Timothy Matis, Milton Smith
Texas Tech University
This project develops an algorithm used for production planning that can make real time adjustments to the scheduling of production operations at B&W Pantex. This algorithm is sensitive to the many variables that affect daily production rates.
Pantex Plant is a government owned and contractor operated manufacturing facility for the US Department of Energy. A research opportunity has been identified to investigate how to improve the existing production and inventory control models used by the weapon operations area to manufacture and store weapons. The facility currently uses technology to schedule a daily production plan that does not consider all resource constraints. Production runs are scheduled around available resources with very complex facility constraints. Between each assembly phase, testing is performed to maintain high levels of product quality. There are a limited number of test facilities; therefore, there is a potential bottleneck in the production process. In addition to physical constraints in the production area, there are a limited number of production technicians that are certified to perform tasks, creating additional constraints on the flexibility of the production schedule.
This project resulted in the modification of C++ code for the near term scheduling of production operations at BWXT/PANTEX. The code has been shown to execute quickly for large problems that are representative of those encountered at the facility.