The Rapid Manufacturing Lab has been involved in a number of research projects that investigate and build technologies that are necessary to achieve cost effective small batch manufacturing; ultimately, aiming for economic ordering quatities of one.
For the last decade, our laboratory has been developing software that facilitates the rapid production of parts for machining and sheet metal bending processes. These systems known as the Intelligent Machining Workstation and the Intelligent Bending Workstation plan all aspects of part production (e.g., machine setup and machine operation) from the part design, machine description and component resources (i.e., tools, robot grippers, and fixtures). The final plans are then sent to a machine controller that executes them directly on the machine, thus producing the first part right. In the first part production, there are extra gaging operations and possibly even human input to account for possible robot errors and part positioning problems. The adjustment information gleaned from the added gaging steps are then fed back to the planner, which updates the plan for the 2nd part. Finally, the extra steps are removed from the plan in order to improve production time. The result is that a first part can be produced in 20 minutes, where several human experts can take 2-3 days. And the 2nd part production time approaches the production cycle times that are achievable by the same human experts.
Our laboratory is now interested in combining these systems with other manufacturing decision systems (e.g., design, scheduling and costing systems). We are interested in collaborative efforts to build these distributed decision systems, in order to achieve corporate improvement goals. In addition, we are also interested in other production domains that would be suited to this rapid production technology.
We are current building an Intelligent (sheet metal) Bending Workstation that can start with a design for a part and plan the sequence of bending operations, the layout of the tooling stages, the robot and repo grasping locations, and the robot motions and bending machine operations required to make the part. Amada America Inc. is sponsoring this R&D project. We also work closely with Design One Software in Pittsburgh.
David Alan Bourne, Senior Scientist in Carnegie Mellon's Robotics Institute, is the head of the Rapid Manufacturing Lab. Current team members include:
Past team members include:
Return to the beginning