JFET_V10_N3_RP2 D. Krishna R. Padma Sree Journal on Future Engineering and Technology 2230 – 7184 10 3 11 17 Artificial Neural Network, Adsorption, Adsorbent, Chromium (VI), Pattern Search Algorithm The new process technologies developed during the past years made it possible to produce biodiesel from recycled edible oils comparable in quality to that of virgin vegetable oil. Biodiesel has an added attractive advantage of being lower in price. Thus, biodiesel produced from recycled edible oils has the same possibilities to be used. From an economic point of view, the production of biodiesel is very feedstock sensitive. From a waste management standpoint, producing biodiesel from used edible oil is environmentally beneficial, since it provides a cleaner way for disposing these products; meanwhile, it can yield valuable cuts in CO2 as well as significant tail-pipe pollution gains. This paper is about the manufacturing of biodiesel from the used vegetable oil. The study aims to define the requirements for biodiesel production by the esterification process, testing its quality by determining some parameters such as Degree API, Gross Calorific Value, Flash point, Specific Gravity and Fire point and comparing it to the commercial Diesel fuel, and the strategic issues to be considered to assess its feasibility, or likelihood of success. The experimentation was carried out for varying booster dosages from 0.2 to 1 gram at 60oC. The experimental results show that the biodiesel obtained at the conditions of oil: alcohol ratio, 6:1, at catalyst dosage 1 gram, at a temperature of 60oC and booster dosages of 0.2 to 1 gram was of good quality.Artificial Neural Network (ANN) was developed by a single layer feed forward back propagation network to the batch experimental data to develop and validate a model that can predict Cr (VI) removal efficiency. ANN is effective in modeling and simulation of highly non linear multivariable relationships. A well-designed network can converge even on multiple numbers of variables at a time without any complex modeling or empirical calculations. The prediction of removal of Cr (VI) from wastewater has been made using variables of pH, adsorbent dosage and initial chromium (VI) concentration. Different types of ANN architecture were tested by varying the neuron number of entrance and the hidden layers, resulting in an excellent agreement between the experimental data and the predicted values. The high 2 correlation coefficient (R =0.992) between the ANN model and the experimental data showed that the model was able to find out the percentage removal of chromium (VI) proficiently. Pattern search method in genetic algorithm was used to obtain the optimum values of input parameters for the maximum percentage removal of chromium (VI). February - April 2015 Copyright © 2015 i-manager publications. All rights reserved. i-manager Publications http://www.imanagerpublications.com/Article.aspx?ArticleId=3344