Clustering Crude Oil Samples Using Swarm Intelligence F. Ferreira 1 , T. Ciodaro 1 , J. M. de Seixas 1 , G. Xavier 2 , and A. Torres 3 1 Signal Processing Lab, COPPE/Poli - Federal University of Rio de Janeiro fferreira,ciodaro,seixas@lps.ufrj.br 2 PETROBRAS Research & Development Center gilberto.xavier@petrobras.com.br 3 FAT - State University of Rio de Janeiro artorres.uerj@gmail.com Abstract. The identification patterns in the crude oil intrinsic qualities provides useful information for the refinery operation and logistics. The a priori information concerning the characteristics expected by a given crude oil improves the logistic concerning which refineries should process this crude, together with pricing and marketing. This article presents the results of data mining models applied to a generic database of crude oil samples. Only information available in the crude oil intrinsic qualities is used. Clustering techniques based on bio-inspired algorithms are applied to the data samples in order to extract structured patterns from data. Three algorithms were used: PSO, FSS and ABC. Particles and fishes represent the possible clustering solutions. ABC represents solutions as food sources to be evaluated by bees. The silhouette index was used as the fitness function to be optimized. The results were later evaluated using other clustering quality index. The algorithms were able to find patterns beyond the standard oil classification, which considers only the oil density measurement. Keywords: Oil and Gas, Swarm Clustering, Data Mining, Pattern Recognition 1 Introduction The oil logistic and refining processes are dependent on the type and charac- teristics of the crude oil feedstock. In fact, the refinery operation is designed to process crude oils with certain characteristics, followed by the transport logistic to the refinery. Depending on the conditions, different crude oils are blended in order to match the characteristics supported by a given refinery. Crude oil (a.k.a petroleum) is the world’s most important fossil fuel today and has invaluable economical importance. The current amount of exploitable quantities of conventional crude oil is estimated at billions barrels. Although, the proven quantities are huge, oil sources are obviously not inexhaustible. According to the International Energy Agency this will lead to shortness in the exploitable