This award provides funding for the development and implementation of an adaptive, closed-loop production control algorithm and concomitant hybrid pull system. The algorithm and its associated tactical production control software will regulate production in such a way as to meet the production quota with a small amount of excessive inventory. A recursive control methodology featuring the use of a Kalman Filter will be employed to estimate the state of the production system. The Kalman Filter statistically weighs differences between actual and expected number of good pieces produced, and hence, indicates, on a near real-time basis, the adjustments to the product flow that will meet the production quota. The adaptive process control algorithm will be implemented on one of two identical lines. An experiment will compare the line controlled with the proposed system against the line with the traditional push system. The measures of effectiveness will include meeting the production quota on time, the amount of work-in-process inventory, and accumulated shipping inventory. The primary goal of this work is to determine the parameters that will affect accumulation and starvation of product, and to design adaptive techniques that will regulate production flow. If successful, the results of this research will lead to an integrated environment for agile and lean production control. This can positively impact industry with sequential manual assembly; particularly those with high turnover, unstable supply chains and variable yield rates. The adaptive, recursive production control system will help these manufacturers to supply small quantities of a large mix of products rapidly, economically and on-demand, while maintaining surge capacity to meet peaks and valleys in demand levels. The research will also contribute to the analytical tools and modeling techniques for production and inventory control problems.
