This document discusses using a back propagation neural network (BPNN) to predict carbon monoxide (CO) emissions from a diesel engine. It begins by providing background on BPNN and how it was applied in this study. Experimental data on engine parameters and CO emissions were collected from tests. The data were divided into training and testing sets to train and validate the BPNN. Different combinations of engine parameters were used as inputs to the BPNN in various "strategies" to determine the best parameters for accurately predicting CO emissions. The BPNN architecture and training parameters were optimized to minimize error between predicted and actual CO emissions. The goal was to develop a method for predicting emissions to better control engine parameters and reduce pollution.