Increasing young population and rising unemployment has led to employment and related indicators of serious problems and controversial country. In conjunction with the study and research on employment provinces, a variety of variables and factors involved and It seems that the lack of balance in the distribution of these variables, social inequality, economic, political and cultural between the provinces follow. Indices of employment in the provinces spatial distribution of employment index shows. The main purpose of this study is to clarification social inequalities of employment for equal distribution of possibilities in provinces. In this paper 55 softened are used. Variables dropped to 10 by means of advanced statistics methods such as factor analysis and presented in combination with meaningful factors. The importance of every factor is given too. Then the country provinces were divided into five equal classes by the cluster analysis model. The research results indicate this fact that there are serious imbalances in the country. So that in terms of employment Tehran province is in the first and south Khorasan province in the last place.

Today, the knowledge of regions’ strong and weak points and the existing gaps between regions is necessary for better planning and programming to deal with them. Using such indexes as economic, sociologic, cultural, sanitary and other indexes which can be good criteria for the position and place determination of these quotients, is the main solving problem factor to achieve economic welfare, sociologic health in assessing stable development. That is the more careful this process is the more effectiveness is expected. Thus in this study three methods of numerical taxonomy, factor analysis and fuzzy logic have been used for the identification of Khuzestan province counties development ranking in 1379 and in 1386. Besides this ranking, comparative analysis of these methods has been done. Results show that the development opportunities of many counties in this province in 1386 compared to 1379 have been destroyed. Results also indicate that condense variability in numerical taxonomy is less than the two other methods. In fact, this method stands between the two methods in the ranking. Perhaps the difference in multiplier is due to standardization which can omit many clustering aspects of criteria. Indeed, it seems that fuzzy logic ranking is more reliable.