A NEURAL NETWORK ARCHITECTURE FOR PROCESS TIME ESTIMATION - AN APPLICATION IN PETROLEUM WELLS OPERATIONS

Abstract This paper presents a connectionist methodology that can be used to approach the total time estimation of complex engineering processes. A number of parameters correlated with the total time of the process, were selected from a database. Correlations and regularities were detected using a competitive neural network. A feedforward neural network was trained to estimate the average, standard deviation and total time wasted in the accomplishment of the process. At the end of the paper some experiments for time assessment in drilling and completion operations of deep-water oil wells are presented in order to investigate the possibilities of the method.