﺕ ﻣﺮﺗﺒﻂ ﺑﺎ ﻣﻮﺿﻮﻉ ﺍﯾﻦ‬
‫ﺁﻥ ﻣﯽﭘﺮﺩﺍﺯﺩ و ﻗﺴﻤﺖ ﺩوﻡ ﻧﯿﺰ‪ ،‬ﭘﺎﺳﺦ ﻣﺴﺎﻟﻪ ﻣﻄﺮﺡ ﺷﺪﻩ ﺭﺍ‬ ‫ﺗﺤﻘﯿﻖ‪ ،‬ﻣﻨﺤﺼﺮﺍً ﻋﺎﻣﻞ ﻣﺸﺎﺑﻬﺖٍ ﺷﺮﺍﯾﻂ ﻓﺮوﺷﻨﺪﮔﺎﻥ و‬
 ‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪1385‬‬ ‫‪572‬‬

‫ﺗﺮﺗﯿﺐ ﺩﺭ ﻣﺴﺎﻟﮥ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ و ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ‬ ‫ﺩﺭﺑﺮﺩﺍﺭﺩ‪ .‬ﻣﺸﺨﺼﺎﺕ ﯾﮏ "ﻣﻮﺭﺩ" ﺑﺎ وﯾﮋﮔﯽﻫﺎی‪ 11‬ﺗﺸﮑﯿﻞ‬
‫‪ : sim -‬ﺗﺎﺑﻊ ﻣﺤﺎﺳﺒﮥ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﺑﯿﻦ ﺩو ﻣﻘﺪﺍﺭ‬ ‫ﺩﻫﻨﺪۀ ﺁﻥ ﺗﻌﯿﯿﻦ ﻣﯽﮔﺮﺩﺩ و ﻣﻘﺎﺩﯾﺮ ﺍﺧﺘﺼﺎﺹ ﯾﺎﻓﺘﻪ ﺑﻪ ﺍﯾﻦ‬
‫‪ f Ri‬و ‪f Ii‬‬ ‫وﯾﮋﮔﯽﻫﺎ‪ ،‬وﺿﻌﯿﺖ ﺁﻥ "ﻣﻮﺭﺩ" ﺭﺍ ﻧﺸﺎﻥ ﻣﯽﺩﻫﻨﺪ‪.‬‬
‫ﺗﺎﺑﻊ ‪ sim‬ﺑﺮﺍی ﻣﻘﺎﺩﯾﺮ ﻋﺪﺩی ﻋﻤﻮﻣﺎً ﺑﺼﻮﺭﺕ ﺭﺍﺑﻄﮥ ﺯﯾﺮ‬ ‫ﺯﻣﺎﻧﯽ ﮐﻪ ﯾﮏ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ ﻣﻄﺮﺡ ﻣﯽﺷﻮﺩ‪ ،‬ﺷﺮﺍﯾﻂ ﺁﻥ ﺑﺎ‬
‫ﺗﻌﺮﯾﻒ ﻣﯽﺷﻮﺩ‪:‬‬ ‫وﺿﻌﯿﺖ ﻣﺴﺎﺋﻞ ﺣﻞ ﺷﺪۀ ﻗﺒﻠﯽ ﻣﻘﺎﯾﺴﻪ ﺷﺪﻩ و ﺑﺎ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ‬
‫| ‪| f Ii − f Ri‬‬ ‫ﻣﮑﺎﻧﯿﺰﻡﻫﺎی ﺗﻄﺒﯿﻖ‪ ،12‬ﺷﺒﯿﻪﺗﺮﯾﻦ ﻣﻮﺭﺩﻫﺎی ﻗﺒﻠﯽ ﺑﺎﺯﯾﺎﺑﯽ‬
‫‪sim (f Ii , f Ri ) = 1 −‬‬ ‫] ‪, f Ii , f Ri ∈ [α i , β i‬‬
‫‪βi − αi‬‬ ‫)‪(2‬‬ ‫ﻣﯽﺷﻮﻧﺪ‪ .‬ﺳﭙﺲ ﺍﺯ ﻣﻮﺭﺩﻫﺎی ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ ﺑﺮﺍی ﺍﺭﺍﺋﻪ ﭘﺎﺳﺦ ﺑﻪ‬
‫ﮐﻪ ﺩﺭ ﺁﻥ ‪ α i‬و ‪ βi‬ﺑﻪ ﺗﺮﺗﯿﺐ ﻧﺸﺎﻥ ﺩﻫﻨﺪۀ ﺣﺪ ﭘﺎﺋﯿﻦ‬ ‫ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ ﺍﺳﺘﻔﺎﺩﻩ ﺷﺪﻩ و ﭘﺎﺳﺦ ﭘﯿﺸﻨﻬﺎﺩی ﺗﻬﯿﻪ‬
‫و ﺑﺎﻻی ﻣﻘﺎﺩﯾﺮ ﻣﻌﯿﺎﺭ ‪ i‬ﻫﺴﺘﻨﺪ‪ .‬ﺩﺭ ﺑﺮﺧﯽ ﻣﻮﺍﺭﺩ ﻧﯿﺰ ﺍﺯ‬ ‫ﻣﯽﮔﺮﺩﺩ‪ .‬ﺩﺭ ﺻﻮﺭﺕ ﻧﯿﺎﺯ‪ ،‬ﭘﺎﺳﺦ ﭘﯿﺸﻨﻬﺎﺩی ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ‬
‫ﻣﻌﮑﻮﺱ "ﻓﺎﺻﻠﮥ ﺍﻗﻠﯿﺪوﺳﯽ ﻣﻮﺯوﻥ"‪ 15‬ﺑﻪ ﺻﻮﺭﺕ ﺯﯾﺮ )ﺑﺎ‬ ‫ﻣﻮﻗﻌﯿﺖ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ ﻣﻮﺭﺩ ﺑﺎﺯﺑﯿﻨﯽ ﻗﺮﺍﺭ ﻣﯽﮔﯿﺮﺩ و ﺩﺭ‬
‫ﻓﺮﺽ ‪ (∑Wi=1‬ﺑﺮﺍی ﻣﺤﺎﺳﺒﮥ ‪ SIR‬ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﺷﻮﺩ‪:‬‬ ‫ﻧﻬﺎﯾﺖ‪ ،‬ﻣﻮﺭﺩ ﺟﺪﯾﺪ )ﯾﻌﻨﯽ ﻣﺴﺎﻟﮥ ﻣﻄﺮوﺣﻪ و ﭘﺎﺳﺦ ﺁﻥ( ﺑﺮﺍی‬
‫‪‬‬ ‫‪n‬‬
‫‪−1‬‬
‫‪‬‬ ‫ﺍﺳﺘﻔﺎﺩﻩﻫﺎی ﺁﺗﯽ ﺩﺭ "ﻣﺨﺰﻥ ﻣﻮﺭﺩ"‪ 13‬ﻧﮕﻬﺪﺍﺭی ﻣﯽﮔﺮﺩﺩ‪.‬‬
‫‪‬‬
‫‪S IR = ‬‬ ‫∑‬ ‫‪w i × (f Ii − f Ri )  ; if f Ii ≠ f Ri for some‬‬
‫‪2‬‬
‫‪‬‬
‫‪i =1‬‬ ‫‪‬‬ ‫ﺩﺍﻣﻨﮥ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭوﺵ ‪ CBR‬ﮔﺴﺘﺮﺩﻩ ﺍﺳﺖ ﮐﻪ ﺍﺯ ﺁﻥ‬
‫‪‬‬
‫‪1‬‬ ‫‪; if f Ii = f Ri for all i‬‬ ‫ﺟﻤﻠﻪ ﻣﯽﺗﻮﺍﻥ ﺑﻪ ﮐﺎﺭﺑﺮﺩ ﺁﻥ ﺩﺭ ﺷﻨﺎﺳﺎﺋﯽ ﻋﯿﺐ ﺳﯿﺴﺘﻤﻬﺎ‬
‫)‪(3‬‬
‫]‪ [31‬ﺗﺸﺨﯿﺺ ﺑﯿﻤﺎﺭیﻫﺎ]‪ ،[32‬ﺯﻣﺎﻧﺒﻨﺪی ﺗﻮﻟﯿﺪ]‪،[33‬‬
‫ﻫﻤﺎﻧﮕﻮﻧﻪ ﮐﻪ ﺭوﺍﺑﻂ )‪ (2‬و )‪ (3‬ﻧﯿﺰ ﻧﺸﺎﻥ ﻣﯽﺩﻫﻨﺪ ﻣﻼک‬
‫ﻃﺮﺣﻬﺎی ﺑﺎﺯﺍﺭﯾﺎﺑﯽ ]‪ [34‬و ﻏﯿﺮﻩ ﺍﺷﺎﺭﻩ ﻧﻤﻮﺩ‪.‬‬
‫ﺳﻨﺠﺶ ﻣﺸﺎﺑﻬﺖ – ﮐﻪ ﺩﺭ وﺍﻗﻊ ﻣﻌﯿﺎﺭ ﺑﺎﺯﯾﺎﺑﯽ ﻣﺴﺎﻟﮥ ﺣﻞ‬
‫ﯾﮑﯽ ﺍﺯ ﭘﯿﭽﯿﺪﻩﺗﺮﯾﻦ ﺍﺑﻌﺎﺩ ﺍﺟﺮﺍﺋﯽ ‪ CBR‬ﻧﺤﻮۀ ﻣﺤﺎﺳﺒﮥ‬
‫ﺷﺪۀ ﻗﺒﻠﯽ و ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﭘﺎﺳﺦ ﺁﻥ ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ‬
‫ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ ﺑﺎ ﻣﺴﺎﺋﻞ ﺣﻞ ﺷﺪۀ ﻗﺒﻠﯽ ﺍﺳﺖ‪.‬‬
‫ﺍﺳﺖ ‪ -‬ﺻﺮﻓﺎً ﻓﺎﺻﻠﻪﺍی ﻣﯽﺑﺎﺷﺪ ﮐﻪ ﻣﻘﺎﺩﯾﺮ ﻣﻌﯿﺎﺭﻫﺎ‬
‫ﻏﺎﻟﺐ ﺭوﺷﻬﺎﺋﯽ ﮐﻪ ﺑﺮﺍی ﻣﺤﺎﺳﺒﮥ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﺑﮑﺎﺭ‬
‫)ﻣﺸﺨﺼﻪﻫﺎ( ﺩﺭ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ و ﻣﺴﺎﻟﮥ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ ﺍﺯ ﻫﻢ‬
‫ﻣﯽﺭوﻧﺪ ﺍﺯ ﺗﺎﺑﻊ ﻣﻄﺎﺑﻘﺖٍ "ﻧﺰﺩﯾﮑﺘﺮﯾﻦ ﻫﻤﺴﺎﯾﻪ"‪ 14‬ﺍﺳﺘﻔﺎﺩﻩ‬
‫ﺩﺍﺭﻧﺪ و ﻣﺜﺒﺖ )ﺳﻮﺩ( ﯾﺎ ﻣﻨﻔﯽ )ﻫﺰﯾﻨﻪ( ﺑﻮﺩﻥ ﻣﻌﯿﺎﺭﻫﺎی‬
‫ﻣﯽﮐﻨﻨﺪ‪ .‬ﺩﺭ ﺍﯾﻦ ﺭوﺷﻬﺎ ﺍﺑﺘﺪﺍ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ ﺑﺎ‬
‫ﺳﻨﺠﺶ ﻣﺸﺎﺑﻬﺖ ﺩﺭ ﺍﻧﺘﺨﺎﺏ ﻣﺴﺎﻟﮥ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ )ﺑﺮﺍی ﺣﻞ‬
‫ﻣﺴﺎﺋﻞ ﻗﺒﻠﯽ ﺩﺭ ﺭﺍﺑﻄﻪ ﺑﺎ ﺗﮏ ﺗﮏ ﻣﻌﯿﺎﺭﻫﺎی ﺳﻨﺠﺶ‬
‫ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ( ﻣﻮﺭﺩ ﺗﻮﺟﻪ ﻗﺮﺍﺭ ﻧﻤﯽﮔﯿﺮﺩ‪ .‬ﻫﻤﺎﻧﮕﻮﻧﻪ ﮐﻪ ﺩﺭ‬
‫ﻣﺸﺎﺑﻬﺖ‪ ،‬ﺍﻧﺪﺍﺯﻩﮔﯿﺮی ﻣﯽﺷﻮﺩ و ﺳﭙﺲ ﻣﯿﺎﻧﮕﯿﻦ ﻣﻮﺯوﻥ‬
‫ﺍﺩﺍﻣﮥ ﻣﻄﺎﻟﺐ ﻧﺸﺎﻥ ﺧﻮﺍﻫﯿﻢ ﺩﺍﺩ‪ ،‬ﺩﺭ ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ‬
‫ﻣﻘﺎﺩﯾﺮ ﺣﺎﺻﻠﻪ ﺑﺼﻮﺭﺕ ﺯﯾﺮ ﻣﺤﺎﺳﺒﻪ و ﻣﻼک ﻋﻤﻞ ﻗﺮﺍﺭ‬
‫ﺗﺎﻣﯿﻦﮐﻨﻨﺪﻩ ﺑﺮوﺵ ‪ ،CBR‬ﺍﺗﮑﺎء ﺻﺮﻑ ﺑﻪ ﻋﺎﻣﻞ ﻣﺸﺎﺑﻬﺖ ﺑﺮ‬
‫ﻣﯽﮔﯿﺮﺩ]‪.[35‬‬
‫ﺭوی ﮐﺎﺭﺍﺋﯽ و ﺩﻗﺖٍ ﺍﻧﺘﺨﺎﺏ ﻧﻬﺎﺋﯽ ﺗﺎﺛﯿﺮ ﻣﻨﻔﯽ ﻣﯽﮔﺬﺍﺭﺩ‪ .‬ﺑﻪ‬ ‫‪n‬‬

‫ﻫﻤﯿﻦ ﺩﻟﯿﻞ ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ ﺭوﺵ ‪ TOPSIS‬ﺭﺍ ﮐﻪ ﯾﮑﯽ ﺍﺯ‬ ‫‪∑w‬‬ ‫‪i‬‬ ‫) ‪× sim(f Ii , f Ri‬‬
‫= ‪SIR‬‬ ‫‪i =1‬‬
‫‪n‬‬
‫ﻣﺘﺪﻫﺎی ﺷﻨﺎﺧﺘﻪ ﺷﺪﻩ ﺩﺭ ﺗﺼﻤﯿﻢﮔﯿﺮی ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ‬ ‫‪∑w‬‬
‫‪i =1‬‬
‫‪i‬‬ ‫)‪(1‬‬
‫ﻣﺤﺴﻮﺏ ﻣﯽﺷﻮﺩ‪ ،‬ﺑﺮﺍی ﺍﺭﺯﯾﺎﺑﯽ ﻧﻬﺎﺋﯽ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺑﺎﺯﯾﺎﺑﯽ‬ ‫ﮐﻪ ﺩﺭ ﺁﻥ‪:‬‬
‫ﺷﺪﻩ و ﺍوﻟﻮﯾﺖﺑﻨﺪی ﺁﻧﻬﺎ )ﺑﻪ ﻣﻨﻈﻮﺭ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪۀ‬ ‫‪ : SIR -‬ﺩﺭﺟﻪ ﯾﺎ ﺷﺎﺧﺺ ﻣﺸﺎﺑﻬﺖ ﺑﯿﻦ ﺷﺮﺍﯾﻂ ﻣﺴﺎﻟﮥ‬
‫ﻣﻨﺎﺳﺐ( ﭘﯿﺸﻨﻬﺎﺩ و ﺑﮑﺎﺭ ﺧﻮﺍﻫﯿﻢ ﺑﺴﺖ‪.‬‬ ‫ﺟﺪﯾﺪ ‪ I‬و ﻣﺴﺎﻟﮥ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪۀ ‪ (0≤SIR≤1) R‬ﮐﻪ ﺩﺭ‬
‫ﺁﻥ ‪ 1‬ﻧﺸﺎﻥ ﺩﻫﻨﺪۀ ﻣﺸﺎﺑﻬﺖ ﺻﺪ ﺩﺭ ﺻﺪی ﯾﺎ ﺗﻄﺒﯿﻖ‬
‫ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی و ﮐﺎﺭﺑﺮﺩ ﺁﻥ ﺩﺭ‬ ‫ﮐﺎﻣﻞ ﺍﺳﺖ و ﻣﻘﺎﺩﯾﺮ ﮐﻤﺘﺮ ﺍﺯ ‪ ، 1‬ﺑﯿﺎﻧﮕﺮ ﺗﻄﺒﯿﻖ ﯾﺎ‬
‫‪CBR‬‬ ‫ﻣﺸﺎﺑﻬﺖ ﺟﺰﺋﯽ ﺍﺳﺖ‬
‫ﺩﺭ ﺑﺴﯿﺎﺭی ﺍﺯ ﺷﺮﺍﯾﻂ‪ ،‬ﺍﻃﻼﻋﺎﺕ ﺩﻗﯿﻖ و ﺩﺍﺩﻩﻫﺎی ﻣﻌﯿﻦ‬ ‫‪ : I -‬ﺍﯾﻨﺪﮐﺲ ﻣﺴﺎﻟﮥ ﺟﺪﯾﺪ‬
‫ﺑﺮﺍی ﻣﺪﻝ ﮐﺮﺩﻥ وﺿﻌﯿﺖ وﺍﻗﻌﯽ ﻧﺎﮐﺎﻓﯽ ﻫﺴﺘﻨﺪ‪ ،‬ﺯﯾﺮﺍ ﻗﻀﺎوﺕ‬ ‫‪ : R -‬ﺍﯾﻨﺪﮐﺲ ﻣﺴﺎﻟﮥ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ‬
‫ﺍﻓﺮﺍﺩ ﻣﻌﻤﻮﻻً ﺑﺼﻮﺭﺕ ﻣﺒﻬﻢ ﺍﻧﺠﺎﻡ ﻣﯽﮔﯿﺮﺩ و ﻣﻘﺎﺩﯾﺮ ﺩﻗﯿﻖ و‬ ‫‪ : i -‬ﺍﯾﻨﺪﮐﺲ ﻣﺸﺨﺼﻪ )ﻣﻌﯿﺎﺭ( )‪(i=1,2,…,n‬‬
‫ﻣﺸﺨﺺ ﻋﺪﺩی ﺑﺮﺍی ﺑﯿﺎﻥ ﻧﻈﺮﺍﺕ ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﻧﻤﯽﺷﻮﺩ‪ .‬ﺑﻪ‬ ‫‪ : Wi -‬وﺯﻥ ﻣﺸﺨﺼﻪ )ﻣﻌﯿﺎﺭ( ‪i‬ﺍﻡ )ﻣﻌﻤﻮﻻً ‪(∑Wi=1‬‬
‫ﻣﻨﻈﻮﺭ ﻓﺎﺋﻖ ﺁﻣﺪﻥ ﺑﺮ ﭼﻨﯿﻦ ﺗﻔﮑﺮ ﻏﯿﺮ ﺩﻗﯿﻘﯽ‪ ،‬ﭘﺮوﻓﺴﻮﺭ‬ ‫‪ : f Ii , f Ri -‬ﻣﻘﺪﺍﺭ ﺍﻣﺘﯿﺎﺯ ﯾﺎ ﺍﺭﺯﯾﺎﺑﯽ ﻣﺸﺨﺼﮥ ‪ i‬ﺑﻪ‬
 ‫‪573‬‬ ‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ ‪.....‬‬

‫)‪µn~ (x‬‬ ‫ﻟﻄﻔﯽ ﺯﺍﺩﻩ ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی ﺭﺍ ﻣﻌﺮﻓﯽ ﻧﻤﻮﺩ ﮐﻪ‬
‫‪1‬‬ ‫ﻧﻮﺁوﺭی ﺍﺻﻠﯽ ﺁﻥ ﺍﯾﺠﺎﺩ ﺍﻣﮑﺎﻥ ﻧﻤﺎﯾﺶ ﻣﻔﺎﻫﯿﻢ ﻏﯿﺮ ﺩﻗﯿﻖ و‬
‫ﻣﺒﻬﻢ ﺍﺳﺖ‪.‬‬
‫ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی ﭼﻬﺎﺭﭼﻮﺏ ﻣﻨﺎﺳﺒﯽ ﺭﺍ ﺑﺮﺍی‬
‫ﺗﻮﺻﯿﻒ و ﺑﯿﺎﻥ ﺭﻓﺘﺎﺭ ﺣﺎﻻﺕ ﻏﯿﺮ ﺻﺮﯾﺢ و ﻧﺎﺩﻗﯿﻖ ﻓﺮﺍﻫﻢ‬
‫‪0‬‬ ‫‪x‬‬
‫ﻣﯽﺁوﺭﺩ‪ .‬ﺑﻌﻨﻮﺍﻥ ﻣﺜﺎﻝ ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ ﺳﻔﺎﺭﺵ ﺗﻮﺳﻂ ﯾﮏ‬
‫~‪.‬‬
‫ﺷﮑﻞ ‪ : 3‬ﻧﻤﺎﯾﺶ ﻋﺪﺩ ﻓﺎﺯی ‪n‬‬ ‫ﻓﺮوﺷﻨﺪﻩ ﻣﯽﺗﻮﺍﻧﺪ ﺑﺎ ﻋﺒﺎﺭﺍﺗﯽ ﻧﻈﯿﺮ ﻋﺒﺎﺭﺍﺕ ﺯﯾﺮ ﺑﯿﺎﻥ ﺷﻮﺩ‪:‬‬
‫‪ -‬ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ ﺳﻔﺎﺭﺵ ﺗﻘﺮﯾﺒﺎً ‪ tm‬ﺍﺳﺖ؛ ﺍﻣﺎ ﺑﻄﻮﺭ‬
‫ﻓﺮﻣﻬﺎی ﺧﺎﺻﯽ ﺍﺯ ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﮐﻪ ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﻣﺜﻠﺜﯽ و‬
‫ﻣﺸﺨﺺ ﺍﺯ ‪ tl‬ﮐﻤﺘﺮ و ﺍﺯ ‪ tu‬ﺑﯿﺸﺘﺮ ﻧﯿﺴﺖ‪.‬‬
‫ﺫوﺯﻧﻘﻪﺍی ﻧﺎﻣﯿﺪﻩ ﻣﯽﺷﻮﻧﺪ ﺍﺑﺰﺍﺭﻫﺎی ﻣﺮﺳﻮﻣﯽ ﺑﺮﺍی ﺑﯿﺎﻥ‬
‫‪ -‬ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ ﺳﻔﺎﺭﺵ ﺑﺎ ﺩﺭﺟﮥ ﻋﻀﻮﯾﺖ ﺑﺎﻻﺋﯽ ﺩﺭ‬
‫ﺣﺎﻻﺕ ﻏﯿﺮ ﺻﺮﯾﺢ و ﻣﺒﻬﻢ ﻫﺴﺘﻨﺪ‪ .‬ﺩﺭ ﺷﮑﻞ )‪ ،(2‬ﻧﻤﺎﯾﮥ‬
‫ﻣﺤﺪوﺩۀ ] ‪ [t ′l , t ′u‬ﻗﺮﺍﺭﺩﺍﺭﺩ‪ ،‬ﻫﻤﭽﻨﯿﻦ ﺍﺯ ‪ tl‬ﮐﻤﺘﺮ و ﺍﺯ ‪tu‬‬
‫)ﺍﻟﻒ( ﻧﺸﺎﻥ ﺩﻫﻨﺪۀ ﯾﮏ ﻋﺪﺩ ﻓﺎﺯی ﻣﺜﻠﺜﯽ ﻧﺮﻣﺎﻝ ﺍﺳﺖ ﮐﻪ‬
‫ﺑﯿﺸﺘﺮ ﻧﯿﺴﺖ‪.‬‬
‫ﺑﺼﻮﺭﺕ ] ‪ [t l , t m , t u‬ﻧﺸﺎﻥ ﺩﺍﺩﻩ ﻣﯽﺷﻮﺩ و ﻧﻤﺎﯾﮥ )ﺏ( ﻧﯿﺰ ﺩﺭ‬
‫ﭼﻨﯿﻦ ﻋﺒﺎﺭﺍﺗﯽ ﻣﯽﺗﻮﺍﻧﻨﺪ ﺑﮑﻤﮏ ﻣﺠﻤﻮﻋﻪﻫﺎی ﻓﺎﺯی و‬
‫وﺍﻗﻊ ﯾﮏ ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﻧﺮﻣﺎﻝ ﺍﺳﺖ ﮐﻪ ﺍﺯ ﻧﻤﺎﺩ‬
‫ﺑﺼﻮﺭﺗﯽﮐﻪ ﺩﺭ ﺷﮑﻞ )‪ (2‬ﺁﻣﺪﻩ ﺍﺳﺖ‪ ،‬ﺗﻮﺻﯿﻒ ﺷﻮﻧﺪ‪.‬‬
‫] ‪ [t l , t′l,t′u , t u‬ﺑﺮﺍی ﻧﻤﺎﯾﺶ ﺁﻥ ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﺷﻮﺩ‪.‬‬ ‫‪µT‬‬ ‫‪µT‬‬
‫ﺍﺯ ﺁﻧﺠﺎﺋﯿﮑﻪ ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ ﺍﺯ ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﺑﺮﺍی‬
‫‪1‬‬ ‫‪1‬‬
‫ﺑﯿﺎﻥ ﻋﺪﻡ ﻗﻄﻌﯿﺖ ﺍﺳﺘﻔﺎﺩﻩ ﺧﻮﺍﻫﯿﻢ ﮐﺮﺩ‪ ،‬ﻟﺬﺍ ﺩﺭ ﺍﯾﻨﺠﺎ ﺑﻪ‬
‫ﺑﯿﺎﻥ ﭼﮕﻮﻧﮕﯽ ﺍﻧﺠﺎﻡ ﻋﻤﻠﯿﺎﺕ ﺭﯾﺎﺿﯽ ﺑﺮ ﺭوی ﺍﯾﻦ ﺍﻋﺪﺍﺩ‬
‫ﻣﯽﭘﺮﺩﺍﺯﯾﻢ‪ .‬ﺍﯾﻦ ﻣﻄﺎﻟﺐ ﺍﺯ ﻣﺮﺟﻊ ]‪ [38‬ﺫﮐﺮ ﻣﯽﺷﻮﻧﺪ‪.‬‬ ‫زﻣﺎن‬ ‫زﻣﺎن‬
‫‪tl‬‬ ‫‪tm‬‬ ‫‪tu‬‬ ‫‪tl‬‬ ‫‪t′l‬‬ ‫‪t′u‬‬ ‫‪tu‬‬
‫ﺑﺮ ﺍﺳﺎﺱ ﺍﺻﻞ ﮔﺴﺘﺮﺵ‪ 16‬ﺣﺎﺻﻞ ﺟﻤﻊ ﻓﺎﺯی ⊕ و ﻧﯿﺰ‬ ‫)اﻟﻒ(‬ ‫)ب(‬
‫ﺣﺎﺻﻞ ﺗﻔﺮﯾﻖ ﻓﺎﺯی ‪ Ө‬ﻫﺮ ﺩو ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی‪ ،‬ﯾﮏ ﻋﺪﺩ‬ ‫ﺷﮑﻞ ‪ : 2‬ﻧﻤﺎﯾﺶ ﻓﺎﺯی ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ ﺳﻔﺎﺭﺵ ﻏﯿﺮ ﺩﻗﯿﻖ‪.‬‬

‫ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﺍﺳﺖ‪ .‬ﺩو ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی‬

‫ﺑﻪ ﻣﻨﻈﻮﺭ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﻗﺎﺑﻠﯿﺘﻬﺎی ﺗﺌﻮﺭی ﻣﺠﻤﻮﻋﻪﻫﺎی‬
‫ﻣﺜﺒﺖ ] ‪~ = [m , m , m , m‬‬
‫‪ m‬و ] ‪ ~n = [n 1 , n 2 , n 3 , n 4‬ﺭﺍ‬ ‫‪1‬‬ ‫‪2‬‬ ‫‪3‬‬ ‫‪4‬‬
‫ﻓﺎﺯی ﺩﺭ ﺗﺤﻘﯿﻖ ﺣﺎﺿﺮ‪ ،‬ﺩﺭ ﺍﯾﻦ ﺑﺨﺶ ﺑﻪ ﺍﺟﻤﺎﻝ ﺗﻌﺎﺭﯾﻒ ﭘﺎﯾﮥ‬
‫ﺩﺭ ﻧﻈﺮ ﺑﮕﯿﺮﯾﺪ‪ .‬ﭘﺎﺭﺍﻣﺘﺮﻫﺎی ﺣﺎﺻﻠﺠﻤﻊ ﻓﺎﺯی و ﺣﺎﺻﻞ ﺗﻔﺮﯾﻖ‬
‫ﻣﻮﺭﺩ ﻧﯿﺎﺯ ﺭﺍ ﺍﺯ ﻧﻈﺮ ﻣﯽﮔﺬﺭﺍﻧﯿﻢ]‪:[37،36‬‬
‫ﻓﺎﺯی ﺍﯾﻦ ﺩو ﻋﺪﺩ ﺑﻪ ﺻﻮﺭﺕ ﺯﯾﺮ ﻣﺤﺎﺳﺒﻪ ﻣﯽﺷﻮﺩ‪:‬‬ ‫~‬
‫ﺗﻌﺮﯾﻒ ‪ – 1‬ﻣﺠﻤﻮﻋﮥ ﻓﺎﺯی ‪ A‬ﺩﺭ ﻣﺠﻤﻮﻋﮥ ﻣﺮﺟﻊ ‪X‬‬
‫~⊕ ~‬
‫‪m‬‬ ‫] ‪n = [m 1 + n 1 , m 2 + n 2 , m 3 + n 3 , m 4 + n 4‬‬
‫ﺑﻮﺳﯿﻠﮥ ﯾﮏ ﺗﺎﺑﻊ ﻋﻀﻮﯾﺖ )‪ µ A~ (x‬ﺗﻮﺻﯿﻒ ﻣﯽﺷﻮﺩ ﮐﻪ ﺑﺮﺍی‬
‫)‪(4‬‬
‫ﻫﺮ ‪ x∈X‬ﯾﮏ ﻋﺪﺩ ﺩﺭ ﺑﺎﺯۀ ﺑﺴﺘﮥ ]‪ [0,1‬ﻗﺮﺍﺭ ﻣﯽﺩﻫﺪ ﮐﻪ‬
‫~‪~ Θ‬‬
‫‪m‬‬ ‫] ‪n = [m1 − n 4 , m 2 − n 3 , m 3 − n 2 , m 4 − n1‬‬ ‫~‬
‫ﺑﯿﺎﻧﮕﺮ ﺩﺭﺟﮥ ﻋﻀﻮﯾﺖ ‪ x‬ﺩﺭ ﻣﺠﻤﻮﻋﮥ ﻓﺎﺯی ‪ A‬ﺍﺳﺖ‪.‬‬
‫)‪(5‬‬ ‫~‬
‫ﺗﻌﺮﯾﻒ ‪ - 2‬ﻣﺠﻤﻮﻋﮥ ﻓﺎﺯی ‪ A‬ﺩﺭ ﻣﺠﻤﻮﻋﮥ ﻣﺮﺟﻊ ‪X‬‬
‫ﺍﺯ ﺣﺎﺻﻠﻀﺮﺏ ﻓﺎﺯی ⊗ ﯾﮏ ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﺩﺭ‬ ‫ﻣﺤﺪﺏ ﺍﺳﺖ ﺍﮔﺮ و ﻓﻘﻂ ﺍﮔﺮ ﺑﺮﺍی ﻫﺮ ‪ x1‬و ‪ x2‬ﺩﺭ ‪ X‬ﺩﺍﺷﺘﻪ‬
‫ﯾﮏ ﻋﺪﺩ ﺣﻘﯿﻘﯽ ﻣﺜﺒﺖ )ﻣﺜﻞ ‪ (r‬ﻧﯿﺰ ﯾﮏ ﻋﺪﺩ ﻓﺎﺯی‬ ‫ﺑﺎﺷﯿﻢ‪:‬‬
‫ﺫوﺯﻧﻘﻪﺍی ﺑﺪﺳﺖ ﻣﯽﺁﯾﺪ؛ ﺍﯾﻦ ﺩﺭ ﺣﺎﻟﯽ ﺍﺳﺖ ﮐﻪ ﺣﺎﺻﻞﺿﺮﺏ‬ ‫)) ‪ µ A~ (λ x 1 + (1 − λ )x 2 ) ≥ Min(µ A~ (x 1 ),µ A~ (x 2‬ﮐﻪ‬
‫ﺩو ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی‪ ،‬ﺗﻘﺮﯾﺒﺎً ﯾﮏ ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی‬
‫ﺩﺭ ﺁﻥ ]‪. λ ∈ [0,1‬‬
‫ﺍﺳﺖ‪ .‬ﭘﺎﺭﺍﻣﺘﺮﻫﺎی ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﺣﺎﺻﻞ ﺍﺯ ﺿﺮﺏ ﺩو‬ ‫~‬
‫ﺗﻌﺮﯾﻒ ‪ - 3‬ﻣﺠﻤﻮﻋﮥ ﻓﺎﺯی ‪ A‬ﺩﺭ ﻣﺠﻤﻮﻋﮥ ﻣﺮﺟﻊ ‪X‬‬
‫ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی و ﻧﯿﺰ ﯾﮏ ﻋﺪﺩ ﺣﻘﯿﻘﯽ ﺩﺭ ﯾﮏ ﻋﺪﺩ‬
‫ﻧﺮﻣﺎﻝ ﻧﺎﻣﯿﺪﻩ ﻣﯽﺷﻮﺩ ﻫﺮ ﮔﺎﻩ ‪∃x i ∈ X , µ A~ (x i ) = 1 :‬‬
‫ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﺑﻪ ﺷﺮﺡ ﺯﯾﺮ ﺑﺪﺳﺖ ﻣﯽﺁﯾﻨﺪ‪:‬‬
‫ﺗﻌﺮﯾﻒ ‪ – 4‬ﻋﺪﺩﻓﺎﺯی ﯾﮏ ﺯﯾﺮ ﻣﺠﻤﻮﻋﮥ ﻓﺎﺯی ﺩﺭ‬
‫~⊗ ~‬
‫‪m‬‬ ‫] ‪n = [m1n1 , m 2 n 2 , m 3 n 3 , m 4 n 4‬‬
‫ﻣﺠﻤﻮﻋﮥ ﻣﺮﺟﻊ ‪ X‬ﺍﺳﺖ ﮐﻪ ﻫﻢ ﻣﺤﺪﺏ ﺍﺳﺖ و ﻫﻢ ﻧﺮﻣﺎﻝ‪.‬‬
‫)‪(6‬‬
‫ﺩﺭ ﺷﮑﻞ )‪ (3‬ﯾﮏ ﻋﺪﺩ ﻓﺎﺯی ﻧﺸﺎﻥ ﺩﺍﺩﻩ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬
‫]‪~ ⊗ r = [m r, m r, m r, m r‬‬
‫‪m‬‬ ‫)‪(7‬‬
‫‪1‬‬ ‫‪2‬‬ ‫‪3‬‬ ‫‪4‬‬
 ‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪1385‬‬ ‫‪574‬‬

‫ﺧﺮﯾــﺪ ﺷــﻤﺎرۀ ‪1 0 0‬‬ ‫ﻧﺎم ﻣــﻮرد‬

‫ﺧﺮﯾــﺪ ﺷــﻤﺎرۀ ‪1 0 8‬‬

‫ﻗﯿﻤــﺖ‬ ‫‪: XY‬‬ ‫‪ : XX‬ﻗﯿﻤــﺖ‬

‫ﮐﯿﻔﯿــﺖ‬ ‫‪: YY‬‬ ‫ﻣﺸﺨﺼــﺎت‬ ‫‪ : YX‬ﮐﯿﻔﯿــﺖ‬ ‫اﻃﻼﻋــﺎت ﺧﺮﯾــﺪﻫﺎی‬
‫)ﻣﻌﯿﺎرﻫــﺎ(‬ ‫ﻗﺒﻞ‬
‫‪ : X- Y- X‬ﺗﺤﻮﯾــﻞ‬ ‫‪ : X-X-X‬ﺗﺤﻮﯾــﻞ‬
‫‪: XXYX‬‬ ‫‪: XYYX‬‬
‫‪ : X-Y-Y‬ﻓﺮوﺷــﻨﺪه‬ ‫ﭘﺎﺳﺦ‬ ‫‪ : X-X-Y‬ﻓﺮوﺷــﻨﺪه‬
‫ﺧﺮﯾــﺪ ﺟﺪﯾــﺪ‬ ‫ﺧﺮﯾــﺪ ﺟﺪﯾــﺪ‬

‫‪ : XXX‬ﻗﯿﻤــﺖ‬ ‫ﻗﯿﻤــﺖ‬ ‫‪: XXX‬‬

‫‪ : YYY‬ﮐﯿﻔﯿــﺖ‬ ‫ﯾــﺎﻓﺘﻦ ﺧﺮﯾــﺪ ﻣﺸــﺎﺑﻪ‬
‫آزﻣــﻮن ﭘــﺬﯾﺮش ﻓﺮوﺷــﻨﺪۀ‬ ‫ﮐﯿﻔﯿــﺖ‬ ‫‪: YYY‬‬
‫ﻣــﺮﺑﻮط ﺑــﻪ ﺧﺮﯾــﺪ ﻣﺸــﺎﺑﻪ‬
‫‪ : Y- Y- X‬ﺗﺤﻮﯾــﻞ‬ ‫‪ : Y- Y- X‬ﺗﺤﻮﯾــﻞ‬
‫‪: XXYX‬‬ ‫‪: XXYX‬‬
‫ﻓﺮوﺷــﻨﺪه‬ ‫?‪:‬‬ ‫‪ : XYX‬ﻓﺮوﺷــﻨﺪه‬
‫ﺧﺮﯾــﺪ ﺷــﻤﺎرۀ ‪1 0 3‬‬ ‫ﺧﺮﯾــﺪ ﺷــﻤﺎرۀ ‪1 1 0‬‬

‫‪ : X‬ﻗﯿﻤــﺖ‬ ‫‪ : Y‬ﻗﯿﻤــﺖ‬ ‫ﭘﺎﺳﺦ‬

‫‪:Y‬‬
‫ﮐﯿﻔﯿــﺖ‬
‫‪. . .‬‬ ‫ﮐﯿﻔﯿــﺖ‬ ‫‪:X‬‬ ‫اﻃﻼﻋــﺎت ﺧﺮﯾــﺪﻫﺎی‬
‫ﻗ ﺒﻞ‬
‫‪ : X-Y‬ﺗﺤﻮﯾــﻞ‬ ‫ﺗﺤﻮﯾــﻞ‬ ‫‪: X-X-Y‬‬
‫‪: XYX‬‬ ‫‪: XYYX‬‬
‫‪ :‬ﻓﺮوﺷــﻨﺪه‬ ‫‪X‬‬ ‫‪ :‬ﻓﺮوﺷــﻨﺪه‬ ‫‪Y‬‬

‫ﺷﮑﻞ ‪ : 4‬ﻣﺪﻝ ﻋﻤﻮﻣﯽ ﻧﺤﻮۀ ﮐﺎﺭﺑﺮﺩ ﺭوﺵ ‪ CBR‬ﺑﺮﺍی ﺣﻞ ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ‪.‬‬

‫ﻣﻌﺮﻓﯽ ﻣﺪﻝ ﻃﺮﺍﺣﯽ ﺷﺪﻩ‬ ‫ﻫﻤﭽﻨﯿﻦ ﻓﺎﺻﻠﮥ ﺩو ﻋﺪﺩ ﻓﺎﺯی ﻓﻮﻕ ﺑﺎ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭوﺵ‬
‫ﭼﻬﺎﺭﭼﻮﺏ ﮐﻠﯽ ﻣﺪﻟﯽ ﮐﻪ ﺩﺭ ﺍﯾﻦ ﺑﺨﺶ ﺍﺭﺍﺋﻪ ﻣﯽﺷﻮﺩ ﺩﺭ‬ ‫ﺭﺃﺳﯽ‪ 17‬ﺑﺼﻮﺭﺕ ﺯﯾﺮ ﻗﺎﺑﻞ ﻣﺤﺎﺳﺒﻪ ﺍﺳﺖ‪:‬‬
‫~ ‪~,‬‬
‫‪d v (m‬‬ ‫= )‪n‬‬
‫ﺷﮑﻞ )‪ (4‬ﺑﻪ ﺗﺼﻮﯾﺮ ﮐﺸﯿﺪﻩ ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﻣﺪﻝ ﺍﺭﺍﺋﻪ ﺷﺪﻩ ﺩﺭ‬
‫‪1‬‬
‫] ‪[(m1 − n 1 ) 2 + (m 2 − n 2 ) 2 + (m 3 − n 3 ) 2 + (m 4 − n 4 ) 2‬‬
‫ﺍﯾﻦ ﺗﺤﻘﯿﻖ‪ ،‬ﻫﻤﺎﻧﮕﻮﻧﻪ ﮐﻪ ﺩﺭ ﺍﯾﻦ ﺷﮑﻞ ﻧﯿﺰ ﻣﺸﺨﺺ ﺷﺪﻩ‬ ‫‪4‬‬

‫ﺍﺳﺖ‪ ،‬ﺑﻪ ﺍﯾﻦ ﺻﻮﺭﺕ ﻋﻤﻞ ﻣﯽﮐﻨﺪ ﮐﻪ ﺑﺎ ﻣﻄﺮﺡ ﺷﺪﻥ ﯾﮏ‬ ‫)‪(8‬‬
‫~‬
‫‪n‬‬ ‫~‬
‫‪m‬‬
‫ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﺍﯾﻦ ﺭوﺵ‪ ،‬ﺩو ﻋﺪﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی و‬
‫ﺧﺮﯾﺪ ﺟﺪﯾﺪ‪ ،‬ﺷﺮﺍﯾﻂ ﺁﻥ ﺭﺍ ﺑﺎ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪﻫﺎی ﺍﻧﺠﺎﻡ‬
‫~‪~,‬‬
‫‪ d v (m‬ﺑﺎﺷﺪ‪.‬‬ ‫ﯾﮑﺴﺎﻥ ﺗﻠﻘﯽ ﻣﯽﺷﻮﻧﺪ ﻫﺮ ﮔﺎﻩ ‪n) = 0‬‬
‫ﺷﺪۀ ﻗﺒﻠﯽ ﻣﻄﺎﺑﻘﺖ ﻣﯽﺩﻫﺪ و ﻓﺮﺷﻨﺪﻩﺍی ﮐﻪ ﺷﺮﺍﯾﻂ ﻓﺮوﺵ‬
‫~‬
‫~ ﺑﻪ ﻋﺪﺩ ﻓﺎﺯی ‪m‬‬ ‫~ ﻧﺴﺒﺖ ﺑﻪ ‪p‬‬ ‫ﻫﻤﭽﻨﯿﻦ ﻋﺪﺩ ﻓﺎﺯی ‪n‬‬
‫ﺁﻥ ﺑﺎ ﻧﯿﺎﺯﻣﻨﺪیﻫﺎی ﺍﻋﻼﻡ ﺷﺪﻩ ﺍﺯ ﻃﺮﻑ ﺧﺮﯾﺪﺍﺭ ﻣﺸﺎﺑﻬﺖ‬
‫ﺩﺍﺷﺘﻪ ﺑﺎﺷﺪ ﺭﺍ ﺍﻧﺘﺨﺎﺏ ﻣﯽﻧﻤﺎﯾﺪ‪ .‬ﺍﯾﻦ ﻣﺪﻝ ﺩﺭ ‪ 6‬ﻣﺮﺣﻠﻪ‬ ‫ﻧﺰﺩﯾﮑﺘﺮ ﺧﻮﺍﻫﺪ ﺑﻮﺩ ﺍﮔﺮ ﺩﺍﺷﺘﻪ ﺑﺎﺷﯿﻢ‪:‬‬
‫~ ‪~,‬‬
‫‪d v (m‬‬ ‫~ ‪~,‬‬
‫‪n) < d v (m‬‬ ‫)‪p‬‬
‫ﭘﯿﺎﺩﻩﺳﺎﺯی ﻣﯽﺷﻮﺩ‪:‬‬
‫ﺑﻬﺮ ﺣﺎﻝ ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﺍﻣﮑﺎﻥ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﻣﻔﺎﻫﯿﻢ ﻓﺎﺯی‪،‬‬
‫ﻣﯽﺗﻮﺍﻥ ﺭوﺷﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ‪ CBR‬ﺭﺍ ﺩﺭ ﺯﻣﯿﻨﻪﻫﺎی ﺯﯾﺮ‬
‫ﺑﺮﭘﺎﺋﯽ ﭘﺎﯾﮕﺎﻩ )ﻣﺨﺰﻥ( ﺧﺮﯾﺪﻫﺎ‬
‫ﺗﻮﺳﻌﻪ ﺩﺍﺩ‪:‬‬
‫"ﻣﺨﺰﻥ ﺧﺮﯾﺪﻫﺎ" ﺩﺭ وﺍﻗﻊ ﺳﺎﺧﺘﺎﺭی ﺍﺳﺖ ﮐﻪ‬
‫‪ – 1‬ﻣﻨﻄﻖ ﻓﺎﺯی ﺍﻣﮑﺎﻥ ﻧﻤﺎﯾﺶ "ﻣﻮﺭﺩ"ﻫﺎﺋﯽ ﮐﻪ ﻣﻘﺎﺩﯾﺮ‬
‫ﺩﺭﺑﺮﮔﯿﺮﻧﺪۀ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪﻫﺎی ﺍﻧﺠﺎﻡ ﺷﺪۀ ﻗﺒﻠﯽ و‬
‫ﻫﻤﻪ ﯾﺎ ﺑﺮﺧﯽ ﺍﺯ ﻣﺸﺨﺼﻪﻫﺎی ﺁﻥ ﺑﺼﻮﺭﺕ ﻏﯿﺮﺩﻗﯿﻖ ﺍﺭﺯﯾﺎﺑﯽ‬
‫ﻓﺮوﺷﻨﺪﻩﻫﺎی ﻣﻨﺘﺨﺐ ﻣﺮﺗﺒﻂ ﺑﺎ ﻫﺮ ﮐﺪﺍﻡ ﺍﺯ ﺁﻧﻬﺎﺳﺖ‪.‬‬
‫ﻣﯽﺷﻮﻧﺪ‪ ،‬ﺭﺍ ﻓﺮﺍﻫﻢ ﻣﯽﺁوﺭﺩ‪.‬‬
‫ﻫﻤﺎﻥﮔﻮﻧﻪ ﮐﻪ ﺩﺭ ﻣﺮوﺭ ﻣﻄﺎﻟﻌﺎﺗﯽ ﻧﯿﺰ ﺗﺎﮐﯿﺪ ﺷﺪ‪ ،‬ﻏﺎﻟﺒﺎً‬
‫‪ – 2‬ﺩﺭ ﺑﺮﺧﯽ ﺍﺯ ﻣﻮﺍﻗﻊ ﺧﻮﺩ "ﻣﺨﺰﻥ ﻣﻮﺭﺩﻫﺎ" ﻧﯿﺮ‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺑﺎ ﻋﻨﺎﯾﺖ ﺑﻪ ﭼﻨﺪﯾﻦ ﻣﻌﯿﺎﺭ ﺍﻧﺠﺎﻡ ﻣﯽﭘﺬﯾﺮﺩ‬
‫ﺑﺼﻮﺭﺕ ﺯﯾﺮ ﻣﺠﻤﻮﻋﮥ ﻓﺎﺯی ﻣﯽﺗﻮﺍﻧﺪ ﺩﺭ ﻧﻈﺮ ﮔﺮﻓﺘﻪ ﺷﻮﺩ؛ ﺑﻪ‬
‫و ﺑﻪ ﻫﻤﯿﻦ ﺩﻟﯿﻞ ﻧﯿﺰ‪ ،‬ﻣﺴﺎﻟﮥ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺩﺭ ﻗﺎﻟﺐ‬
‫ﺍﯾﻦ ﺗﺮﺗﯿﺐ ﮐﻪ ﻣﻮﺭﺩﻫﺎی ﺗﺸﮑﯿﻞ ﺩﻫﻨﺪۀ ﺁﻥ ﺑﺎ ﺩﺭﺟﻪﺍی ﺍﺯ‬
‫ﻣﺴﺎﺋﻞ ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ ﺗﻘﺴﯿﻢ ﺑﻨﺪی ﻣﯽﺷﻮﺩ‪ .‬ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ‪،‬‬
‫ﺍﻫﻤﯿﺖ ﺑﺮﺍی ﺍﺳﺘﻔﺎﺩﻩﻫﺎی ﺁﺗﯽ ﻣﻔﯿﺪ ﺑﺎﺷﻨﺪ‪.‬‬
‫ﻣﻌﯿﺎﺭﻫﺎی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺭﺍ ﺑﺼﻮﺭﺕ ﯾﮏ ﺳﻠﺴﻠﻪ ﻣﺮﺍﺗﺒﯽ‬
‫ﺩﺭ ﺗﺤﻘﯿﻖ ﺣﺎﺿﺮ ﺑﺎ ﺗﺎﮐﯿﺪ ﺑﺮ ﺯﻣﯿﻨﮥ ﺑﻨﺪ ‪ 1‬ﻓﻮﻕ‪ ،‬ﻧﺤﻮۀ‬
‫ﺩﺭ ﻧﻈﺮ ﻣﯽﮔﯿﺮﯾﻢ ﮐﻪ ﺍﺯ ﻋﻨﺎﺻﺮ ﺯﯾﺮ ﺗﺸﮑﯿﻞ ﺷﺪﻩ ﺍﺳﺖ‪:‬‬
‫ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺭوﺵ ‪ CBR‬ﺭﺍ ﺑﺮﺍی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺩﺭ‬
‫‪ –1‬ﻣﻌﯿﺎﺭ ﻗﯿﻤﺖ ؛ ﺷﺎﻣﻞ ‪:‬‬
‫ﻣﺤﯿﻄﻬﺎی ﻓﺎﺯی ﺍﺭﺍﺋﻪ ﻣﯽﮐﻨﯿﻢ‪.‬‬
 ‫‪575‬‬ ‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ ‪.....‬‬

‫ﺑﺮﭘﺎﺋﯽ ﺭوﺵ ﺑﺎﺯﯾﺎﺑﯽ‬ ‫‪ -1–1‬ﻫﺰﯾﻨﮥ ﺳﻔﺎﺵ ﺩﻫﯽ‬

‫ﺑﺎﺯﯾﺎﺑﯽ ﺧﺮﯾﺪﻫﺎی ﻗﺒﻠﯽ ﺑﺮﺍی ﯾﺎﻓﺘﻦ ﻓﺮوﺷﻨﺪﻩ ﯾﺎ‬ ‫‪ -2–1‬ﻫﺰﯾﻨﮥ ﺧﺮﯾﺪ ﻫﺮ وﺍﺣﺪ‬
‫ﻓﺮوﺷﻨﺪﻩﻫﺎﺋﯽ ﺍﺳﺖ ﮐﻪ ﺷﺮﺍﯾﻂ ﻓﺮوﺵ ﺁﻥ)ﻫﺎ( ﻣﺸﺎﺑﻬﺖ‬ ‫‪ -3-1‬ﻫﺰﯾﻨﮥ ﺣﻤﻞ ﻧﻘﻞ ﺑﻪ ﺍﺯﺍی ﻫﺮ وﺍﺣﺪ‬
‫ﺑﯿﺸﺘﺮی ﺑﺎ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺩﺍﺷﺘﻪ ﺑﺎﺷﺪ‪ .‬ﺑﺮﺍی ﺑﺎﺯﯾﺎﺑﯽ‬ ‫‪ –2‬ﻣﻌﯿﺎﺭ ﺗﺤﻮﯾﻞ ؛ ﻣﺘﺸﮑﻞ ﺍﺯ ‪:‬‬
‫ﺧﺮﯾﺪﻫﺎی ﻣﺸﺎﺑﻪ‪ ،‬ﺑﺎﯾﺪ ﺷﺮﺍﯾﻂ ﺍﻋﻼﻡ ﺷﺪﻩ ﺑﺮﺍی ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬ ‫‪ –1–2‬ﻣﺪﺕ ﺯﻣﺎﻥ ﻣﻮﺭﺩ ﻧﯿﺎﺯ ﺑﺮﺍی ﺗﺤﻮﯾﻞ ﮐﺎﻻ‬
‫ﺭﺍ ﺑﺎ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪﻫﺎی ﻗﺒﻠﯽ ﻣﻄﺎﺑﻘﺖ ﺩﺍﺩﻩ و ﻣﻮﺍﺭﺩ ﻣﺸﺎﺑﻪ‬ ‫‪ –2–2‬ﺍﻧﻌﻄﺎﻑ ﭘﺬﯾﺮی ﺩﺭ ﺗﻐﯿﯿﺮ ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ‬
‫ﺭﺍ ﺑﺎ ﺷﺎﺧﺼﯽ ﮐﻪ ﺩﺭﺟﮥ ﻣﺸﺎﺑﻬﺖ ﺭﺍ ﻧﺸﺎﻥ ﻣﯽﺩﻫﺪ‪ ،‬ﺍﺳﺘﺨﺮﺍﺝ‬ ‫‪ -3‬ﻣﻌﯿﺎﺭ ﮐﯿﻔﯿﺖ ؛ ﺩﺭ ﺑﺮﮔﯿﺮﻧﺪۀ ‪:‬‬
‫ﻧﻤﻮﺩ‪.‬‬ ‫‪ –1–3‬ﺩﺭﺻﺪ ﮐﺎﻻی ﻣﻌﯿﻮﺏ ﺩﺭ ﺍﻗﻼﻡ ﺗﺤﻮﯾﻠﯽ‬
‫ﺭوﺵ ﻣﺸﺎﺑﻬﺖﺳﻨﺠﯽ ﭘﯿﺸﻨﻬﺎﺩی ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ ﺑﺮ‬ ‫‪ –2–3‬ﻣﺘﻮﺳﻂ ﺯﻣﺎﻥ ﻻﺯﻡ ﺑﺮﺍی ﻣﺮﺍﺟﻌﻪ ﺟﻬﺖ ﺭﻓﻊ ﻣﺸﮑﻞ‬
‫ﺍﺳﺎﺱ ﺭﺍﺑﻄﮥ )‪ (1‬ﻋﻤﻞ ﻣﯽﻧﻤﺎﯾﺪ‪ .‬ﺍﺑﺘﺪﺍ ﻻﺯﻡ ﺍﺳﺖ ﻣﯿﺰﺍﻥ‬ ‫‪ –3–3‬ﺑﺮﺧﻮﺭﺩﺍﺭی ﺍﺯ ﺗﻮﺍﻧﺎﺋﯽ ﻓﻨﯽ ﺩﺭ ﺭﻓﻊ ﻣﺸﮑﻼﺕ‬
‫ﺍﻫﻤﯿﺖ ﯾﺎ وﺯﻥ ﻫﺮ ﮐﺪﺍﻡ ﺍﺯ ﻣﻌﯿﺎﺭﻫﺎی ﺧﺮﯾﺪ ﻣﺸﺨﺺ ﮔﺮﺩﺩ‪.‬‬ ‫ﺩﻟﯿﻞ ﺍﻧﺘﺨﺎﺏ ﺍﯾﻦ ﻣﻌﯿﺎﺭﻫﺎ‪ ،‬ﺗﻌﺪﺩ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺁﻧﻬﺎ ﺩﺭ‬
‫ﺍﺯ ﺭوﺷﻬﺎی ﻣﺘﻌﺪﺩی ﺑﺮﺍی ﺗﻌﯿﯿﻦ وﺯﻥ ﻣﻌﯿﺎﺭﻫﺎ ﻣﯽﺗﻮﺍﻥ‬ ‫ﺗﺤﻘﯿﻘﺎﺕ ﭘﯿﺸﯿﻦ ﺍﺳﺖ و ﻫﻤﺎﻧﮕﻮﻧﻪ ﮐﻪ ﭘﯿﺶﺗﺮ ﻧﯿﺰ ﺍﺷﺎﺭﻩ‬
‫ﺍﺳﺘﻔﺎﺩﻩ ﻧﻤﻮﺩ ﮐﻪ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﻧﻈﺮﺍﺕ ﻣﺴﺘﻘﯿﻢ ﺗﺼﻤﯿﻢ‬ ‫ﮐﺮﺩﯾﻢ ﻧﻮﻉ ﻣﻌﯿﺎﺭﻫﺎ ﺑﺴﺘﻪ ﺑﻪ ﺷﺮﺍﯾﻂ ﻣﺴﺎﻟﻪ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ‪،‬‬
‫ﮔﯿﺮﻧﺪﮔﺎﻥ و ﯾﺎ ﺭوﺷﻬﺎی ﺗﺤﻠﯿﻠﯽ ﻫﻤﭽﻮﻥ ﺭوﺵ ﻓﺮﺁﯾﻨﺪ‬ ‫ﻣﺘﻐﯿﺮ ﻣﯽﺑﺎﺷﺪ‪ .‬ﻧﺤﻮۀ ﺍﺭﺯﯾﺎﺑﯽ ﻣﻌﯿﺎﺭﻫﺎی ﻓﻮﻕ ﺩﺭ ﺟﺪوﻝ )‪(1‬‬
‫ﺗﺤﻠﯿﻞ ﺳﻠﺴﻠﻪ ﻣﺮﺍﺗﺒﯽ ﮐﻪ ﺑﺮ ﭘﺎﯾﮥ ﻣﻘﺎﯾﺴﺎﺕ ﺯوﺟﯽ ﺍﺳﺘﻮﺍﺭ‬ ‫ﺗﻮﺿﯿﺢ ﺩﺍﺩﻩ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬
‫ﺍﺳﺖ‪ ،‬ﻧﻤﻮﻧﻪﻫﺎﺋﯽ ﺍﺯ ﺁﻥ ﺍﺳﺖ‪ .‬ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ ﺭوﺵ ﺍﺧﯿﺮ‬ ‫ﺑﺮﺍی ﮐﻤِﯽ ﻧﻤﻮﺩﻥ ﻣﺘﻐﯿﺮﻫﺎی ﮐﻼﻣﯽ ﺍﺯ ﺗﺌﻮﺭی ﻣﻨﻄﻖ‬
‫)‪ (AHP‬ﺑﺮﺍی ﺗﻌﯿﯿﻦ ﺍوﺯﺍﻥ ﭘﯿﺸﻨﻬﺎﺩ ﻣﯽﺷﻮﺩ‪ .‬ﺩﺭ ﻗﺪﻡ ﺑﻌﺪ‬ ‫ﻓﺎﺯی ﺑﻬﺮﻩ ﮔﺮﻓﺘﻪ و ﺁﻧﻬﺎ ﺭﺍ ﺑﺎ ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﻣﻄﺎﺑﻖ‬
‫ﺑﺎﯾﺪ ﻧﺤﻮۀ ﺳﻨﺠﺶ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﻣﻘﺎﺩﯾﺮ ﻣﻌﯿﺎﺭﻫﺎ ﺑﺼﻮﺭﺕ‬ ‫ﺁﻧﭽﻪ ﮐﻪ ﺩﺭ ﺷﮑﻞ )‪ (5‬ﻧﺸﺎﻥ ﺩﺍﺩﻩ ﺷﺪﻩ‪ ،‬ﺑﺮ ﺭوی ﻣﺠﻤﻮﻋﮥ‬
‫ﺗﮑﯽ ﻣﺸﺨﺺ ﮔﺮﺩﺩ ﮐﻪ ﺫﯾﻼً ﺑﻪ ﺁﻥ ﻣﯽﭘﺮﺩﺍﺯﯾﻢ‪:‬‬ ‫ﻣﺮﺟﻊ ]‪ U=[0,10‬ﻧﻤﺎﯾﺶ ﻣﯽﺩﻫﯿﻢ‪ .‬ﺑﻪ ﻋﻨﻮﺍﻥ ﻣﺜﺎﻝ ﺑﺮﺍی‬
‫ﻧﺸﺎﻥ ﺩﺍﺩﻥ وﺿﻌﯿﺖ "ﻣﺘﻮﺳﻂ ﺭو ﺑﻪ ﺑﺎﻻ" ﺍﺯ ﻋﺪﺩ ﻓﺎﺯی‬
‫ﺳﻨﺠﺶ ﻣﺸﺎﺑﻬﺖ ﻣﻌﯿﺎﺭﻫﺎی ﻗﻄﻌﯽ‬ ‫]‪ [5,6,7,8‬ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﮐﻨﯿﻢ‪.‬‬
‫درﺟﮥ‬
‫ﺑﺮﺍی ﺳﻨﺠﺶ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﻣﻘﺎﺩﯾﺮ ﻣﺮﺑﻮﻁ ﺑﻪ ﺍﯾﻦ‬ ‫ﻋﻀــﻮﯾﺖ‬
‫ﻣﻌﯿﺎﺭﻫﺎ‪ ،‬ﺍﺯ ﺭﺍﺑﻄﮥ )‪ (2‬ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﮐﻨﯿﻢ ﮐﻪ ﺩﺭ ﻋﻤﺪۀ ﻣﺮﺍﺟﻊ‬ ‫ﺧﯿﻠﯽ‬ ‫ﻣﺘﻮﺳﻂ ﺭو‬ ‫ﻣﺘﻮﺳﻂ ﺭو‬
‫ﺿﻌﯿﻒ‬ ‫ﺿﻌﯿﻒ‬ ‫ﺑﻪ ﭘﺎﺋﯿﻦ‬ ‫ﻣﺘﻮﺳﻂ‬ ‫ﺑﻪ ﺑﺎﻻ‬ ‫ﺧﻮﺏ‬ ‫ﻋﺎﻟﯽ‬
‫ﻣﺮﺑﻮﻁ ﺑﻪ ﺭوﺵ ‪ CBR‬ﻧﯿﺰ ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬ ‫‪1‬‬

‫ﺳﻨﺠﺶ ﻣﺸﺎﺑﻬﺖ ﻣﻌﯿﺎﺭﻫﺎی ﻓﺎﺯی‬

‫ﺑﺮﺍی ﺳﻨﺠﺶ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﺑﯿﻦ ﻣﻘﺪﺍﺭ ﻣﺮﺑﻮﻁ ﺑﻪ ﻣﻌﯿﺎﺭ‬
‫~‬
‫ﻓﺎﺯی ‪ j‬ﺩﺭ ﺧﺮﯾﺪ ﻗﺒﻠﯽ ‪R‬ﺍﻡ‪ f Rj = [f Rj1 , f Rj2 , f Rj3 , f Rj4 ] ،‬و‬ ‫‪1‬‬ ‫‪2‬‬ ‫‪3‬‬ ‫‪4‬‬ ‫‪5‬‬ ‫‪6‬‬ ‫‪7‬‬ ‫‪8‬‬ ‫‪9‬‬ ‫‪10‬‬
‫ﻣﻘﺪﺍﺭ ﻣﻮﺭﺩ ﻧﻈﺮ ﺧﺮﯾﺪﺍﺭ ﺩﺭ ﻣﻮﺭﺩ ﻫﻤﯿﻦ ﻣﻌﯿﺎﺭ‪ ،‬ﯾﻌﻨﯽ‬ ‫ﺷﮑﻞ ‪ : 5‬ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﻣﺘﻨﺎﻇﺮ ﺑﺎ ﻣﺘﻐﯿﺮﻫﺎی ﮐﻼﻣﯽ‪.‬‬
‫~‬
‫] ‪ ، f Ij = [f Ij1 , f Ij2 , f Ij3 , f Ij4‬ﺍﺯ ﺭوﺵ ﺍﺭﺍﺋﻪ ﺷﺪﻩ ﺩﺭ ﻣﺮﺟﻊ‬
‫‪18‬‬
‫]‪ [39‬ﻣﻮﺳﻮﻡ ﺑﻪ ﺭوﺵ ﻓﺎﺻﻠﮥ ﻣﯿﺎﻧﮕﯿﻦ ﻣﺪﺭﺝ ﺗﻠﻔﯿﻘﯽ‬ ‫ﺑﺎ ﺛﺒﺖ و ﻧﮕﻬﺪﺍﺭی ﺍﻃﻼﻋﺎﺕ ﺧﺮﯾﺪﻫﺎی ﺍﻧﺠﺎﻡ ﺷﺪۀ ﻗﺒﻠﯽ‬
‫ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﮐﻨﯿﻢ ﮐﻪ ﺗﺎﺑﻊ ﺯﯾﺮ ﺭﺍ ﺑﺮﺍی ﻣﺤﺎﺳﺒﮥ ﻣﯿﺰﺍﻥ‬ ‫ﮐﻪ ﺷﺎﻣﻞ ﻣﻘﺪﺍﺭ ﺍﺭﺯﯾﺎﺑﯽ ﻣﻌﯿﺎﺭﻫﺎی ﺧﺮﯾﺪ و ﻧﺎﻡ ﻓﺮوﺷﻨﺪۀ‬
‫ﻣﺸﺎﺑﻬﺖ ﺑﯿﻦ ﺩو ﻣﻘﺪﺍﺭ ﻓﻮﻕ ﭘﯿﺸﻨﻬﺎﺩ ﺩﺍﺩﻩ ﺍﺳﺖ‪:‬‬ ‫ﻣﻨﺘﺨﺐ ﻫﺮ ﺧﺮﯾﺪ ﺍﺳﺖ‪ ،‬ﻣﺨﺰﻥ ﺧﺮﯾﺪ ﺗﺸﮑﯿﻞ ﻣﯽﮔﺮﺩﺩ‪.‬‬
‫ﻫﻤﭽﻨﯿﻦ ﺍﻣﮑﺎﻥ وﺍﺭﺩ ﮐﺮﺩﻥ ﺍﻃﻼﻋﺎﺕ ﻓﺮوﺷﻨﺪﮔﺎﻥ و‬
‫~ ~‬ ‫~‬ ‫~‬ ‫ﭘﯿﺸﻨﻬﺎﺩﺩﻫﻨﺪﮔﺎﻥ ﺟﺪﯾﺪ ﺩﺭ ﻣﺨﺰﻥ ﺍﻃﻼﻋﺎﺕ ﺧﺮﯾﺪ وﺟﻮﺩ‬
‫‪sim ( f Ij , f Rj ) = [ 1+ | P( f Ij ) − P( f Rj ) | ]−1‬‬
‫)‪( 9‬‬ ‫ﺩﺍﺭﺩ‪ .‬ﺩﺭ ﺟﺪوﻝ )‪ (2‬ﺳﺎﺧﺘﺎﺭ ﯾﮏ ﭘﺎﯾﮕﺎﻩ )ﻣﺨﺰﻥ( ﺧﺮﯾﺪ ﺑﻪ‬
‫ﺗﺼﻮﯾﺮ ﮐﺸﯿﺪﻩ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬
‫ﮐﻪ ﺩﺭ ﺁﻥ ‪:‬‬
 ‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪1385‬‬ ‫‪576‬‬

‫ﺟﺪوﻝ ‪ : 1‬ﻧﺤﻮۀ ﺍﺭﺯﯾﺎﺑﯽ ﻣﻌﯿﺎﺭﻫﺎی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ‪.‬‬

‫ﻣﺜﺎل‬ ‫ﻧﻮع ارزﯾﺎﺑﯽ‬ ‫ﻧﺎم زﯾﺮ ﻣﻌﯿﺎر‬ ‫ﻧﺎم ﻣﻌﯿﺎر‬
‫‪245‬‬ ‫ﻗﻄﻌﯽ‬ ‫ﻫﺰﯾﻨﮥ ﺳﻔﺎﺭﺵ ﺩﻫﯽ )ﺩﻻﺭ(‬
‫‪1.95‬‬ ‫ﻗﻄﻌﯽ‬ ‫ﻫﺰﯾﻨﮥ ﺧﺮﯾﺪ )ﺩﻻﺭ(‬ ‫ﻗﯿﻤﺖ‬
‫]‪[0.30،0.40،0.45،0.50‬‬ ‫ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی‬ ‫ﻫﺰﯾﻨﮥ ﺍﻧﺘﻘﺎﻝ )ﺩﻻﺭ(‬
‫]‪[3.0،5.0،5.0،7.0‬‬ ‫ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی‬ ‫ﻣﺪﺕ ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ )ﺭوﺯ(‬
‫ﺗﺤﻮﯾﻞ‬
‫ﻣﺘﻮﺳﻂ‬ ‫ﻣﺘﻐﯿﺮ ﮐﻼﻣﯽ‬ ‫ﺍﻧﻌﻄﺎﻑ ﭘﺬﯾﺮی ﺩﺭ ﺗﺤﻮﯾﻞ‬
‫‪3.9‬‬ ‫ﻗﻄﻌﯽ‬ ‫ﺩﺭﺻﺪ ﮐﺎﻻی ﻣﻌﯿﻮﺏ )‪(%‬‬
‫]‪[3.5،4.0،5.5،7.0‬‬ ‫ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی‬ ‫ﺯﻣﺎﻥ ﻣﺮﺍﺟﻌﻪ ﺟﻬﺖ ﺭﻓﻊ ﺍﺷﮑﺎﻝ )ﺳﺎﻋﺖ(‬ ‫ﮐﯿﻔﯿﺖ‬
‫ﺧﻮﺏ‬ ‫ﻣﺘﻐﯿﺮ ﮐﻼﻣﯽ‬ ‫ﺗﻮﺍﻧﺎﺋﯽ ﻓﻨﯽ‬

‫ﺟﺪوﻝ ‪ : 2‬ﺳﺎﺧﺘﺎﺭ ﻧﮕﻬﺪﺍﺭی ﺍﻃﻼﻋﺎﺕ ﺧﺮﯾﺪﻫﺎ )ﻣﺨﺰﻥ ﺧﺮﯾﺪﻫﺎ(‪.‬‬

‫ﻣﺸﺨﺼﺎت ﺧﺮﯾﺪﻫﺎی ﻗﺒﻠﯽ‬ ‫ﻧﻮع‬
‫ﻧﺎم ﻣﺸﺨﺼﻪ )ﻣﻌﯿﺎر(‬
‫‪....‬‬ ‫ﺧﺮﯾﺪ ‪2‬‬ ‫ﺧﺮﯾﺪ ‪1‬‬ ‫ﻣﻌﯿﺎر‬
‫‪....‬‬ ‫‪240‬‬ ‫‪245‬‬ ‫ﻗﻄﻌﯽ‬ ‫ﻫﺰﯾﻨﮥ ﺳﻔﺎﺭﺵ ﺩﻫﯽ )ﺩﻻﺭ(‬
‫‪....‬‬ ‫‪1.90‬‬ ‫‪1.95‬‬ ‫ﻗﻄﻌﯽ‬ ‫ﻫﺰﯾﻨﮥ ﺧﺮﯾﺪ )ﺩﻻﺭ(‬ ‫ﻗﯿﻤﺖ‬
‫‪....‬‬ ‫]‪[0.30،0.37،0.45،0.55‬‬ ‫]‪[0.30،0.40،0.45،0.50‬‬ ‫ﻓﺎﺯی‬ ‫ﻫﺰﯾﻨﮥ ﺍﻧﺘﻘﺎﻝ )ﺩﻻﺭ(‬
‫‪....‬‬ ‫]‪[4.0،5.0،5.0،7.0‬‬ ‫]‪[3.0،5.0،5.0،7.0‬‬ ‫ﻓﺎﺯی‬ ‫ﻣﺪﺕ ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ )ﺭوﺯ(‬ ‫ﺗﺤﻮﯾ‬
‫‪....‬‬ ‫ﺧﻮﺏ‬ ‫ﻣﺘﻮﺳﻂ‬ ‫ﮐﻼﻣﯽ‬ ‫ﺍﻧﻌﻄﺎﻑ ﭘﺬﯾﺮی ﺩﺭ ﺗﺤﻮﯾﻞ‬ ‫ل‬
‫‪....‬‬ ‫‪3.8‬‬ ‫‪3.9‬‬ ‫ﻗﻄﻌﯽ‬ ‫ﺩﺭﺻﺪ ﮐﺎﻻی ﻣﻌﯿﻮﺏ )‪(%‬‬
‫ﮐﯿﻔﯿ‬
‫‪....‬‬ ‫]‪[3.0،4.0،5.0،6.0‬‬ ‫]‪[3.5،4.0،5.5،7.0‬‬ ‫ﻓﺎﺯی‬ ‫ﺯﻣﺎﻥ ﺭﻓﻊ ﺍﺷﮑﺎﻝ )ﺳﺎﻋﺖ(‬
‫ت‬
‫‪....‬‬ ‫ﻣﺘﻮﺳﻂ ﺭو ﺑﻪ ﺑﺎﻻ‬ ‫ﺧﻮﺏ‬ ‫ﮐﻼﻣﯽ‬ ‫ﺗﻮﺍﻧﺎﺋﯽ ﻓﻨﯽ‬
‫‪....‬‬ ‫ﻓﺮوﺷﻨﺪۀ ‪B‬‬ ‫ﻓﺮوﺷﻨﺪۀ ‪A‬‬ ‫ﻧﺎم ﻓﺮوﺷﻨﺪۀ ﻣﻨﺘﺨﺐ‬

‫ﻣﺨﺰﻥ ﺧﺮﯾﺪﻫﺎ‪ ،‬ﺑﺎﺯﯾﺎﺑﯽ و ﺍﻃﻼﻋﺎﺕ ﺁﻧﻬﺎ ﺑﻪ ﺷﺮﺡ ﺯﯾﺮ ﺍﺳﺘﺨﺮﺍﺝ‬ ‫~‬ ‫‪f Ij1 + 2f Ij2 + 2f Ij3 + f Ij4‬‬
‫= ) ‪P( fIj‬‬
‫ﻣﯽﺷﻮﺩ‪:‬‬ ‫‪6‬‬
‫~‬ ‫‪f Rj + 2f Rj + 2f Rj3 + f Rj4‬‬
‫‪1‬‬ ‫‪2‬‬

‫‪ : m‬ﺗﻌﺪﺍﺩ ﺧﺮﯾﺪﻫﺎی ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪۀ ﻣﻮﺭﺩ ﺍﺭﺯﯾﺎﺑﯽ )ﺑﻪ ﺍﻧﺘﺨﺎﺏ‬
‫ﺗﺼﻤﯿﻢﮔﯿﺮﻧﺪﻩ( ﮐﻪ ﺩﺭ ﺁﻥ ‪M>m>1‬‬
‫‪ : i‬ﺷﺎﺧﺺ )ﺷﻤﺎﺭۀ( ﺧﺮﯾﺪ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ )‪(i=1,2,…,m‬‬
‫‪ : n‬ﺗﻌﺪﺍﺩ ﻣﻌﯿﺎﺭﻫﺎ ﯾﺎ ﻣﺸﺨﺼﻪﻫﺎی ﺧﺮﯾﺪ‬
‫‪ : j‬ﺷﺎﺧﺺ )ﺷﻤﺎﺭۀ( ﻣﻌﯿﺎﺭ )‪(j=1,2,…,n‬‬
‫~‬
‫‪ : fij‬ﻣﻘﺪﺍﺭ ﻣﺸﺨﺼﻪ ﯾﺎ ﻣﻌﯿﺎﺭ ‪ j‬ﺩﺭ ﺧﺮﯾﺪ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ ‪i‬ﺍﻡ‬
‫‪ : Wj‬ﺍﻫﻤﯿﺖ ﻧﺴﺒﯽ ﯾﺎ وﺯﻥ ﻣﻌﯿﺎﺭ ‪j‬ﺍﻡ )ﻣﺤﺎﺳﺒﻪ ﺷﺪﻩ ﺩﺭ‬
‫ﻣﺮﺣﻠﮥ ‪(2-5‬‬
‫~‬
‫‪ : D=[ fij ]mxn‬ﻣﺎﺗﺮﯾﺲ ﺗﺼﻤﯿﻢ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ‬
‫‪ 2-5‬ﺑﺮ ﺭوی ﺗﻤﺎﻣﯽ ﻣﻮﺍﺭﺩ ﺧﺮﯾﺪ ﻗﺒﻠﯽ ﺍﻋﻤﺎﻝ ﻣﯽﺷﻮﺩ‪ .‬ﺩﺭ‬
‫ﺍوﻟﻮﯾﺖﺑﻨﺪی ﻓﺮوﺷﻨﺪﮔﺎﻥ‬
‫ﺍﺟﺮﺍی ﻣﺮﺣﻠﮥ ﻗﺒﻞ‪ ،‬ﺩﺭ وﺍﻗﻊ ﺑﻪ ﻓﺮﺍﻫﻢ ﺁﻣﺪﻥ ﯾﮏ ﻣﺴﺎﻟﮥ‬
‫ﺗﺼﻤﯿﻢﮔﯿﺮی ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ ﺑﺮﺍی ﺍﺭﺯﯾﺎﺑﯽ و ﺍوﻟﻮﯾﺖﺑﻨﺪی‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ و ﺩﺭ ﻧﻬﺎﯾﺖ ﺍﻧﺘﺨﺎﺏ ﻧﻬﺎﺋﯽ ﻓﺮوﺷﻨﺪﻩ )ﺍﺯ ﻣﯿﺎﻥ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﻣﺮﺑﻮﻁ ﺑﻪ ‪ m‬ﺧﺮﯾﺪ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ( ﻣﻨﺠﺮ ﻣﯽﺷﻮﺩ‪.‬‬
‫= ) ‪P( fRj‬‬ ‫)‪(10‬‬
‫‪6‬‬
‫ﺣﺎﻝ ﺑﺎ ﻣﺸﺨﺺ ﺷﺪﻥ ﻧﺤﻮۀ ﻣﺤﺎﺳﺒﮥ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖٍ ﻣﻘﺎﺩﯾﺮ‬
‫ﻣﻌﯿﺎﺭﻫﺎ ﺑﺼﻮﺭﺕ ﻣﻨﻔﺮﺩ‪ ،‬ﻣﯽﺗﻮﺍﻥ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﺷﺮﺍﯾﻂ‬
‫ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺑﺎ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪﻫﺎی ﻣﻮﺟﻮﺩ ﺩﺭ ﻣﺨﺰﻥ‬
‫ﺧﺮﯾﺪﻫﺎ ﺭﺍ ﺑﻪ ﮐﻤﮏ ﺭﺍﺑﻄﮥ )‪ (1‬ﺑﺪﺳﺖ ﺁوﺭﺩ‪.‬‬
‫ﺍٍﻋﻤﺎﻝ ﺭوﺵ ﺑﺎﺯﯾﺎﺑﯽ ﺑﺮ ﺭوی ﮐﻠﯿﻪ ﻣﻮﺍﺭﺩ ﺧﺮﯾﺪ‬
‫ﺩﺭ ﺍﯾﻦ ﻣﺮﺣﻠﻪ ﺑﺎ ﻣﺸﺨﺺ ﺷﺪﻥ ﻧﯿﺎﺯﻣﻨﺪیﻫﺎی ﺧﺮﯾﺪ‬
‫ﺟﺪﯾﺪ ﺗﻮﺳﻂ ﺧﺮﯾﺪﺍﺭ‪ ،‬ﺭوﺵ ﺑﺎﺯﯾﺎﺑﯽ ﻣﻌﺮﻓﯽ ﺷﺪﻩ ﺩﺭ ﻣﺮﺣﻠﮥ‬
‫ﺻﻮﺭﺗﯿﮑﻪ ﻓﺮوﺷﻨﺪﻩﻫﺎی ﺟﺪﯾﺪی ﻏﯿﺮ ﺍﺯ ﻣﻮﺍﺭﺩی ﮐﻪ ﺍﻃﻼﻋﺎﺕ‬
‫ﺁﻧﻬﺎ ﺩﺭ ﻣﺨﺰﻥ ﺧﺮﯾﺪﻫﺎ ﻣﻮﺟﻮﺩ ﺍﺳﺖ‪ ،‬ﺑﺮﺍی ﻓﺮوﺵ ﮐﺎﻻی‬
‫ﻣﻮﺭﺩ ﻧﯿﺎﺯ ﺧﺮﯾﺪﺍﺭ ﺍﻋﻼﻡ ﺁﻣﺎﺩﮔﯽ ﮐﺮﺩﻩ ﺑﺎﺷﻨﺪ‪ ،‬ﺭوﺵ ﺑﺎﺯﯾﺎﺑﯽ‬
‫ﺑﺮ ﺭوی ﺁﻧﻬﺎ ﻧﯿﺰ ﺍﻋﻤﺎﻝ ﻣﯽﮔﺮﺩﺩ‪ .‬ﺩﺭ ﻧﻬﺎﯾﺖ ﺑﺮ ﺍﺳﺎﺱ ﻧﻈﺮ‬
‫ﺗﺼﻤﯿﻢﮔﯿﺮﻧﺪﻩ‪ ،‬ﺗﻌﺪﺍﺩ ‪ m‬ﺧﺮﯾﺪ ﺍﺯ ﻣﯿﺎﻥ ‪ M‬ﺧﺮﯾﺪ ﻣﻮﺟﻮﺩ ﺩﺭ‬
 ‫‪577‬‬ ‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ ‪.....‬‬

‫ﮔﺎم ﺳﻮم ‪ -‬ﺗﻌﯿﯿﻦ ﻣﺸﺨﺼﺎﺕ ﻧﺮﻣﺎﻝ ﺷﺪۀ ﺧﺮﯾﺪﻫﺎی ﺍﯾﺪﻩﺁﻝ‬ ‫ﺑﺮﺍی ﺣﻞ ﺍﯾﻦ ﻣﺴﺎﻟﻪ‪ ،‬ﺭوﺵ ‪ TOPSIS‬ﻓﺎﺯی ﭘﯿﺸﻨﻬﺎﺩ‬
‫و ﺿﺪ ‪ -‬ﺍﯾﺪﻩﺁﻝ‬ ‫ﻣﯽﺷﻮﺩ ﮐﻪ ﺍوﻟﻮﯾﺖﺑﻨﺪی ﻓﺮوﺷﻨﺪﮔﺎﻥ ﻣﺮﺗﺒﻂ ﺑﺎ ﺧﺮﯾﺪﻫﺎی‬
‫ﻣﺸﺨﺼﺎﺕ ﻧﺮﻣﺎﻝ ﺷﺪۀ ﺧﺮﯾﺪ ﺍﯾﺪﻩﺁﻝ و ﺧﺮﯾﺪ ﺿﺪ‪-‬ﺍﯾﺪﺍﻩﺍﻝ‬ ‫ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ ﺭﺍ ﺩﺭ ‪ 5‬ﮔﺎﻡ ﺑﻪ ﺷﺮﺡ ﺯﯾﺮ ﺑﻪ ﺍﻧﺠﺎﻡ ﻣﯽﺭﺳﺎﻧﺪ‪.‬‬
‫~( = ‪ A +‬و ) ‪A − = (~y1− , ~y 2− ,..., ~y n−‬‬
‫ﮐﻪ ﺑﺼﻮﺭﺕ ) ‪y1+ , ~y 2+ ,..., ~y n+‬‬ ‫ﺗﻮﺻﯿﻪ ﻣﯽﺷﻮﺩ ﻗﺒﻞ ﺍﺯ ﻣﻄﺎﻟﻌﮥ ﺍﺩﺍﻣﮥ ﻣﻄﺎﻟﺐ ﻣﻘﻮﻟﮥ‬
‫ﻧﺸﺎﻥ ﺩﺍﺩﻩ ﻣﯽﺷﻮﻧﺪ‪ ،‬ﺍﺯ ﻣﺎﺗﺮﯾﺲ ﻧﺮﻣﺎﻝ ﺷﺪۀ‬ ‫ﺗﺼﻤﯿﻢﮔﯿﺮی ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ و ﺭوﺵ ‪ TOPSIS‬ﻓﺎﺯی ﺍﺯ ﻣﺮﺍﺟﻊ‬
‫~‬ ‫]‪ [40‬و ]‪ [41‬ﻣﻄﺎﻟﻌﻪ ﺷﻮﺩ‪.‬‬
‫‪ (Y‬ﺑﺪﺳﺖ ﻣﯽﺁﯾﻨﺪ‪:‬‬ ‫ﻣﻮﺯوﻥ )‬

‫] ‪~y + = [ y + , y + , y + , y +‬‬ ‫ﺑﻬﺮ ﺣﺎﻝ ﻫﻤﺎﻧﮕﻮﻧﻪ ﮐﻪ ﻋﻨﻮﺍﻥ ﺷﺪ ﻣﺎﺗﺮﯾﺲ ﺗﺼﻤﯿﻢ‬

‫‪j‬‬ ‫‪j‬‬ ‫‪j‬‬ ‫‪j‬‬ ‫‪j‬‬
‫ﺣﺎﺻﻠﻪ ﺩﺭ ﻣﺮﺣﻠﻪ ‪ 3-5‬ﺑﺼﻮﺭﺕ ﺯﯾﺮ ﻣﯽﺑﺎﺷﺪ‪:‬‬
‫‪y +j = max{y ij4 } ; i = 1,2,..., m ; j = 1,2,..., n‬‬ ‫~‬ ‫~‬ ‫~‬
‫‪i‬‬ ‫)‪(15‬‬ ‫‪f11‬‬ ‫‪f12‬‬ ‫‪...‬‬ ‫‪f1n‬‬
‫~‬ ‫~‬ ‫‪...‬‬ ‫~‬
‫] ‪~y− = [ y− , y− , y− , y−‬‬ ‫‪f.21‬‬ ‫‪f.22‬‬ ‫‪f‬‬
‫=‪D‬‬ ‫‪..‬‬ ‫‪..‬‬ ‫‪.‬‬ ‫‪.. 2n‬‬
‫‪j‬‬ ‫‪j‬‬ ‫‪j‬‬ ‫‪j‬‬ ‫‪j‬‬ ‫‪.‬‬ ‫‪.‬‬
‫~‬ ‫~‬ ‫‪.‬‬ ‫~‬
‫‪fm1‬‬ ‫‪fm2‬‬ ‫‪...‬‬ ‫‪fmn‬‬
‫‪y−j = min{y1ij} ; i = 1,2,..., m ; j = 1,2,..., n‬‬ ‫)‪(16‬‬
‫‪i‬‬
‫~‬
‫‪19‬‬ ‫ﺍﺳﺖ و ﻣﻘﺎﺩﯾﺮ ﻗﻄﻌﯽ‬ ‫] ‪fij = [f ij1 , f ij2 , f ij3 , f ij4‬‬ ‫ﮐﻪ ﺩﺭ ﺁﻥ‬
‫ﮔﺎم ﭼﻬﺎرم ‪ -‬ﻣﺤﺎﺳﺒﮥ ﺩﺭﺟﮥ ﻣﻔﺎﺭﻗﺖ‬
‫ﻣﺎﺗﺮﯾﺲ ﻓﻮﻕ ﺑﺼﻮﺭﺕ ﺍﻋﺪﺍی ﻓﺎﺯی ﺫوﺯﻧﻘﻪﺍی ﺑﺎ ﭘﺎﺭﻣﺘﺮﻫﺎی‬
‫ﺩﺭ ﺍﯾﻦ ﻣﺮﺣﻠﻪ ﺩﺭﺟﮥ ﻣﻔﺎﺭﻗﺖ ﻣﺸﺨﺼﺎﺕ ﻧﺮﻣﺎﻝ ﺷﺪۀ‬
‫ﯾﮑﺴﺎﻥ )ﺑﻪ ﺍﻧﺪﺍﺯۀ ﻣﻘﺪﺍﺭ ﻏﯿﺮ ﻓﺎﺯی( ﺩﺭ ﻧﻈﺮ ﮔﺮﻓﺘﻪ ﺧﻮﺍﻫﻨﺪ‬
‫ﺧﺮﯾﺪﻫﺎی ﺩﺭ ﺣﺎﻝ ﺍﺭﺯﯾﺎﺑﯽ ﺑﺎ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺍﯾﺪﻩﺁﻝ ) ‪( D i+‬‬
‫ﺷﺪ‪.‬‬
‫و ﻧﯿﺰ ﺧﺮﯾﺪ ﺿﺪ‪-‬ﺍﯾﺪﻩﺁﻝ ) ‪ ( Di−‬ﻣﻄﺎﺑﻖ ﺭﺍﺑﻄﮥ )‪ (8‬ﻣﺤﺎﺳﺒﻪ‬
‫~‬
‫ﻣﯽﺷﻮﺩ‪:‬‬ ‫)‪(R‬‬ ‫ﮔﺎم اول – ﺗﺸﮑﯿﻞ ﻣﺎﺗﺮﯾﺲ ﻧﺮﻣﺎﻝ ﺷﺪﻩ ﻓﺎﺯی‬
‫ﻋﻨﺎﺻﺮ ﺗﺸﮑﯿﻞ ﺩﻫﻨﺪۀ ﺍﯾﻦ ﻣﺎﺗﺮﯾﺲ‪ ،‬ﯾﻌﻨﯽ ‪ ~rij‬ﻫﺎ‪،‬‬
‫‪n‬‬
‫) ‪D i+ = ∑ d v (~y ij , ~y +j‬‬ ‫ﺑﺼﻮﺭﺕ ﺯﯾﺮ ﻣﺤﺎﺳﺒﻪ ﻣﯽﺷﻮﺩ‪:‬‬
‫‪j=1‬‬
‫‪; i = 1,2,..., m‬‬ ‫‪1‬‬ ‫‪2‬‬ ‫‪3‬‬ ‫‪4‬‬
‫‪n‬‬
‫] ‪~r = [ f , f , f , f‬‬
‫) ‪D i− = ∑ d v (~y ij , ~y −j‬‬
‫‪ij‬‬ ‫‪ij‬‬ ‫‪ij‬‬ ‫‪ij‬‬
‫‪ij‬‬ ‫‪4‬‬
‫‪f f f f‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫‪4‬‬ ‫ﺑﺮﺍی ﻣﻌﯿﺎﺭﻫﺎی )‪ (j‬ﻣﺜﺒﺖ‬
‫‪j=1‬‬ ‫‪+j‬‬ ‫‪+j‬‬ ‫‪+j‬‬ ‫‪+j‬‬
‫)‪(17‬‬ ‫} ‪f = max {f‬‬
‫‪4‬‬ ‫‪4‬‬ ‫)ﯾﺎ ﺳﻮﺩ(‬
‫‪+j‬‬ ‫‪ij‬‬
‫‪20‬‬
‫ﮔﺎم ﭘﻨﺠﻢ ‪ -‬ﻣﺤﺎﺳﺒﮥ ﺷﺎﺧﺺ ﻗﺮﺍﺑﺖ‬
‫‪i‬‬ ‫)‪(11‬‬
‫‪1‬‬ ‫‪1‬‬ ‫‪1‬‬ ‫‪1‬‬
‫ﺩﺭ ﺍﯾﻦ ﻗﺪﻡ‪ ،‬ﺷﺎﺧﺺ ﻗﺮﺍﺑﺖ ﻣﺸﺨﺼﺎﺕ ﻫﺮ ﮐﺪﺍﻡ ﺍﺯ ﺧﺮﯾﺪﻫﺎی‬ ‫] ‪~r = [ f − j , f − j , f − j , f − j‬‬
‫‪ij‬‬ ‫ﺑﺮﺍی ﻣﻌﯿﺎﺭﻫﺎی )‪ (j‬ﻣﻨﻔﯽ‬
‫‪f ij4 f ij3 f ij2 f ij1‬‬
‫ﺩﺭ ﺣﺎﻝ ﺍﺭﺯﯾﺎﺑﯽ ﺑﻪ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺍﯾﺪﻩﺁﻝ‪ ،‬ﺑﺼﻮﺭﺕ ﺯﯾﺮ ﻣﺤﺎﺳﺒﻪ‬ ‫)ﯾﺎ ﻫﺰﯾﻨﻪ(‬
‫}‪f -1j = min {f ij1‬‬ ‫)‪(12‬‬
‫‪i‬‬
‫ﻣﯽﺷﻮﺩ‪:‬‬
‫ﻣﻌﯿﺎﺭ ﻣﺜﺒﺖ ﺑﻪ ﻣﻌﯿﺎﺭی ﺍﻃﻼﻕ ﻣﯽﺷﻮﺩ ﮐﻪ ﻣﻘﺎﺩﯾﺮ‬
‫‪−‬‬
‫‪D‬‬ ‫ﺍﺭﺯﯾﺎﺑﯽِ ﺑﯿﺸﺘﺮ ﻣﻄﻠﻮﺑﯿﺖ ﺑﺎﻻﺗﺮی ﺩﺍﺭﻧﺪ )ﻫﻤﭽﻮﻥ ﻣﻌﯿﺎﺭ‬
‫= ‪Ci‬‬ ‫‪i‬‬
‫;‬ ‫‪i = 1,2,..., m‬‬
‫‪D + D i−‬‬
‫‪+‬‬
‫‪i‬‬ ‫"ﺩﺭﺻﺪ ﺗﺤﻮﯾﻞ ﺑﻤﻮﻗﻊ" ﮐﻪ ﻣﻘﺪﺍﺭ ﺁﻥ ﻫﺮ ﭼﻘﺪﺭ ﺑﯿﺸﺘﺮ ﺑﺎﺷﺪ‬
‫)‪(18‬‬
‫ﺑﻬﺘﺮ ﺍﺳﺖ( و ﺩﺭ ﺭﺍﺑﻄﻪ ﺑﺎ ﻣﻌﯿﺎﺭﻫﺎی ﻣﻨﻔﯽ )ﻫﻤﭽﻮﻥ ﺩﺭﺻﺪ‬
‫ﻣﻘﺎﺩﯾﺮ ﺑﺰﺭﮔﺘﺮ ‪ Ci‬ﮐﻪ ﺍﺯ ﺍﯾﻦ ﺑﻪ ﺑﻌﺪ ﺑﻪ ﺁﻥ ﺷﺎﺧﺺ‬
‫ﮐﺎﻻی ﻣﻌﯿﻮﺏ(‪ ،‬ﺑﺮ ﺧﻼﻑ ﻣﻌﯿﺎﺭﻫﺎی ﻣﺜﺒﺖ‪ ،‬ﻣﻘﺎﺩﯾﺮ ﺍﺭﺯﯾﺎﺑﯽِ‬
‫ﺍﺭﺟﺤﯿﺖ ﺧﺮﯾﺪ ‪ i‬ﺍﻃﻼﻕ ﺧﻮﺍﻫﺪ ﺷﺪ‪ ،‬ﺑﯿﺎﻧﮕﺮ ﺍﺭﺟﺤﯿﺖ ﺑﯿﺸﺘﺮ‬
‫ﮐﻤﺘﺮ ﻣﻄﻠﻮﺏﺗﺮﻧﺪ‪.‬‬
‫ﯾﺎ ﺍوﻟﻮﯾﺖ ﺑﺎﻻﺗﺮ ﻓﺮوﺷﻨﺪۀ ﻣﺮﺑﻮﻃﻪ ﺑﺮﺍی ﺍﻧﺠﺎﻡ ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬ ‫~‬
‫ﮔﺎم دوم ‪ -‬ﺍﯾﺠﺎﺩ ﻣﺎﺗﺮﯾﺲ ﻧﺮﻣﺎﻝ ﺷﺪۀ ﻣﻮﺯوﻥ )‪(Y‬‬
‫ﺍﺳﺖ‪.‬‬
‫ﺍﯾﻦ ﻣﺎﺗﺮﯾﺲ ﺍﺯ ﺣﺎﺻﻠﻀﺮﺏ ﻋﻨﺎﺻﺮ ﺳﺘﻮﻧﯽ ﻣﺎﺗﺮﯾﺲ ‪ R‬ﺩﺭ‬
‫ﺗﻄﺒﯿﻖ ﺧﺮﯾﺪ‬ ‫وﺯﻥ ﻣﺘﻨﺎﻇﺮ ﺑﺎ ﻣﻌﯿﺎﺭ ﻫﺮ ﺳﺘﻮﻥ ﺑﻪ ﺷﺮﺡ ﺯﯾﺮ ﺑﺪﺳﺖ ﻣﯽﺁﯾﺪ‪:‬‬
‫ﺩﺭ ﺍﯾﻦ ﻣﺮﺣﻠﻪ ﺧﺮﯾﺪﻫﺎی ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ ﺑﻪ ﺗﺮﺗﯿﺐ ﺍوﻟﻮﯾﺖ‬ ‫~‬
‫‪Y = [~yij ]m×n ; i = 1,2,..., m ; j = 1,2,..., n‬‬
‫ﻣﺤﺎﺳﺒﻪ ﺷﺪﻩ ﺩﺭ ﻣﺮﺣﻠﮥ ‪ 4-5‬ﻣﻮﺭﺩ ﺍﺭﺯﯾﺎﺑﯽ ﻧﻬﺎﺋﯽ ﻗﺮﺍﺭ‬ ‫)‪(13‬‬
‫ﻣﯽﮔﯿﺮﻧﺪ ﺗﺎ ﻣﯿﺰﺍﻥ ﺍﻧﻄﺒﺎﻕ ﺁﻧﻬﺎ ﺑﺎ ﻧﯿﺎﺯﻣﻨﺪیﻫﺎی ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬ ‫ﮐﻪ ﺩﺭ ﺁﻥ‪:‬‬
‫ﻣﺸﺨﺺ ﺷﻮﺩ‪ .‬ﻣﺮﺣﻠﮥ ﺗﻄﺒﯿﻖ ﺧﺮﯾﺪ ﮐﻪ ﻣﻌﻤﻮﻻً ﺗﻮﺍﻡ ﺑﺎ‬ ‫‪~y = W ⊗ ~r ; i = 1,2,..., m ; j = 1,2,..., n‬‬
‫‪ij‬‬ ‫‪j‬‬ ‫‪ij‬‬
‫] ‪~y = [ y1 , y 2 , y 3 , y 4‬‬ ‫)‪(14‬‬
‫ﻣﺬﺍﮐﺮﻩ ﺑﺎ ﻓﺮوﺷﻨﺪﮔﺎﻥ ﺍﺳﺖ‪ ،‬ﻣﻨﺠﺮ ﺑﻪ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩﺍی‬ ‫‪ij‬‬ ‫‪ij‬‬ ‫‪ij‬‬ ‫‪ij‬‬ ‫‪ij‬‬
 ‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪1385‬‬ ‫‪578‬‬
‫ﺍﺩﺍﻣﮥ ﺟﺪوﻝ ‪ : 3‬ﻣﻘﺎﯾﺴﺎﺕ ﺯوﺟﯽ و ﻣﻘﺎﺩﯾﺮ ﻧﻬﺎﺋﯽ ﺍوﺯﺍﻥ‬ ‫ﻣﯽﺷﻮﺩ ﮐﻪ ﻧﯿﺎﺯﻫﺎی ﺍﻋﻼﻡ ﺷﺪﻩ ﺑﺮﺍی ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺭﺍ ﺑﺮﺁوﺭﺩﻩ‬
‫ﻣﻌﯿﺎﺭﻫﺎ‪.‬‬ ‫ﻣﯽﮐﻨﺪ‪.‬‬
‫ﺍوﺯﺍﻥ‬ ‫ﺍوﺯﺍﻥ‬
‫ﺗﺤﻮﯾﻞ‬ ‫ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ‬ ‫ﺍﻧﻌﻄﺎﻑ ﭘﺬﯾﺮی‬
‫ﻣﺤﻠﯽ‬ ‫ﻧﻬﺎﺋﯽ‬
‫ﺯﻣﺎﻥ‬ ‫‪0.71‬‬ ‫‪0.14‬‬ ‫ﺍﺿﺎﻓﻪ ﻧﻤﻮﺩﻥ ﻃﻼﻋﺎﺕ ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺩﺭ ﻣﺨﺰﻥ ﺧﺮﯾﺪﻫﺎ‬
‫‪1‬‬ ‫‪0.3‬‬
‫ﺗﺤﻮﯾﻞ‬ ‫‪4‬‬ ‫‪6‬‬ ‫ﺩﺭ ﺁﺧﺮﯾﻦ ﻣﺮﺣﻠﻪ‪ ،‬ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺑﻪ ﻫﻤﺮﺍﻩ‬
‫ﺍﻧﻌﻄﺎﻑ‬ ‫‪0.28‬‬ ‫‪0.05‬‬
‫‪1‬‬ ‫ﻓﺮوﺷﻨﺪۀ ﻣﻨﺘﺨﺐ ﺑﺮﺍی ﺁﻥ ﺩﺭ ﻣﺨﺰﻥ ﺧﺮﯾﺪﻫﺎ ﺫﺧﯿﺮﻩ و‬
‫ﭘﺬﯾﺮی‬ ‫‪6‬‬ ‫‪9‬‬
‫‪ = 0.00‬ﻧﺮﺥ ﻧﺎﭘﺎﯾﺪﺍﺭی‬ ‫ﻧﮕﻬﺪﺍﺭی ﻣﯽﺷﻮﺩ ﺗﺎ ﺑﺮﺍی ﺧﺮﯾﺪﻫﺎی ﺁﺗﯽ ﻣﻮﺭﺩ ﺍﺳﺘﻔﺎﺩﻩ ﻗﺮﺍﺭ‬
‫ﺩﺭﺻﺪ‬ ‫ﺯﻣﺎﻥ‬ ‫ﺗﻮﺍﻧﺎﺋﯽ‬ ‫ﺍوﺯﺍﻥ‬ ‫ﺍوﺯﺍﻥ‬ ‫ﮔﯿﺮﺩ‪ .‬ﺩﺭ ﺷﮑﻞ )‪ (6‬ﻧﺤﻮۀ ﺍﺟﺮﺍی ﺭوﺵ ﭘﯿﺸﻨﻬﺎﺩی ﺑﻪ ﺗﺼﻮﯾﺮ‬
‫ﮐﯿﻔﯿﺖ‬
‫ﻣﻌﯿﻮﺏ‬ ‫ﻣﺮﺍﺟﻌﻪ‬ ‫ﻓﻨﯽ‬ ‫ﻣﺤﻠﯽ‬ ‫ﻧﻬﺎﺋﯽ‬
‫ﮐﺸﯿﺪﻩ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬
‫ﺩﺭﺻﺪ‬ ‫‪0.46‬‬ ‫‪0.13‬‬
‫‪1‬‬ ‫‪1.5‬‬ ‫‪2.0‬‬
‫ﻣﻌﯿﻮﺏ‬ ‫‪3‬‬ ‫‪4‬‬
‫ﺯﻣﺎﻥ‬ ‫‪0.29‬‬ ‫‪0.08‬‬ ‫ﯾﮏ ﻣﺜﺎﻝ‬
‫‪1‬‬ ‫‪1.2‬‬
‫ﻣﺮﺍﺟﻌﻪ‬ ‫‪8‬‬ ‫‪7‬‬
‫ﺩﺭ ﺍﯾﻦ ﺑﺨﺶ ﺑﻤﻨﻈﻮﺭ ﺗﺸﺮﯾﺢ ﻣﺪﻝ ﺍﺭﺍﺋﻪ ﺷﺪﻩ‪ ،‬ﯾﮏ ﻣﺜﺎﻝ‬
‫ﺗﻮﺍﻧﺎﺋﯽ‬ ‫‪0.24‬‬ ‫‪0.07‬‬
‫‪1‬‬
‫ﻓﻨﯽ‬ ‫‪0‬‬ ‫‪0‬‬ ‫ﺑﯿﺎﻥ ﻣﯽﮐﻨﯿﻢ‪ .‬ﺳﺎﺧﺘﺎﺭ ‪ Case-Base‬ﻣﻮﺭﺩ ﺍﺳﺘﻔﺎﺩﻩ ﺩﺭ ﺍﯾﻦ‬
‫‪ = 0.00‬ﻧﺮﺥ ﻧﺎﭘﺎﯾﺪﺍﺭی‬ ‫ﻣﺜﺎﻝ ﻣﻄﺎﺑﻖ ﺳﺎﺧﺘﺎﺭ ﺟﺪوﻝ )‪ (1‬ﺩﺭ ﻧﻈﺮ ﮔﺮﻓﺘﻪ ﻣﯽﺷﻮﺩ‪.‬‬
‫ﺗﺼﻮﺭ ﮐﻨﯿﺪ ﺩﺭ ﺍﻧﺒﺎﺭۀ ﺧﺮﯾﺪ‪ ،‬ﺍﻃﻼﻋﺎﺕ ﻣﺮﺑﻮﻁ ﺑﻪ ‪10‬‬ ‫ﺑﺮﺍی ﺗﻌﯿﯿﻦ ﻣﯿﺰﺍﻥ ﺍﻫﻤﯿﺖ ﻫﺮ ﮐﺪﺍﻡ ﺍﺯ ﻣﻌﯿﺎﺭﻫﺎ ﺩﺭ ﺍﻧﺘﺨﺎﺏ‬
‫ﺧﺮﯾﺪ ﺛﺒﺖ و ﺑﻪ ﺷﺮﺡ ﺟﺪوﻝ )‪ (4‬ﻧﮕﻬﺪﺍﺭی ﻣﯽﺷﻮﺩ‪ .‬ﺑﺪﯾﻬﯽ‬ ‫ﻓﺮوﺷﻨﺪﻩ ﺍﺯ ﺭوﺵ ‪ AHP‬ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﮐﻨﯿﻢ‪ .‬ﻣﺠﻤﻮﻋﻪ‬
‫ﺍﺳﺖ ﺩﺭ ﯾﮏ ﻣﺜﺎﻝ وﺍﻗﻌﯽ‪ ،‬ﺍﻃﻼﻋﺎﺕ ﻣﺮﺑﻮﻁ ﺑﻪ ﺻﺪﻫﺎ ﺧﺮﯾﺪ‬ ‫ﻣﻘﺎﯾﺴﺎﺕ ﺯوﺟﯽ ﻣﻮﺭﺩ ﻧﯿﺎﺯ ﺍﯾﻦ ﺭوﺵ ﺩﺭ ﺟﺪوﻝ )‪ (3‬ﺁوﺭﺩﻩ‬
‫ﮔﺬﺷﺘﻪ ﻣﯽﺗﻮﺍﻧﺪ ﻧﮕﻬﺪﺍﺭی ﺷﻮﺩ‪ .‬ﻫﻤﭽﻨﯿﻦ ﻓﺮﺽ ﮐﻨﯿﺪ ﺧﺮﯾﺪ‬ ‫ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﺍوﺯﺍﻥ ﻧﻬﺎﺋﯽ ﻣﻌﯿﺎﺭﻫﺎی ﭘﺎﺋﯿﻦﺗﺮﯾﻦ ﺳﻄﺢ ﺳﻠﺴﻠﻪ‬
‫ﺟﺪﯾﺪی ﺑﺎ ﻣﺸﺨﺼﺎﺕ ﻣﻄﺮﺡ ﺷﺪﻩ ﺩﺭ ﺳﻄﺮ ﺁﺧﺮ ﺍﯾﻦ ﺟﺪوﻝ‬ ‫ﻣﺮﺍﺗﺒﯽ )ﺯﯾﺮ ﻣﻌﯿﺎﺭﻫﺎ( ﺩﺭ ﺗﻌﯿﯿﻦ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ‬
‫ﻣﻄﺮﺡ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬ ‫ﻧﯿﺎﺯﻣﻨﺪیﻫﺎی ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺑﺎ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪﻫﺎی ﻗﺒﻠﯽ ﺑﺮ‬
‫ﺣﺎﻝ ﺭوﺵ ﺑﺎﺯﯾﺎﺑﯽ ﭘﯿﺸﻨﻬﺎﺩی ﺭﺍ ﺑﺮ ﺭوی ﺍﻃﻼﻋﺎﺕ ﻣﻮﺟﻮﺩ‬ ‫ﺍﺳﺎﺱ ﺭﺍﺑﻄﮥ )‪ (1‬ﺑﮑﺎﺭ ﮔﺮﻓﺘﻪ ﺧﻮﺍﻫﺪ ﺷﺪ‪.‬‬
‫ﺍﻋﻤﺎﻝ ﻣﯽﮐﻨﯿﻢ‪ .‬ﺩﺭ ﺟﺪوﻝ )‪ (5‬ﻧﺤﻮۀ ﺍﻋﻤﺎﻝ ﺭوﺵ ﭘﯿﺸﻨﻬﺎﺩی‬
‫ﺟﺪوﻝ ‪ : 3‬ﻣﻘﺎﯾﺴﺎﺕ ﺯوﺟﯽ و ﻣﻘﺎﺩﯾﺮ ﻧﻬﺎﺋﯽ ﺍوﺯﺍﻥ ﻣﻌﯿﺎﺭﻫﺎ‪.‬‬
‫ﺑﺮﺍی ﺳﻨﺠﺶ ﻣﯿﺰﺍﻥ ﻣﺸﺎﺑﻬﺖ ﺑﯿﻦ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺷﻤﺎﺭۀ‬
‫‪ 1‬و ﻧﯿﺎﺯﻣﻨﺪیﻫﺎی ﺧﺮﯾﺪ ﺟﺪﯾﺪ‪ ،‬ﺗﻮﺿﯿﺢ ﺩﺍﺩﻩ ﺷﺪﻩ ﺍﺳﺖ‪.‬‬ ‫ﻫﺪف‬ ‫ﻗﯿﻤﺖ‬ ‫ﺗﺤﻮﯾﻞ‬ ‫ﮐﯿﻔﯿﺖ‬ ‫وﺯﻥ ﻣﺤﻠﯽ )ﻧﻬﺎﺋﯽ(‬
‫ﺩﺭ ﺟﺪوﻝ )‪ (6‬ﻧﺘﯿﺠﮥ ﻧﻬﺎﺋﯽ ﺍﻋﻤﺎﻝ ﺭوﺵ ﭘﯿﺸﻨﻬﺎﺩی ﺑﺮ ﺭوی‬ ‫ﻗﯿﻤﺖ‬ ‫‪1‬‬ ‫‪2.5‬‬ ‫‪1.7‬‬ ‫‪0.504‬‬
‫ﮐﻠﯿﮥ ﺧﺮﯾﺪﻫﺎ ﺁوﺭﺩﻩ ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﺍﻃﻼﻋﺎﺕ ﺍﯾﻦ ﺟﺪوﻝ ﻧﺸﺎﻥ‬ ‫ﺗﺤﻮﯾﻞ‬ ‫‪1‬‬ ‫‪0.7‬‬ ‫‪0.205‬‬
‫ﻣﯽﺩﻫﺪ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺷﻤﺎﺭۀ ‪ 7‬ﺷﺒﯿﻪﺗﺮﯾﻦ ﻣﻮﺭﺩ ﺑﻪ‬ ‫ﮐﯿﻔﯿﺖ‬ ‫‪1‬‬ ‫‪0.291‬‬
‫‪ = 0.00‬ﻧﺮﺥ ﻧﺎﭘﺎﯾﺪﺍﺭی‬
‫ﻣﺸﺨﺼﺎﺕ ﺍﻋﻼﻡ ﺷﺪﻩ ﺗﻮﺳﻂ ﺧﺮﯾﺪﺍﺭ ﺑﺮﺍی ﺍﻧﺠﺎﻡ ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬
‫ﺍﺳﺖ و ﺧﺮﯾﺪﻫﺎی ﺷﻤﺎﺭۀ ‪ 3‬و ‪ 5‬ﺑﺼﻮﺭﺕ ﺗﻮﺍﻣﺎﻥ ﺩﺭ ﺭﺗﺒﮥ ﺑﻌﺪ‬ ‫ﻫﺰﯾﻨﮥ‬ ‫ﻫﺰﯾﻨﮥ‬ ‫ﻫﺰﯾﻨﮥ‬ ‫ﺍوﺯﺍﻥ‬ ‫ﺍوﺯﺍﻥ‬
‫ﻗﯿﻤﺖ‬
‫ﻗﺮﺍﺭﺩﺍﺭﻧﺪ‪ .‬ﺭﺗﺒﮥ ‪ 3‬و ‪ 4‬ﺷﺒﺎﻫﺖ ﻧﯿﺰ ﺑﻪ ﺗﺮﺗﯿﺐ ﻣﺘﻌﻠﻖ ﺑﻪ‬ ‫ﺳﻔﺎﺭﺵﺩﻫﯽ‬ ‫ﺧﺮﯾﺪ‬ ‫ﺣﻤﻞ و ﻧﻘﻞ‬ ‫ﻣﺤﻠﯽ‬ ‫ﻧﻬﺎﺋﯽ‬
‫ﺧﺮﯾﺪﻫﺎی ﺷﻤﺎﺭۀ ‪ 8‬و ‪ 6‬ﺍﺳﺖ‪ .‬ﻫﻤﺎﻧﮕﻮﻧﻪ ﮐﻪ ﺍﻃﻼﻋﺎﺕ ﺟﺪوﻝ‬ ‫ﻫﺰﯾﻨﮥ‬
‫‪1‬‬ ‫‪0.3‬‬ ‫‪2.0‬‬ ‫‪0.216‬‬ ‫‪0.109‬‬
‫ﺳﻔﺎﺭﺵﺩﻫﯽ‬
‫)‪ (4‬ﻧﺸﺎﻥ ﻣﯽﺩﻫﺪ ﻫﻤﮥ ﻣﺸﺨﺼﺎﺕ ﺩو ﺧﺮﯾﺪ ‪ 3‬و ‪ 5‬ﻏﯿﺮ ﺍﺯ‬ ‫ﻫﺰﯾﻨﮥ‬
‫ﻣﺸﺨﺼﮥ "ﺍﻧﻌﻄﺎﻑﭘﺬﯾﺮی ﺩﺭ ﺗﺤﻮﯾﻞ" ﺑﺎ ﻫﻢ ﯾﮑﺴﺎﻥ ﻫﺴﺘﻨﺪ‪.‬‬ ‫‪1‬‬ ‫‪7.0‬‬ ‫‪0.682‬‬ ‫‪0.343‬‬
‫ﺧﺮﯾﺪ‬
‫ﺩﺭ ﺣﺎﻟﯿﮑﻪ ﻓﺮوﺷﻨﺪۀ "ﺧﺮﯾﺪ ﺷﻤﺎﺭۀ ‪ "5‬ﺑﻮﺍﺳﻄﮥ ﺩﺍﺷﺘﻦ‬ ‫ﻫﺰﯾﻨﮥ‬
‫‪1‬‬ ‫‪0.103‬‬ ‫‪0.052‬‬
‫ﺣﻤﻞ و ﻧﻘﻞ‬
‫ﺍﻧﻌﻄﺎﻑﭘﺬﯾﺮی "ﺧﻮﺏ"‪ ،‬ﺍﺯ ﻓﺮوﺷﻨﺪۀ "ﺧﺮﯾﺪ ﺷﻤﺎﺭۀ ‪ "3‬ﺑﺎ‬
‫‪ = 0.00‬ﻧﺮﺥ ﻧﺎﭘﺎﯾﺪﺍﺭی‬
‫ﺍﻧﻌﻄﺎﻑﭘﺬﯾﺮی "ﺿﻌﯿﻒ" ﻣﻄﻠﻮﺑﺘﺮ ﺍﺳﺖ‪ ،‬ﻟﯿﮑﻦ ﺑﻮﺍﺳﻄﮥ ﺍﯾﻨﮑﻪ‬
‫ﻣﺸﺎﺑﻬﺖ ﺻﺮﻓﺎً ﺑﺮ ﺍﺳﺎﺱ ﻣﻌﯿﺎﺭ ﻓﺎﺻﻠﻪ ﺍﺳﺖ‪ ،‬ﻟﺬﺍ ﻓﺎﺻﻠﮥ‬
‫ﺍﺭﺯﯾﺎﺑﯽﻫﺎی "ﺧﻮﺏ" و "ﺿﻌﯿﻒ" ﻧﺴﺒﺖ ﺑﻪ "ﻣﺘﻮﺳﻂ"‬
‫ﯾﮑﺴﺎﻥ ﻣﻨﻈﻮﺭ ﻣﯽﺷﻮﺩ‪ .‬ﻫﻤﯿﻦ ﻣﻮﺿﻮﻉ ﺩﺭ ﻣﻮﺭﺩ ﺧﺮﯾــﺪﻫﺎی‬
 ‫‪579‬‬ ‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ ‪.....‬‬
‫ﺟﺪوﻝ ‪ : 4‬ﺍﻃﻼﻋﺎﺕ ﺧﺮﯾﺪﻫﺎی ﻗﺒﻞ و ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺟﺪﯾﺪ‪.‬‬
‫ﻧﺎم‬ ‫ﮐﯿﻔﯿﺖ‬ ‫ﺗﺤﻮﯾﻞ‬ ‫ﻗﯿﻤﺖ‬ ‫ﻣﻌﯿﺎر‬

‫ﻓﺮوﺷﻨﺪه‬ ‫ﺗﻮان ﻓﻨﯽ‬ ‫زﻣﺎن ﻣﺮاﺟﻌﻪ‬ ‫اﻧﻌﻄﺎف ﭘﺬﯾﺮی درﺻﺪ ﻣﻌﯿﻮب‬ ‫ﻣﺪت ﺗﺤﻮﯾﻞ‬ ‫ﻫﺰﯾﻨﮥ ﺣﻤﻞ‬ ‫ﻫﺰﯾﻨﮥ ﺳﻔﺎرش ﻫﺰﯾﻨﮥ ﺧﺮﯾﺪ‬ ‫ﺷﻤﺎرۀ ﺧﺮﯾﺪ‬

‫‪A‬‬ ‫ﺿﻌﯿﻒ‬ ‫]‪[3.5,4.0,5.5,7.0‬‬ ‫‪3.9‬‬ ‫ﻣﺘﻮﺳﻂ‬ ‫]‪[3.0,5.0,5.0,7.0‬‬ ‫]‪[0.30,0.40,0.45,0.50‬‬ ‫‪2.15‬‬ ‫‪245‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪1‬‬

‫‪B‬‬ ‫ﻣﺘﻮﺳﻂ رو ﺑﻪ ﺑﺎﻻ‬ ‫]‪[3.0,4.0,5.0,6.0‬‬ ‫‪3.8‬‬ ‫ﺧﻮب‬ ‫]‪[4.0,5.0,5.0,7.0‬‬ ‫]‪[0.30,0.37,0.45,0.55‬‬ ‫‪1.90‬‬ ‫‪240‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪2‬‬

‫‪D‬‬ ‫ﺧﻮب‬ ‫]‪[2.0,6.0,6.0,7.0‬‬ ‫‪3.9‬‬ ‫ﺿﻌﯿﻒ‬ ‫]‪[2.0,3.0,4.0,5.0‬‬ ‫]‪[0.35,0.38,0.42,0.58‬‬ ‫‪2.10‬‬ ‫‪247‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪3‬‬

‫‪C‬‬ ‫ﺧﯿﻠﯽ ﺿﻌﯿﻒ‬ ‫]‪[3.0,5.0,5.0,7.0‬‬ ‫‪4.0‬‬ ‫ﻣﺘﻮﺳﻂ رو ﺑﻪ ﭘﺎﺋﯿﻦ‬ ‫]‪[4.0,5.0,5.0,6.0‬‬ ‫]‪[0.28,0.35,0.43,0.62‬‬ ‫‪1.95‬‬ ‫‪243‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪4‬‬

‫‪E‬‬ ‫ﺧﻮب‬ ‫]‪[2.0,6.0,6.0,7.0‬‬ ‫‪3.9‬‬ ‫ﺧﻮب‬ ‫]‪[2.0,3.0,4.0,5.0‬‬ ‫]‪[0.35,0.38,0.42,0.58‬‬ ‫‪2.10‬‬ ‫‪247‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪5‬‬

‫‪G‬‬ ‫ﻣﺘﻮﺳﻂ‬ ‫]‪[1.0,2.0,4.0,6.0‬‬ ‫‪4.2‬‬ ‫ﻣﺘﻮﺳﻂ رو ﺑﻪ ﺑﺎﻻ‬ ‫]‪[1.0,3.0,5.0,7.0‬‬ ‫]‪[0.41,0.45,0.48,0.57‬‬ ‫‪1.97‬‬ ‫‪242‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪6‬‬

‫‪F‬‬ ‫ﺧﻮب‬ ‫]‪[2.0,5.0,6.0,7.0‬‬ ‫‪3.8‬‬ ‫ﻋﺎﻟﯽ‬ ‫]‪[2.0,3.0,4.0,5.0‬‬ ‫]‪[0.24,0.39,0.42,0.58‬‬ ‫‪1.99‬‬ ‫‪250‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪7‬‬

‫‪H‬‬ ‫ﺧﻮب‬ ‫]‪[2.0,6.0,6.0,7.0‬‬ ‫‪3.9‬‬ ‫ﻋﺎﻟﯽ‬ ‫]‪[2.0,3.0,4.0,5.0‬‬ ‫]‪[0.35,0.38,0.42,0.58‬‬ ‫‪2.10‬‬ ‫‪247‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪8‬‬

‫‪I‬‬ ‫ﺧﯿﻠﯽ ﺿﻌﯿﻒ‬ ‫]‪[2.0,4.0,4.0,6.0‬‬ ‫‪3.6‬‬ ‫ﺧﻮب‬ ‫]‪[3.0,5.0,7.0,9.0‬‬ ‫]‪[0.27,0.35,0.35,0.43‬‬ ‫‪1.98‬‬ ‫‪247‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪9‬‬

‫‪J‬‬ ‫ﻣﺘﻮﺳﻂ رو ﺑﻪ ﭘﺎﺋﯿﻦ‬ ‫]‪[1.0,3.0,3.0,4.0‬‬ ‫‪3.8‬‬ ‫ﻋﺎﻟﯽ‬ ‫]‪[5.0,6.0,6.0,7.0‬‬ ‫]‪[0.30,0.34,0.38,0.42‬‬ ‫‪1.98‬‬ ‫‪252‬‬ ‫ﺧﺮﯾﺪ ﺷﻤﺎرۀ ‪10‬‬

‫?‬ ‫ﺧﻮب‬ ‫]‪[1.0,3.0,4.0,6.0‬‬ ‫‪4.0‬‬ ‫ﻣﺘﻮﺳﻂ‬ ‫]‪[2.0,3.0,4.0,5.0‬‬ ‫]‪[0.35,0.40,0.45,0.50‬‬ ‫‪1.84‬‬ ‫‪250‬‬ ‫ﻣﺸﺨﺼﺎت ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬

‫ﺗﻌﯿﯿﻦ‬ ‫ﺟﺮﯾــﺎن اﻃﻼﻋــﺎت‬

‫ﻣﺸﺨﺼﺎﺕ‬ ‫ﺟﺮﯾــﺎن ﻓﺮآﯾﻨــﺪ‬

‫ﭘﯿﺸﻨﻬﺎﺩ ﺩﻫﻨﺪﻩ ﻫﺎی ﺟﺪﯾﺪ‬
‫)ﻣﻌﯿﺎﺭﻫﺎی ﺧﺮﯾﺪ(‬

‫ﺗﻌﯿﯿﻦ ﺳﺎﺧﺘﺎﺭ ﺫﺧﯿﺮۀ ﺧﺮﯾﺪﻫﺎ‬

‫ﻣﺸﺨﺺ ﮐﺮﺩﻥ وﺯﻥ‬

‫ﻣﻌﯿﺎﺭﻫﺎ‬
‫وﺍﺭﺩ ﮐﺮﺩﻥ‬
‫ﺍﻃﻼﻋﺎﺕ‬
‫ﻣﺸﺨﺼﺎﺕ ‪/‬ﻧﯿﺎﺯﻣﻨﺪی ﻫﺎی‬
‫ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬

‫ﻃﺮﺍﺣﯽ ﺭوﺵ‬
‫ﺑﺎﺯﯾﺎﺑﯽ‬

‫ﻣﺨﺰﻥ‬
‫ﺍﻃﻼﻋﺎﺕ ﺧﺮﯾﺪﻫﺎ‬

‫ﺑﺎﺯﯾﺎﺑﯽ ﺧﺮﯾﺪﻫﺎی‬
‫ﻣﺸﺎﺑﻪ‬
‫ﻧﮕﻬﺪﺍﺭی ﺍﻃﻼﻋﺎﺕ‬
‫ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬

‫ﺍوﻟﻮﯾﺖ ﺑﻨﺪی‬
‫ﻓﻬﺮﺳﺖ ﮐﺮﺩﻥ ﺧﺮﯾﺪﻫﺎی‬ ‫ﺗﻄﺒﯿﻖ ﻓﺮوﺷﻨﺪﻩ‬
‫ﺧﺮﯾﺪﻫﺎی ﺑﺎﺯﯾﺎﺑﯽ‬
‫ﻣﺸﺎﺑﻪ ﺑﻪ ﺗﺮﺗﯿﺐ ﻣﯿﺰﺍﻥ‬ ‫ﺑﺎ ﺷﺮﺍﯾﻂ ﺧﺮﯾﺪ‬ ‫ﭘﺎﺳﺦ‬
‫ﻣﺸﺎﺑﻬﺖ‬ ‫ﺷﺪﻩ ﺑﺎ ﺭوﺵ‬
‫‪Fuzzy TOPSIS‬‬ ‫ﺟﺪﯾﺪ‬

‫ﺷﮑﻞ ‪ : 5‬ﻧﻤﻮﺩﺍﺭ ﺟﺮﯾﺎﻥ ﺭوﺵ ﭘﯿﺸﻨﻬﺎﺩی‪.‬‬

‫ﺑﺮﺧﻮﺭﺩﺍﺭ ﺍﺳﺖ‪.‬‬ ‫ﺷﻤﺎﺭۀ ‪ 5‬و ‪ 8‬ﺑﻪ ﻧﻮﻉ ﺩﯾﮕﺮی ﺗﮑﺮﺍﺭ ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﺩﺭ ﺍﯾﻦ‬
‫ﺑﻤﻨﻈﻮﺭ ﺗﺼﺤﯿﺢ ﺍﯾﻦ ﻣﺸﮑﻞ و ﺍﺭﺯﯾﺎﺑﯽ ﺩﻗﯿﻖﺗﺮ‬ ‫ﺧﺮﯾﺪﻫﺎ ﻧﯿﺰ ﻫﻤﮥ ﻣﻘﺎﺩﯾﺮ ﺍﺭﺯﯾﺎﺑﯽ ﻣﺸﺨﺼﻪﻫﺎ ﻏﯿﺮ ﺍﺯ ﻣﺸﺨﺼﮥ‬
‫ﻓﺮوﺷﻨﺪﮔﺎﻥ ﻣﺮﺗﺒﻂ ﺑﺎ ﺧﺮﯾﺪﻫﺎی ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ‪ ،‬ﺭوﺵ‬ ‫"ﺍﻧﻌﻄﺎﻑﭘﺬﯾﺮی" ﺑﺎ ﻫﻢ ﺑﺮﺍﺑﺮ ﺍﺳﺖ‪ .‬ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﻣﻘﺒﻮﻟﯿﺖ‬
‫‪ TOPSIS‬ﺭﺍ ﺑﺮﺍی ‪ 5‬ﺧﺮﯾﺪ ﺑﺎﺯﯾﺎﺑﯽ ﺷﺪﻩ ﺑﺎ ﺩﺭﺟﮥ ﻣﺸﺎﺑﻬﺖ‬ ‫وﺿﻌﯿﺖ ﺍﻧﻌﻄﺎﻑﭘﺬﯾﺮی "ﻋﺎﻟﯽ" ﻧﺴﺒﺖ ﺑﻪ وﺿﻌﯿﺖ‬
‫ﺑﺎﻻ ﺍﺟﺮﺍ ﻣﯽﮐﻨﯿﻢ‪ .‬ﺑﺪﯾﻬﯽ ﺍﺳﺖ ﻣﻌﯿﺎﺭﻫﺎی "ﺍﻧﻌﻄﺎﻑﭘﺬﯾﺮی ﺩﺭ‬ ‫ﺍﻧﻌﻄﺎﻑﭘﺬﯾﺮی "ﺧﻮﺏ"‪ ،‬ﺑﺎﯾﺪ ﺧﺮﯾﺪ ‪ 8‬ﺍوﻟﻮﯾﺖ ﺑﺎﻻﺗﺮی ﮐﺴﺐ‬
‫ﺗﺤﻮﯾﻞ" و ﺗﻮﺍﻧﺎﺋﯽ ﻓﻨﯽ" ﻣﻌﯿﺎﺭﻫﺎی ﻣﺜﺒﺖ و ﻣﺎﺑﻘﯽ ﻣﻌﯿﺎﺭﻫﺎ‪،‬‬ ‫ﻧﻤﺎﯾﺪ وﻟﯽ ﺍﺯ ﺁﻧﺠﺎﺋﯿﮑﻪ ﻓﺎﺻﻠﮥ وﺿﻌﯿﺖ "ﺧﻮﺏ" ﻧﺴﺒﺖ ﺑﻪ‬
‫ﻣﻌﯿﺎﺭﻫﺎی ﻣﻨﻔﯽ ﻣﯽﺑﺎﺷﻨﺪ‪ .‬ﺩﺭ ﺍﯾﻨﺠﺎ ﺍﺯ ﺫﮐﺮ ﺟﺪﺍوﻝ ﻣﯿﺎﻧﯽ‬ ‫وﺿﻌﯿﺖ "ﻋﺎﻟﯽ"‪ ،‬ﺍﺯ وﺿﻌﯿﺖ "ﻣﺘﻮﺳﻂ" ﻣﻮﺭﺩ ﻧﻈﺮ ﺧﺮﯾﺪﺍﺭ‬
‫ﺣﺎﺻﻞ ﺍﺯ ﺍﺟﺮﺍی ﺍﻟﮕﻮﺭﯾﺘﻢ ﺍﺭﺍﺋﻪ ﺷﺪﻩ ﺩﺭ ﺑﺨﺶ ‪ 4-5‬ﺧﻮﺩﺍﺭی‬ ‫ﮐﻤﺘﺮ ﺍﺳﺖ‪ ،‬ﺧﺮﯾﺪ ‪ 5‬ﺍوﻟﻮﯾﺖ ﺑﺎﻻﺗﺮی ﮐﺴﺐ ﻧﻤﻮﺩﻩ ﺍﺳﺖ‪.‬‬
‫ﻧﻤﻮﺩﻩ و ﻣﻨﺤﺼﺮﺍً ﺩﺭ ﺟﺪوﻝ )‪ (7‬ﺑﻪ ﺫﮐﺮ ﻧﺘﯿﺠﮥ ﻧﻬﺎﺋﯽ ﺑﺴﻨﺪﻩ‬ ‫ﺑﺪﯾﻦ ﺗﺮﺗﯿﺐ ﺍﺗﮑﺎء ﺻﺮﻑ ﺑﻪ ﺷﺎﺧﺺ ﻣﺸﺎﺑﻬﺖ ﺣﺎﺻﻠﻪ ﺍﺯ ﺗﻮﺍﺑﻊ‬
‫ﻧﻤﻮﺩﻩﺍﯾﻢ‪:‬‬ ‫ﻓﺎﺻﻠﻪﺍی ﺑﺮﺍی ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺍﺯ ﺩﻗﺖ و ﮐﺎﺭﺍﺋﯽ ﭘﺎﺋﯿﻨﯽ‬
 ‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪1385‬‬ ‫‪580‬‬
‫ﺟﺪوﻝ ‪ : 5‬ﻣﺤﺎﺳﺒﮥ ﻣﯿﺰﺍﻥ ﺷﺒﺎﻫﺖ ﺑﯿﻦ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ‪ 1‬و ﻧﯿﺎﺯﻣﻨﺪیﻫﺎی ﺧﺮﯾﺪ ﺟﺪﯾﺪ‪.‬‬
‫ﻣﺸﺎﺑﻬﺖ‬ ‫ﻣﺸﺎﺑﻬﺖ‬ ‫ﻣﻘﺪار ﻣﺸﺨﺼﻪ‬ ‫ﻣﻘﺪار ﻣﺸﺨﺼﻪ‬ ‫وزن‬
‫ﻧﺎم ﻣﺸﺨﺼﻪ )ﻣﻌﯿﺎر(‬
‫ﻧﻬﺎﺋﯽ )ﮐﻠﯽ(‬ ‫اﻧﻔﺮادی )ﺗﮑﯽ(‬ ‫در ﺧﺮﯾﺪ ‪1‬‬ ‫در ﺧﺮﯾﺪ ﺟﺪﯾﺪ‬ ‫ﻣﺸﺨﺼﻪ‬
‫‪0.583 1‬‬ ‫‪250‬‬ ‫‪245‬‬ ‫‪0.109‬‬ ‫ﻫﺰﯾﻨﮥ ﺳﻔﺎﺭﺵ ﺩﻫﯽ‬
‫‪0.244‬‬ ‫‪1.84‬‬ ‫‪2.15‬‬ ‫‪0.343‬‬ ‫ﻫﺰﯾﻨﮥ ﺧﺮﯾﺪ‬ ‫ﻗﯿﻤﺖ‬
‫‪2‬‬
‫‪0.992‬‬ ‫]‪[0.35,0.40,0.45,0.50‬‬ ‫]‪[0.30,0.40,0.45,0.50‬‬ ‫‪0.052‬‬ ‫ﻫﺰﯾﻨﮥ ﺍﻧﺘﻘﺎﻝ‬
‫‪0.400‬‬ ‫]‪[2.0,3.0,4.0,5.0‬‬ ‫]‪[3.0,5.0,5.0,7.0‬‬ ‫‪0.146‬‬ ‫ﻣﺪﺕ ﺯﻣﺎﻥ ﺗﺤﻮﯾﻞ‬
‫‪0.482 3‬‬ ‫ﺗﺤﻮﯾﻞ‬
‫‪1.000‬‬ ‫ﻣﺘﻮﺳﻂ‬ ‫ﻣﺘﻮﺳﻂ‬ ‫‪0.059‬‬ ‫ﺍﻧﻌﻄﺎﻑ ﭘﺬﯾﺮی‬
‫‪0.875‬‬ ‫‪4‬‬ ‫‪3.9‬‬ ‫‪0.134‬‬ ‫ﺩﺭﺻﺪ ﮐﺎﻻی ﻣﻌﯿﻮﺏ‬
‫‪0.444‬‬ ‫]‪[1.0,3.0,4.0,6.0‬‬ ‫]‪[3.5,4.0,5.5,7.0‬‬ ‫‪0.087‬‬ ‫ﺯﻣﺎﻥ ﻣﺮﺍﺟﻌﻪ‬ ‫ﮐﯿﻔﯿﺖ‬
‫‪0.143‬‬ ‫ﺧﻮﺏ‬ ‫ﺿﻌﯿﻒ‬ ‫‪0.070‬‬ ‫ﺗﻮﺍﻧﺎﺋﯽ ﻓﻨﯽ‬

‫| ‪| 250 − 245‬‬ ‫ﯾﺎدداﺷﺘﻬﺎی ﻣﺤﺎﺳﺒﺎﺗﯽ‪:‬‬

‫‪1- sim (245,250) = 1 −‬‬ ‫)‪= 0.583 (α = 240 and β = 252‬‬
‫‪252 − 240‬‬
‫‪2-‬‬
‫~ ~‬
‫[‬ ‫~‬
‫]‬
‫~‬
‫‪sim ( fI3 , fR 1 3 ) = 1+ | P( fI3 ) − P( fR 1 3 ) | −1= 0.992‬‬
‫~‬ ‫‪0.30 + (2 × 0.40) + (2 × 0.45) + 0.50‬‬ ‫~‬ ‫‪0.35 + (2 × 0.40) + (2 × 0.45) + 0.50‬‬
‫= ) ‪P( fI3‬‬ ‫; ‪= 0.416‬‬ ‫= ) ‪P( fR 1 3‬‬ ‫‪= 0.425‬‬
‫‪6‬‬ ‫‪6‬‬
‫‪ = (0.583 X 0.109) + (0.244 X 0.343) + . . . + (0.444 X 0.087) + (0.143 X 0.070) = 0.482‬ﻣﺸﺎﺑﻬﺖ ﻧﻬﺎﺋﯽ ‪3-‬‬

‫ﺟﺪوﻝ ‪ : 6‬ﻣﯿﺰﺍﻥ ﺷﺒﺎﻫﺖ ﺧﺮﯾﺪﻫﺎی ﻣﻮﺟﻮﺩ ﺩﺭ ﺍﻧﺒﺎﺭۀ ﺧﺮﯾﺪ ﺑﺎ ﻧﯿﺎﺯﻣﻨﺪی ﻫﺎی ﺧﺮﯾﺪ ﺟﺪﯾﺪ‪.‬‬
‫‪J/10‬‬ ‫‪I/9‬‬ ‫‪H/8‬‬ ‫‪G/7‬‬ ‫‪F/6‬‬ ‫‪E/5‬‬ ‫‪D/4‬‬ ‫‪C/3‬‬ ‫‪B/2‬‬ ‫‪A/1‬‬ ‫ﺷﻤﺎرۀ ﺧﺮﯾﺪ‪/‬ﻓﺮوﺷﻨﺪه‬
‫‪0.585‬‬ ‫‪0.535‬‬ ‫‪0.633‬‬ ‫‪0.737‬‬ ‫‪0.625‬‬ ‫‪0.637‬‬ ‫‪0.608‬‬ ‫‪0.637‬‬ ‫‪0.605‬‬ ‫‪0.482‬‬ ‫ﻣﯿﺰان ﻣﺸﺎﺑﻬﺖ‬
‫‪8‬‬ ‫‪9‬‬ ‫‪4‬‬ ‫‪1‬‬ ‫‪5‬‬ ‫‪2‬‬ ‫‪6‬‬ ‫‪2‬‬ ‫‪7‬‬ ‫‪10‬‬ ‫اوﻟﻮﯾﺖ‬

‫ﻧﺘﯿﺠﻪ ﮔﯿﺮی‬ ‫ﺟﺪوﻝ ‪ : 7‬ﻧﺘﺎﯾﺞ ﻧﻬﺎﺋﯽ ﺍﺟﺮﺍی ﺍﻋﻤﺎﻝ ﺭوﺵ ‪ TOPSIS‬ﻓﺎﺯی‪.‬‬

‫ﺷﻤﺎرۀ‬ ‫اوﻟﻮﯾﺖ‬
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺑﻌﻨﻮﺍﻥ ﯾﮑﯽ ﺍﺯ ﺩﻏﺪﻏﻪﻫﺎی ﻗﺎﺑﻞ ﺗﺎﻣﻞ‬ ‫‪Di+‬‬ ‫‪Di−‬‬ ‫‪Ci‬‬
‫ﺧﺮﯾﺪ‪/‬ﻓﺮوﺷﻨﺪه‬ ‫ﻧﻬﺎﺋﯽ‬
‫ﻣﺪﯾﺮﺍﻥ ﭘﺸﺘﯿﺒﺎﻧﯽ ﺷﺮﮐﺘﻬﺎ ﻣﻄﺮﺡ ﺑﻮﺩﻩ و ﺍﺳﺖ‪ .‬ﺩﺭ ﻣﺤﯿﻄﯽ‬ ‫‪G/7‬‬ ‫‪0.16‬‬ ‫‪0.21‬‬ ‫‪0.42‬‬ ‫‪2‬‬
‫ﮐﻪ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﺑﻪ ﮐﺎﺭی ﺭوﺯﻣﺮﻩ و ﺑﺎ ﺩﺭﺟﮥ ﺗﮑﺮﺍﺭ ﺑﺎﻻ‬ ‫‪C/3‬‬ ‫‪0.09‬‬ ‫‪0.27‬‬ ‫‪0.24‬‬ ‫‪5‬‬
‫‪E/5‬‬ ‫‪0.12‬‬ ‫‪0.24‬‬ ‫‪0.34‬‬ ‫‪4‬‬
‫ﺗﺒﺪﯾﻞ ﺷﺪﻩ ﺑﺎﺷﺪ‪ ،‬ﻫﻤﻮﺍﺭﻩ ﺭﯾﺴﮏ ﻧﺎﺷﯽ ﺍﺯ ﺍﺗﺨﺎﺫ ﺗﺼﻤﯿﻢ‬ ‫‪H/7‬‬ ‫‪0.13‬‬ ‫‪0.23‬‬ ‫‪0.36‬‬ ‫‪3‬‬
‫ﺍﺷﺘﺒﺎﻩ وﺟﻮﺩ ﺩﺍﺭﺩ‪ .‬ﺩﺭ ﭼﻨﯿﻦ ﺷﺮﺍﯾﻄﯽ ﮐﻪ ﺑﻮﺍﺳﻄﮥ ﺗﮑﺮﺍﺭی‬ ‫‪F/6‬‬ ‫‪0.18‬‬ ‫‪0.24‬‬ ‫‪0.43‬‬ ‫‪1‬‬
‫ﺑﻮﺩﻥ ﺗﺼﻤﯿﻢ‪ ،‬ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺩﺭﺳﻬﺎی ﺁﻣﻮﺧﺘﻪ ﺷﺪﻩ ﺍﺯ ﺗﺠﺮﺑﯿﺎﺕ‬
‫ﻗﺒﻠﯽ ﺑﻄﻮﺭ ﻃﺒﯿﻌﯽ ﻣﺪ ﻧﻈﺮ ﺗﺼﻤﯿﻢﮔﯿﺮﻧﺪﻩ ﺍﺳﺖ؛ ﻃﺮﺍﺣﯽ و‬ ‫ﺑﻪ ﺍﯾﻦ ﺗﺮﺗﯿﺐ ﺑﺮ ﺧﻼﻑ ﻧﺘﺎﯾﺞ ﺍوﻟﯿﻪ‪ ،‬ﻓﺮوﺷﻨﺪۀ ‪ F‬ﺑﻮﺍﺳﻄﮥ‬
‫ﭘﯿﺎﺩﻩﺳﺎﺯی ﺳﯿﺴﺘﻤﯽ ﻫﻤﭽﻮﻥ ﺭوﺵ ‪ CBR‬ﮐﻪ ﺑﺘﻮﺍﻧﺪ ﺍﯾﻦ‬ ‫ﺩﺍﺷﺘﻦ ﺷﺎﺧﺺ ﺍﺭﺟﺤﯿﺖ ﺑﺎﻻﺗﺮ ﺣﺎﺋﺰ ﺍوﻟﻮﯾﺖ ﺍوﻝ ﮔﺮﺩﯾﺪﻩ و‬
‫ﻧﯿﺎﺯ ﺭﺍ ﺑﻄﻮﺭ ﻧﻈﺎﻡﻣﻨﺪ ﭘﺎﺳﺦ ﮔﻮﯾﺪ‪ ،‬ﺑﻪ ﺍﺗﺨﺎﺫ ﺗﺼﻤﯿﻤﺎﺕ‬ ‫ﻓﺮوﺷﻨﺪﻩﻫﺎی ‪ E ،H ،G‬و ‪ C‬ﺑﻪ ﺗﺮﺗﯿﺐ ﺩﺭ ﺭﺗﺒﻪﻫﺎی ﺑﻌﺪ ﻗﺮﺍﺭ‬
‫ﺳﺮﯾﻊﺗﺮ و ﺻﺤﯿﺢﺗﺮ ﮐﻤﮏ ﺷﺎﯾﺎﻧﯽ ﺧﻮﺍﻫﺪ ﮐﺮﺩ‪.‬‬ ‫ﻣﯽﮔﯿﺮﻧﺪ و ﻟﺬﺍ ﻓﺮوﺷﻨﺪۀ ‪ F‬ﺑﺮﺍی ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﭘﯿﺸﻨﻬﺎﺩ‬
‫ﺍﺯ ﻃﺮﻑ ﺩﯾﮕﺮ‪ ،‬ﺍﻃﻼﻋﺎﺗﯽ ﮐﻪ ﺑﺮﺍی ﺍﺗﺨﺎﺫ ﺗﺼﻤﯿﻢ ﻣﻮﺭﺩ‬ ‫ﻣﯽﺷﻮﺩ‪ .‬ﺩﺭ ﺻﻮﺭﺗﯿﮑﻪ ﻣﺮﺣﻠﮥ ﺗﻄﺒﯿﻖ ﺷﺮﺍﯾﻂ ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺑﺎ‬
‫ﻧﯿﺎﺯ ﻫﺴﺘﻨﺪ ﻫﻤﻮﺍﺭﻩ ﺑﻄﻮﺭ ﻣﺘﻘﻦ و ﻗﻄﻌﯽ ﺩﺭ ﺍﺧﺘﯿﺎﺭ‬ ‫ﻣﺸﺨﺼﺎﺕ ﻓﺮوﺵ ﺍﯾﻦ ﻓﺮوﺷﻨﺪﻩ ﺑﺎ ﻣﻮﻓﻘﯿﺖ ﺍﻧﺠﺎﻡ ﮔﯿﺮﺩ‪ ،‬ﺍﯾﻦ‬
‫ﺗﺼﻤﯿﻢﮔﯿﺮﻧﺪﻩ ﻗﺮﺍﺭ ﻧﻤﯽﮔﯿﺮﻧﺪ و ﺩﺭ ﺑﺴﯿﺎﺭی ﺍﺯ ﺷﺮﺍﯾﻂ وﺍﻗﻌﯽ‪،‬‬ ‫ﻓﺮوﺷﻨﺪﻩ ﺑﺮﺍی ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﺍﻧﺘﺨﺎﺏ ﻣﯽﮔﺮﺩﺩ؛ ﺩﺭ ﻏﯿﺮ‬
‫ﺑﺎ ﻋﺒﺎﺭﺍﺕ و ﺗﺨﻤﯿﻦﻫﺎی ﻧﺎﺩﻗﯿﻖ ﺗﻬﯿﻪ ﻣﯽﮔﺮﺩﻧﺪ‪ .‬ﯾﮑﯽ ﺍﺯ‬ ‫ﺍﯾﻨﺼﻮﺭﺕ ﺑﺎﯾﺪ ﻓﺮوﺷﻨﺪﻩﻫﺎی ﺩﯾﮕﺮ )ﺑﻪ ﺗﺮﺗﯿﺐ ﺷﺎﺧﺺ‬
‫ﺍﺑﺰﺍﺭﻫﺎی ﻣﻌﻤﻮﻝ ﺑﺮﺍی ﺗﺒﯿﯿﻦ ﺍﯾﻦ ﻣﻮﻗﻌﯿﺖﻫﺎ‪ ،‬ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ‬ ‫ﺍﺭﺟﺤﯿﺘﯽ ﮐﻪ ﻣﺤﺎﺳﺒﻪ ﺷﺪﻩ ﺍﺳﺖ‪ ،‬ﯾﻌﻨﯽ ‪ H ،G‬و ‪(. . .‬‬
‫ﻣﻔﺎﻫﯿﻢ ﻣﻨﻄﻖ ﻓﺎﺯی ﺍﺳﺖ و ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ ﺑﺮﺍی ﭘﻮﺷﺶ‬ ‫ﺍﻣﺘﺤﺎﻥ ﺷﻮﻧﺪ ﺗﺎ ﺩﺭ ﻧﻬﺎﯾﺖ ﻓﺮوﺷﻨﺪۀ ﻣﻨﺎﺳﺐ ﺍﻧﺘﺨﺎﺏ ﺷﻮﺩ‪.‬‬
‫ﺩﺍﺩﻥ ﭼﻨﯿﻦ ﺷﺮﺍﯾﻄﯽ‪ ،‬ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺍﻋﺪﺍﺩ ﻓﺎﺯی ﭘﯿﺸﻨﻬﺎﺩ ﺷﺪﻩ‬ ‫ﺑﻬﺮ ﺣﺎﻝ ﺩﺭ ﭘﺎﯾﺎﻥ ﺑﺎﯾﺪ ﻣﺸﺨﺼﺎﺕ ﺧﺮﯾﺪ ﺟﺪﯾﺪ ﻫﻤﺮﺍﻩ ﺑﺎ‬
‫و ﺩﺭ وﺍﻗﻊ ﻧﺤﻮۀ ﭘﯿﺎﺩﻩﺳﺎﺯی ‪ CBR‬ﻓﺎﺯی ﺩﺭ ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ‬ ‫ﻓﺮوﺷﻨﺪۀ ﻣﻨﺘﺨﺐ ﺑﺮﺍی ﺁﻥ‪ ،‬ﺑﻪ ﭘﺎﯾﮕﺎﻩ ﺍﻃﻼﻋﺎﺗﯽ ﺧﺮﯾﺪﻫﺎ‬
‫ﺗﺸﺮﯾﺢ ﺷﺪﻩ ﺍﺳﺖ‪ .‬ﺩﺭ ﺍﺩﺍﻣﻪ‪ ،‬ﻧﻘﺎﻁ ﺿﻌﻒ ﺭوﺷﻬﺎی ﺑﺎﺯﯾﺎﺑﯽ‬ ‫ﺍﻓﺰوﺩﻩ ﮔﺮﺩﺩ‪.‬‬
 581 ..... ‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ‬

.‫ﻗﺮﺍﺭ ﮔﯿﺮﺩ‬
‫– ﻣﺪﻝ ﺍﺭﺍﺋﻪ ﺷﺪﻩ ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ ﺗﻮﺟﻪ ﺍﺻﻠﯽ ﺧﻮﺩ ﺭﺍ ﺑﻪ‬3
‫ﺍﻧﺘﺨﺎﺏ ﻓﺮوﺷﻨﺪﻩ ﻣﻌﻄﻮﻑ ﻧﻤﻮﺩﻩ و ﺗﺨﺼﯿﺺ ﺳﻔﺎﺭﺵ ﺭﺍ ﻣﺪ‬
‫ ﺑﺎ ﺗﻠﻔﯿﻖ ﻧﺘﺎﯾﺞ ﺣﺎﺻﻠﻪ ﺩﺭ ﻣﺪﻟﻬﺎی‬.‫ﻧﻈﺮ ﻗﺮﺍﺭ ﻧﺪﺍﺩﻩ ﺍﺳﺖ‬
‫ ﺍﯾﻦ ﻧﻘﺺ ﺭﺍ ﻣﯽﺗﻮﺍﻥ‬،‫ ﻫﻤﭽﻮﻥ ﻣﺪﻟﻬﺎی ﺭﯾﺎﺿﯽ‬،‫ﮐﻤﮑﯽ‬
.‫ﺑﺮﻃﺮﻑ ﻧﻤﻮﺩ‬
‫ ﺗﻮﺳﻌﮥ ﻣﺮﺣﻠﮥ ﺑﺎﺯﯾﺎﺑﯽ ﺭوﺵ‬،‫– ﺗﺎﮐﯿﺪ ﺍﺻﻠﯽ ﺩﺭ ﺍﯾﻦ ﺗﺤﻘﯿﻖ‬4
.‫ ﺑﻮﺩ و ﺳﺎﯾﺮ ﻣﺮﺍﺣﻞ ﺍﯾﻦ ﺭوﺵ ﺑﻪ ﺍﺟﻤﺎﻝ ﺗﺸﺮﯾﺢ ﺷﺪﻧﺪ‬CBR
‫ﺗﻮﺻﯿﻪ ﻣﯽﺷﻮﺩ ﺩﺭ ﭘﮋوﻫﺶﻫﺎی ﺁﺗﯽ ﻣﺮﺍﺣﻞ ﺩﯾﮕﺮ ﻣﺪﻝ‬
‫ ﺍﺯ ﺟﻤﻠﻪ ﻣﺮﺣﻠﮥ ﺗﻄﺒﯿﻖ ﺧﺮﯾﺪ ﺑﺮﺍی ﻣﻮﺿﻮﻉ ﺗﺤﻘﯿﻖ‬CBR
.‫ﮔﺴﺘﺮﺵ ﯾﺎﺑﺪ‬
‫ ﻧﯿﺰ ﻫﻤﭽﻮﻥ ﺭوﺷﻬﺎی‬CBR ‫ – ﻧﻈﺮ ﺑﻪ ﺍﯾﻨﮑﻪ ﺭوﺵ‬5
،‫ﺩﺍﺩﻩﮐﺎوی ﻣﺒﺘﻨﯽ ﺑﺮ ﺑﻬﺮﻩﺑﺮﺩﺍﺭی ﺍﺯ ﺗﺠﺮﺑﯿﺎﺕ ﮔﺬﺷﺘﻪ ﺍﺳﺖ‬
‫ﭘﯿﺸﻨﻬﺎﺩ ﻣﯽﺷﻮﺩ ﺩﺭ ﺗﺤﻘﯿﻘﺎﺕ ﺁﺗﯽ ﺯﻣﯿﻨﮥ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ‬
‫ﻗﺎﺑﻠﯿﺘﻬﺎی ﺭوﺷﻬﺎی ﺩﺍﺩﻩﮐﺎوی ﺑﻤﻨﻈﻮﺭ ﺩﺳﺘﻪﺑﻨﺪی ﺭﮐﻮﺭﺩﻫﺎی‬
‫ ﻣﻮﺭﺩ‬،‫ﭘﺎﯾﮕﺎﻩ )ﻣﺨﺰﻥ( ﺧﺮﯾﺪﻫﺎ و ﭘﺎﮐﺴﺎﺯی ﺩﺍﺩﻩﻫﺎی ﻧﺎﻣﻨﺎﺳﺐ‬
.‫ﺑﺮﺭﺳﯽ ﻗﺮﺍﺭ ﮔﯿﺮﺩ‬
‫ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ‬،‫ﻣﺮﺳﻮﻡ ﺑﯿﺎﻥ ﺷﺪﻩ و ﺑﺮﺍی ﺑﻬﺒﻮﺩ ﻧﺘﺎﯾﺞ ﺣﺎﺻﻠﻪ‬
‫ ﺑﺮﺍی ﺍوﻟﻮﯾﺖﺑﻨﺪی ﮔﺰﯾﻨﻪﻫﺎی ﺑﺎﺯﯾﺎﺑﯽﺷﺪﻩ‬TOPSIS ‫ﺭوﺵ‬
‫ ﺑﻪ ﺍﯾﻦ ﺗﺮﺗﯿﺐ ﺩﺭ ﺭوﺵ ﭘﯿﺸﻨﻬﺎﺩی ﻋﻼوﻩ‬.‫ﺗﻮﺻﯿﻪ ﺷﺪﻩ ﺍﺳﺖ‬
‫ﺑﺮ ﺍﯾﻨﮑﻪ ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺗﺠﺮﺑﯿﺎﺕٍ ﺧﺮﯾﺪﻫﺎی ﭘﯿﺸﯿﻦ ﻣﻮﺭﺩ ﺗﺎﮐﯿﺪ‬
‫ ﻣﻄﻠﻮﺑﯿﺖ ﻣﻌﯿﺎﺭﻫﺎی ﺗﺼﻤﯿﻢﮔﯿﺮی ﻧﯿﺰ ﺩﺭ ﺍﻧﺘﺨﺎﺏ‬،‫ﻗﺮﺍﺭ ﮔﺮﻓﺘﻪ‬
.‫ﻓﺮوﺷﻨﺪۀ ﻣﻨﺎﺳﺐ ﺩﺧﺎﻟﺖ ﺩﺍﺩﻩ ﺷﺪﻩ ﺍﺳﺖ‬
‫ﺩﺭ ﭘﺎﯾﺎﻥ ﻣﻮﺿﻮﻋﺎﺕ ﺯﯾﺮ ﺑﻌﻨﻮﺍﻥ ﺯﻣﯿﻨﻪﻫﺎی ﭘﮋوﻫﺸﯽ ﺁﺗﯽ‬
:‫ﻣﻌﺮﻓﯽ ﻣﯽﺷﻮﻧﺪ‬
‫– ﻣﺪﻝ ﭘﯿﺸﻨﻬﺎﺩ ﺷﺪﻩ ﺣﺎﻟﺘﯽ ﺳﺎﺩﻩ ﮐﻪ ﺩﺭ ﺁﻥ ﻣﺸﺨﺼﺎﺕ‬1
،‫ﻫﺮ "ﻣﻮﺭﺩ" ﺧﺮﯾﺪ ﻣﺤﺪوﺩ ﺑﻪ ﭼﻨﺪ وﯾﮋﮔﯽ ﻋﻤﻮﻣﯽ ﺍﺳﺖ‬
‫ﻃﺮﺍﺣﯽ ﺷﺪﻩ ﺍﺳﺖ؛ ﺩﺭ ﺻﻮﺭﺗﯿﮑﻪ ﺩﺭ ﻋﻤﻞ و ﺑﺴﺘﻪ ﺑﻪ ﺷﺮﺍﯾﻂ‬
‫ ﺧﺮﯾﺪﺍﺭ وﯾﮋﮔﯽﻫﺎی ﻣﺘﻌﺪﺩ و ﻣﺘﻨﻮﻋﯽ ﺭﺍ ﺑﺮﺍی‬،‫ﻣﺴﺎﻟﻪ‬
‫ ﻣﺪﻝ ﺗﺤﻘﯿﻖ ﺣﺎﺿﺮ ﺭﺍ ﺑﺮﺍی‬.‫ﻓﺮوﺷﻨﺪﻩ ﺩﺭ ﻧﻈﺮ ﻣﯽﮔﯿﺮﺩ‬
.‫ﭼﻨﯿﻦ ﺣﺎﻻﺕ ﭘﯿﭽﯿﺪﻩﺍی ﻣﯽﺗﻮﺍﻥ ﺗﻮﺳﻌﻪ ﺩﺍﺩ‬
‫– ﺍﺳﺘﻔﺎﺩﻩ ﺍﺯ ﺳﺎﯾﺮ ﻣﺘﺪﻫﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﺗﺼﻤﯿﻢﮔﯿﺮیﻫﺎی‬2
‫ﭼﻨﺪ ﻣﻌﯿﺎﺭﻩ ﺑﺮﺍی ﺗﻘﻮﯾﺖ ﻣﺮﺣﻠﮥ ﺑﺎﺯﯾﺎﺑﯽ و ﺍﻧﺘﺨﺎﺏ ﻧﻬﺎﺋﯽ‬
‫ﻓﺮوﺷﻨﺪﻩ ﻣﯽﺗﻮﺍﻧﺪ ﺑﻪ ﻋﻨﻮﺍﻥ ﺯﻣﯿﻨﮥ ﺗﺤﻘﯿﻘﺎﺗﯽ ﺩﯾﮕﺮی ﻣﺪ ﻧﻈﺮ‬

‫ﻣﺮﺍﺟﻊ‬
1 - Ghobadian, A., Stainer, A. and Kiss, T. (1993). “A computerized vendor rating system.” Proceedings of the
First International Symposium on Logistics, The University of Nottingham, Nottingham, UK, PP. 321-328.
2 - Burton, T.T. (1988). “JIT/Repetitive sourcing strategies: tying the knot with your suppliers.” Production
and Inventory Management Journal, 4th Quarter, PP. 38-41.
3 - Dickson, G. W. (1966). “An analysis of vendor selection systems and decisions.” Journal of Purchasing,
Vol. 2, No. 1, PP. 5-17.
4 - Weber, C. A., Current, J. R. and Benton, W. C. (1991). “Vendor selection criteria and methods.” European
Journal of Operational Research, Vol. 50, PP. 2-18.
5 - Mummalaneni, V., Dubas, K. M. and Chao, C. (1996). “Chinese purchasing managers’ preferences and
trade-offs in supplier selection and performance evaluation.” Industrial Marketing Management, Vol. 25,
No.2, PP. 115-24.
6 - Dahel, N. (2003). “Vendor selection and order quantity allocation in volume discount environment.” Supply
Chain Management: An International Journal, Vol. 8, No. 4, PP. 335-342.
7 - Lehmann, D. and O’Shaughnessy, J. (1982). “Decision criteria used in buying different categories of
products.” Journal of Purchasing and Materials Management, Vol. 18, No. 1, PP. 9-14.
8 - Cooper, S. D. (1977). “A total system for measuring of performance.” Journal of Purchasing and Materials
Management, PP. 22-26.
9 - Mazurak, R. E., Rao, S. R. and Scotton, D. W. (1985). “Spreadsheet software application in purchasing.”
Journal of Purchasing and Materials Management, PP. 8-16.
10 - Timmerman, E. (1986). “An approach to vendor performance evaluation.” Journal of Purchasing and
Materials Management, winter, PP. 2-8, 1986.
 1385 ‫ ﻣﻬﺮ ﻣﺎﻩ‬،4 ‫ ﺷﻤﺎﺭﻩ‬،40 ‫ ﺟﻠﺪ‬،‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‬ 582
11 - Min, H. (1994). “International supplier selection: a multi-attribute utility approach.” International Journal
of Physical Distribution & Logistics Management, Vol. 24, No. 5, PP. 24-33.
12 - Narasimhan, R. (1983). “An analytic approach to supplier selection.” Journal of Purchasing and Materials
Management, winter, PP. 27-32.
13 - Barbarosoglu, G. and Yazgaç, T. (1997). “An application of the analytic hierarchy process to the supplier
selection problem.” Production and Inventory Management Journal, 1st Quarter, PP. 14-21.
14 - Gaballa, A. A. (1974). “Minimum cost allocation of tenders.” Operational Research Quarterly, Vol. 25,
No. 3, PP. 398.
15 - Anthony, T. F. and Buffa, F. P. (1977). “Strategic purchase scheduling.” Journal of Purchasing and
Materials Management, Vol. 13, No. 3, PP. 27-31.
16 - Pan, A. C. (1989). “Allocation of order quantity among suppliers.” Journal of Purchasing and Materials
Management, Vol. 25, No. 3, PP. 36-39.
17 - Buffa, F. P. and Jackson, W. M. (1983). “A goal programming model for purchase planning.” Journal of
Purchasing and Materials Management, Vol. 19, No. 3, PP. 27-34.
18 - Sharma, D., Benton, W. C. and Srivastava, R. (1989). “Competitive strategy and purchasing decision”
Proceedings of the 1989 Annual Conference of the Decision Sciences Institute, PP. 1088-1090.
19 - Weber, C. A. (1996). “A data envelopment analysis approach to measuring vendor performance.” Supply
Chain Management, Vol.1, No.1, PP. 28-39.
20 - Easton, L., Murphy, J. D. and Pearson, J. N. (2002). “Purchasing performance evaluation: with data
envelopment analysis.” European Journal of Purchasing & Supply Management, Vol. 8, PP. 123–134.
21 - Ghodsypour, S. H. and O'Brien, C. (1998). “A decision support system for supplier selection using an
integrated analytic hierarchy process and linear programming.” International Journal of Production
Economics, Vol. 56-57, PP. 199-212.
22 - Morlacchi, P. (1997). “Small and medium enterprises in supply chain: a supplier evaluation model and
some empirical results.” Proceedings IFPMM Summer School, August, Saltsburg.
23 - Erol, I., William, G. and Ferrell, Jr. (2003). “A methodology for selection problems with multiple,
conflicting objectives and both qualitative and quantitative criteria.” International Journal of Production
Economics, Vol. 86, PP. 187-199.
24 - Li, C. C., Fun, Y. P. and Hung, J. S. (1997). “A new measure for supplier performance evaluation.” IIE
Transactions on Operations Engineering, Vol. 29, PP. 753-758.
25 - Kumar, M., Vrat, P. and Shankar, R. (2004). “A fuzzy goal programming approach for vendor selection
problem in a supply chain.” Computers & Industrial Engineering, Vol. 46, PP. 69–85.
26 - Cook, R. L. (1997). “Case-based reasoning systems in purchasing: applications and development.”
International Journal of Purchasing and Materials Management, winter, PP. 32–39.
27 - Choy, K. L. and Lee, W. B. (2003). “A generic supplier management tool for outsourcing manufacturing.”
Supply Chain Management: An International Journal. Vol. 8, No. 2, PP. 140-154.
28 - Choy, K. L. and Lee, W. B. (2001) “Multi-agent based virtual enterprise supply chain network for order
management.” Journal of Industrial Engineering Research, Vol. 2, No. 2, PP. 126-141.
29 - Mclover, R. T. and Humphreys, P. K. (2000). “A case-based reasoning approach to the make or buy
decision.” Integrated Manufacturing Systems, Vol. 11, No. 5, PP. 295-310.
 583 ..... ‫ﺍﺭﺍﺋﻪ ﯾﮏ ﻣﺪﻝ‬
30 - Aamodt, A. and Plaza E. (1994). “Case-based reasoning: foundational issues, methodological variations and
system approaches.” AI Communications, Vol. 7, No. 1, PP. 39-59.
31 - Varma, A. and Roddy, N. (1999). “ICARUS: Design and deployment of a case-based reasoning system for
locomotive diagnostics.” Engineering Application of Artificial Intelligence, Vol. 12, No. 6, PP. 681-690.
32 - Montani, S., Bellazzi, R., Portinale, L.. d’Annunzio, G., Fiocchi, S. and Stefanelli, M. (2000). “Diabetic
patients management exploiting case-based reasoning techniques”, Computer Methods and Program in
Biomedicine, Vol. 62, No. 3, PP. 205-218.
33 - Schmidt, G. (1998). “Case-based reasoning for production scheduling.” International Journal of
Production Economics, Vol. 56-57, PP. 537-546.
34 - Changchien, S. W. and Lin, M. C. (2005). “Design and implementation of a case-based reasoning system
for marketing plans.” Expert Systems with Application, Vol. 28, PP. 43-53.
35 - Kolodner, J. (1993). Cased-based reasoning. Morgan Kaufmann, San Mateo, CA.
36 - Kaufmann, A. and Gupta, M. M. (1985). Introduction to Fuzzy Arithmetic: Theory and Applications. Van
Nostrand Reinhold, New York.
37 - Zimmermann, H. J. (1991). Fuzzy set theory and its applications, Second Edition, Kluwer Academic
Publishers, Boston / Dordrecht / London.
38 - Dubois, D. and Parade, H. (1980). Fuzzy sets and systems: theory and applications, Academic Press Inc.,
New York.
39 - Hsieh, C. H. and Chen, S. H. (1999). “Similarity of generalized fuzzy numbers with graded mean
integration representation.” Proceedings of 8th International Fuzzy Systems Association World Congress.,
Vol. 2, Taipei, Taiwan, Republic of China, PP. 551–555.
40 - Hwang, C. L. and Yoon, K. (1981). Multiple attribute decision making: methods and applications,
Springer-Verlag, New York.
41 - Chen, C. T. (2000). “Extensions of the TOPSIS for group decision making under fuzzy environment.”
Fuzzy Sets and Systems, Vol. 114, PP. 1-9.

‫وﺍژﻩ ﻫﺎی ﺍﻧﮕﻠﯿﺴﯽ ﺑﻪ ﺗﺮﺗﯿﺐ ﺍﺳﺘﻔﺎﺩﻩ ﺩﺭ ﻣﺘﻦ‬

1 - Component Parts 14 - Nearest-Neighbor

2 - Linear Weighting method 15 - Weighted Euclidean Distance
3 - Analytical Hierarchy Process (AHP) 16 - Extension Principle
4 - Mixed Integer Programming (MIP) 17 - Vertex Method
5 - Data Envelopment Analysis (DEA) 18- Graded Mean Integration-Representation Distance
6 - Outsourcing 19 - Separation Measure
7 - Make or Buy decision 20 - Closeness
8 - Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)
9 - Retrieve
10 - Case
11 - Attribute (Feature or Criterion)
12 - Matching Method
13 - Case-Base
‫ﻧﺸﺮﯾﻪ ﺩﺍﻧﺸﮑﺪﻩ ﻓﻨﯽ‪ ،‬ﺟﻠﺪ ‪ ،40‬ﺷﻤﺎﺭﻩ ‪ ،4‬ﻣﻬﺮ ﻣﺎﻩ ‪1385‬‬ ‫‪584‬‬