This paper is dedicated to Prof. Paola Gramatica on the occasion of her retirement.Abstract: In this publication, QSAR models were developed to predict analgesic and anti-inflammatory activities of some 2-benzoxazolinone derivatives using multiple linear regression method. The models were validated internally and externally according to the OECD principles. With the help of these models, pronounced molecular properties of these compounds related to activities were also explored. The developed models demonstrated that hydrophobicity, the number of halogens, and the shape of the molecular structure of these candidate drugs are prominent to represent analgesic and anti-inflammatory activities. Based on the previously tested compounds and the developed models, 77 new compounds were designed as potential analgesic and anti-inflammatory drugs. Majority of the newly designed compounds demonstrated promising analgesic and anti-inflammatory activity.(7 of 9) 1800090 Figure 4. Insubria graph. Predicted vs. leverage values of (a) A activity model and (b) AI activity model.