The European Parliament identifies virgin olive oil (VOO) as one of the foods which are often subject to fraudulent activities. Possibilities of adulteration are the application of illegal soft deodorization of extra virgin olive oil (EVOO) or the commercialization of blends of EVOO with soft-deodorized EVOO or refined vegetable oils. Despite the search for possibilities to prove the illegal soft deodorization of EVOO or the addition of cheaper vegetable oils to EVOO, suitable methods are still missing. Therefore, the aim of the study is to develop a new analytical and statistical approach addressing detection of mild deodorization or addition of refined foreign oils. For this purpose, VOOs are treated in lab-scale for 1 h up to 28 days at different temperatures (20, 50, 60, 80,100, 110, and 170°C) in order to simulate and study the effect of heat treatment on known analytical parameters by near infrared spectroscopy (NIR). A logit regression model enabling the calculation of the probability for a heat treatment is developed. This new methodology allows detecting both soft deodorized olive oils and blends of EVOO with cheaper full refined vegetable oils. Adding only 10% of full refined oil could be detected in extra VOO.Practical Applications: NIR methods combined with chemometrics have become one of the most attractive analytical tools to control quality of food. It is a simple, precise, and rapid method. All relevant analytical parameters of oxidative and thermal fat degradation can be determined in a single run and be used to detect adulterated virgin olive oils (VOOs). The use of a simple equation developed from the logistic regression using peroxide value, K-values, p-anisidine value, pyropheophytine, 1,2-diacylglycerols, total polar compounds and monomeric oxidized triacylglycerols, and other well-known parameters allows to detect mild deodorized olive oils or also blends of VOO with soft-deodorized ones or the addition of low amounts of foreign vegetable oils. This technique has potential to be used as a screening method for the detection of adulterated olive oils using both the traditional laboratory methods and the corresponding NIR-methods. R 2 : coefficient of determination; d) RMSEP: root mean square error of prediction; e) MD: limit of Mahalanobis distances.