In some companies, undesirable costs are imposed for provision of services. It is required that current outputs of decision making units are produced at minimal undesirable cost. In this paper, a new method is proposed for evaluating the undesirable cost imposed on the units. Moreover, data is considered in fuzzy number form, due to the observed input-output data are sometimes imprecise in real applications. Applying the -level based approach, the proposed mode is solved to measure minimum undesirable cost. In this method, data information is supposed as triangular fuzzy numbers. Finally, the proposed method is illustrated by a real application. Manuscript profile

Classic Data Envelopment Analysis assumes that the values of inputs and outputs are exactly determined. However, in most real-life problems, the exact values of some of the inputs and outputs are imprecise. One of such imprecise data is ordinal data. In this paper, a new model is presented for measuring the efficiency evaluation of decision making units with ordinal data. The general idea of this model is assigning real values to ordinal data with new approach. Furthermore, another new model for ranking efficient units is presented with the main idea of changes in controlled efficiency. Then, the results of proposed model are compared with Cooper model. Therefore, the efficiency scores obtained from proposed model are more realistic and reasonable than the results obtained from the Cooper's model. Manuscript profile

Mathematical modeling is one of the feasible methods that can be used to solve real problems. Modeling can be done using a variety of methods, including statistical methods that can be used to predict a variety of events. Health is one of the most important areas of research in the world today. Among the various diseases in the health sector, this study concerns stroke which is the second leading cause of death and long-term human disability, that has led to doing this research. The main objective of this research is to design and to build a predictive model of stroke based on symptoms and clinical reports, whether or not stroke occurs in patients in the near future. Using logistic regression technology, the main pathogenic factors of stroke have been found and their incidence has been predicted. In this study, clinical information from 5411 patients was collected and, after applying the LR method, the predictive model was designed. Manuscript profile

Data Envelopment Analysis (DEA) is an eciency measurement tool for evaluation of similar Decision Making Units (DMUs). In DEA, weights are assigned to inputs and outputs and the absolute eciency score is obtained by the ratio of weighted sum of outputs to weighted sum of inputs. In traditional DEA models, this measure is also equivalent with relative eciency score which evaluates DMUs in compare with the most ecient DMU. Recently network DEA models are appeared in the literature, which try to assess DMUs regarding their internal production divisions and intermediate products. In this paper we compare absolute and relative eciency scores in network framework. Since in network DEA models, an ecient DMU does not exist necessarily, the relative eciency model helps us to have at least one ecient DMU in our assessments. Manuscript profile