Diesel engines exhausts vehemently threaten the environment and are one of the significant sources of air pollution. This research paper is aimed at predicting the risk level of carbon monoxide (CO) emission produced by diesel engine generators. Emission data on the diesel engine and the emission parameters were obtained for 105 diesel engine generators. The four input variables used to develop the model were: the installed capacity, the purchase time, the operating time and the time last serviced. Multiple linear regression modeling technique was used to develop the output (risk level of emission). The adjusted coefficient of determination, R2 value of the optimum model, was 0.382 which explains about 38.2% of the variance in CO emission; also the Cook’s distance (0.131) was less than 1 which signifies minor or no problem in the sample size used. Model validation showed a strong correlation of 89.5% between the model predictions and human predictions. At a significance level of 5%, the t-test showed that there was no significant difference between the model prediction risk level and the human prediction risk level. The model in this study is useful to industry, government environmental protection agency and private users during the determination of the risk level of CO emissions from diesel engine generators.
Key words: emission, generator, regression model, risk