Date of Award


Degree Name

Bachelor of Science


Paper Science and Engineering


This study was a limited laboratory scale investigation of whether or not linear programming was a viable technique for determining the least cost furnish blends. It is original in that it used actual laboratory developed data for input to determine the linear programming model, and the results were actually produced in the laboratory to see if constraints were met. The materials used were a bleached hardwood, a bleached softwood, tab cards, clay and TiO2. It was found that requirements of linearity and averaging inherent in the linear programming caused results which were not as accurate as needed. However, by using the technique several times in a successive approximation type procedure, readjusting between uses to compensate for the problems previously noted, results of sufficient accuracy to be realistically depended upon were obtained. It is felt this justifies considerable optimism for this technique as a means of constantly economizing furnish costs.