ارزيابي كارايي مدل تحليل پوششي داده هاي دو مرحله‌اي مبتني بر اعداد فازي شهودي مثلثي و متغيرهاي كمكي

Efficiency Evaluation of Two-Stage Data Envelopment Analysis Model Based on Triangular Intuitionistic Fuzzy numbers andslack variables

ﯾﮑﯽ از ﺑﻬﺘﺮﯾﻦ اﺑﺰارﻫﺎ ﺑﺮاي ارزﯾﺎﺑﯽ ﻋﻤﻠﮑﺮد واﺣﺪﻫﺎي ﺗﺼﻤﯿﻢ ﮔﯿﺮﻧﺪه، ﺗﺤﻠﯿﻞ ﭘﻮﺷﺸﯽ داده ﻫﺎ اﺳﺖ. ﺗﺤﻠﯿﻞ ﭘﻮﺷﺸﯽ داده ﻫﺎي ﺳﻨﺘﯽ در ﺳﻨﺠﺶ واﺣﺪﻫﺎي ﮐﺎرا و ﻧﺎﮐﺎرا و ارزﯾﺎﺑﯽ ﻋﻤﻠﮑﺮد ﺳﯿﺴﺘﻢﻫﺎي ﺷﺒﮑﻪاي ﺑﻪ درﺳﺘﯽ ﻋﻤﻞ ﻧﻤﯽﮐﻨﻨﺪ و ﻣﺪل-ﻫﺎي ﺳﻨﺘﯽ ﺗﺤﻠﯿﻞ ﭘﻮﺷﺸﯽ دادهﻫﺎ ﺑﻪ ﺳﺎﺧﺘﺎرﻫﺎي دروﻧﯽ و ﻣﻘﺎدﯾﺮ ﻣﯿﺎﻧﯽ ﺗﻮﺟﻬﯽ ﻧﺪارﻧﺪ. ﺑﻪ ﻫﻤﯿﻦ ﻋﻠﺖ در ﺳﺎلﻫﺎي اﺧﯿﺮ ﻣﺪلﻫﺎي ﺗﺤﻠﯿﻞ ﭘﻮﺷﺸﯽ دادهﻫﺎ ﻣﻌﺮوف ﺑﻪ ﻣﺪلﻫﺎي ﺗﺤﻠﯿﻞ ﭘﻮﺷﺸﯽ داده ﻫﺎي ﺷﺒﮑﻪاي ﻣﻌﺮﻓﯽ ﺷﺪهاﻧﺪ، اﯾﻦ ﻣﺪلﻫﺎ ﺑﺎ ﻟﺤﺎظ ﻧﻤﻮدن ﻣﻘﺎدﯾﺮ ﻣﯿﺎﻧﯽ، اﯾﻦ ﻧﻘﺺ را ﺑﺮﻃﺮف ﮐﺮدﻧﺪ. در اﯾﻦ ﻣﻘﺎﻟﻪ ﺑﺮاي ﺷﻨﺎﺳﺎﯾﯽ ﮐﺎراﯾﯽ واﺣﺪﻫﺎ از ﺗﺤﻠﯿﻞ ﭘﻮﺷﺸﯽ دادهﻫﺎ ﻣﺒﺘﻨﯽ ﺑﺮ ﻣﺘﻐﯿﺮﻫﺎي ﮐﻤﮑﯽ ﺷﺒﮑﻪاي ﺑﺼﻮرت دو ﻣﺮﺣﻠﻪاي و از دادهﻫﺎي ﻓﺎزي ﺷﻬﻮدي ﻣﺜﻠﺜﯽ، اﺳﺘﻔﺎده ﺷﺪه اﺳﺖ. ﺑﻪ ﻃﻮري ﮐﻪ در اﺑﺘﺪا ﻣﺪل ﺗﺤﻠﯿﻞ ﭘﻮﺷﺸﯽ دادهﻫﺎي دو ﻣﺮﺣﻠﻪاي ﻣﻌﺮﻓﯽ ﻣﯽﮔﺮدد و ﺳﭙﺲ ﻣﺪل ﺑﺮ اﺳﺎس ﺿﺮاﯾﺐ ﻣﺘﻐﯿﺮﻫﺎي ﻓﺎزي ﺷﻬﻮدي ﺗﺒﺪﯾﻞ ﮔﺮدﯾﺪه و ﺳﺮاﻧﺠﺎم ﺑﻪ ﮐﻤﮏ ﻋﻤﻠﮕﺮﻫﺎي ﺣﺴﺎﺑﯽ روي دادهﻫﺎي ﻓﺎزي ﺷﻬﻮدي ﺑﻪ ﺳﺎﺧﺘﺎرﻫﺎي دو ﻣﺮﺣﻠﻪ اي ﻗﻄﻌﯽ ﺗﺒﺪﯾﻞ ﻣﯽﺷﻮد. اﻫﻤﯿﺖ اﯾﻦ ﻣﺪل اﻧﺪازه ﮔﯿﺮي ﻣﻘﺎدﯾﺮ ﻣﺘﻐﯿﺮﻫﺎي ﮐﻤﺒﻮد و ﻣﺎزاد ﻣﯽﺑﺎﺷﺪ، ﮐﻪ ﺑﺮ اﺳﺎس ﻣﺪل ﺗﻮن و ﺗﺴﻮﺗﺴﻮي ﺑﻪ ﺑﻬﯿﻨﻪ ﺳﺎزي ﻣﻘﺎدﯾﺮ ﻣﯿﺎﻧﯽ ﺑﺮاي واﺣﺪﻫﺎي ﻧﺎﮐﺎرا ﻣﯽﭘﺮدازد و ﻧﻬﺎﯾﺘﺎ ﻧﺎﮐﺎراﯾﯽ را ﺑﻬﺘﺮ ﻧﺸﺎن ﻣﯽدﻫﺪ. ﺳﺮاﻧﺠﺎم ﻣﻘﺎدﯾﺮ ﺑﻬﯿﻨﻪ ﺷﺪهي ﻣﯿﺎﻧﯽ در ﻣﺪل ﻣﻄﺮح ﺷﺪه ﻟﺤﺎظ ﮔﺮدﯾﺪه و ﺑﺪﯾﻦ ﻃﺮﯾﻖ ﻣﻨﺠﺮ ﺑﻪ ﺑﻬﺒﻮد ﻣﻘﺪار ﮐﺎراﯾﯽ ﮐﻞ ﺳﯿﺴﺘﻢ ﻣﯽﺷﻮﻧﺪ.

Data Envelopment Analysis (DEA) is one of the best tools for evaluating the performance of decision-making units. Traditional DEA fails to measure efficient and inefficient units and evaluate the performance of network systems, and traditional models of DEA do not pay attention to internal structures and intermediate values. For this reason, in recent years, DEA model, known as network Data Envelopment Analysis models, have been introduced, this models eliminated this deficiency by considering intermediate values. In this paper, DEA based on network two-stage and slack variables and triangular intuitionistic fuzzy data is used to identify the efficiency of units. At first, the two-stage DEA model is introduced and then the model is transformed based on intuitionistic fuzzy coefficients and variables and finally it is converted to crisp two-stage structures by arithmetic operators on intuitionistic fuzzy data. The importance of this model is to measure the values of slack variables, which based on the Tone and Tsutsui model optimizes the intermediate values for inefficient units and ultimately shows better inefficiency. Finally, the optimized intermediate values are considered in the proposed model and thus are improved the overall efficiency of the system.