Due to the global competition, the survival of enterprises must face the major test. Since the poor management of the market will increase number of companies, so the crisis early warning model of business has the necessary to establish. The cause of financial crisis is the main source of financial situation of the deteriorating. Therefore, if we could analysis the facets and weights of potential affect factors through the financial and managerial situation of business to judge the crisis cause of a corporation and establish the early warning model is worth to discuss deeply. The precious year of companies¡¦ data that this study collecting are from the Taiwan Securities Exchange 2006/01/01-2008/12/31 which have been out of the open security market based on the analysis standards and omitting the less information and banking, have total 36 enterprises data for analysis. The application of total variables, this study pre-adapts the TEJ business credit risk indicators to integrate the documentation and analysis the fundamental variables. It can be seen that all the factors have the relationship with each other through this study. This highlights a very important message, and also to the crisis among the factors and normal company with a considerable fluctuations. The judging results of DEA-DA model show that most of the company might be affected with some important factors of interpretation in abnormal situation to let the company in crisis cluster. Through Logistic regression analysis results show that our study forecasting model has the great explanatory power to meet the behavior of interpretation with the crisis and normal companies. By the enterprises crisis model of this study building to assess and forecast the crisis situation have the same results with the simplified model constructing with key factors to affect the original model direction. This study shows a very important fact that the crisis forecasting models will not be simplified to change the outcome which also indicating to increase of variables won¡¦t change the results of the assessment. In accordance with this study proposed model, if value positive that would be show more and more vulnerable to crisis. By other words, if value negative that would be more small vulnerable to crisis.

There is much debate among genocide scholars as to the causes and even accurate definitions of genocide. Early warning developed to address the increasing need for humanitarian intervention in violent conflicts around the world. As a subset of genocide studies, early warning seeks to go beyond explaining the causes of genocide. The early warning model created here uses six indicator variables - government, leaders/elites, followers, non-followers/bystanders, outsider group, and environment - to detect the likelihood of genocide within a given case study. Four cases were chosen - Kenya, Nigeria, Yemen, and Ethiopia - and analyzed using the indicator variables to determine if these violent conflicts may already be or may become genocides. Preliminary findings show that the civilian outsider group is a vital component when determining whether or not a conflict is or may become a "limited-genocide" and that genocides are a function of the interaction of the six indicator variables and not just their presence. Other implications for sovereignty and humanitarian intervention are discussed.