Data mining play a vital role in computer field . A huge and valuable knowledge is extracted from the large collection of data. outlier detection is currently an important and active research problem in many fields and is involved in numerous applications. This paper applies minimum volume ellipsoid (MVE) with principle component analysis (PCA) extension, a powerful algorithm for detecting multivariate outliers. If the data points exceed the cut-off value, the cook’s distance is used for the outliers. The paper also compares the performance of the suggested frame work with statistical methods to demonstrate its validity through simulation and experimental applications for incident detection in the field of agriculture. Keywords: outlier detection, PCA, MVE, cook’s distance.