Overview
As manufacturing companies increase the level of customization in their product offerings, move towards smaller lot production and experiment with more flexible customer/supplier arrangements such as those made possible by Electronic Data Interchange (EDI)/Electronic Commerce (EC), they increasingly require the ability to respond quickly, accurately and competitively to customer requests for bids on new products and efficiently work out supplier/subcontractor arrangements for these new products. This in turn requires the ability to rapdily convert standard-based product specifications into process plans and integrate new orders with their process plans into existing production schedules across the supply chain.

As part of the Agile Manufacturing Initiative sponsored by DARPA's Defense Sciences Office (under BAA #94-31), the Raytheon Company and Carnegie Mellon University have joined together in the development of an Integrated Process Planning/Production Scheduling (IP3S) shell for agile manufacturing [Sadeh et al., 98].

The IP3S architecture is designed around an innovative blackboard architecture that emphasizes:

1. concurrent development and dynamic revision of integrated process planning and production scheduling solutions
2. mixed-initiative functionalities that support the maintenance of multiple problem instances and solutions, and allow the user to control the development of solutions by exploring alternative tradeoffs and imposing or retracting various assumptions
3. workflow management functionalities to alert and remind the user of new events or existing conditions that have not yet been addressed in a particular solution
4. coordination with outside information sources such as customer and supplier sites
5. the use of a common representation for exchanging process planning and production scheduling information
6. portability and ease of integration with legacy systems

The IP3S project builds on advanced process planning and dynamic production scheduling technologies developed in the context of Raytheon's IPPI planning system and Carnegie Mellon's Micro-Boss scheduling system. The IP3S shell is scheduled for initial evaluation in a large and highly dynamic machine shop at Raytheon's Andover, MA manufacturing facility.

Expected benefits of this project include significant reductions in production leadtimes and significant improvements in delivery date accuracy, resource utilization, end-user productivity, and coordination with outside information sources. With its emphasis on portability and extensibility, the resulting IP3S technology should improve the ability of manufacturing enterprises to compete in a climate of continuous and often rapid change.

Motivation
In practice, process planning and production scheduling activities are typically handled independently and are carried out in a rigid, sequential manner with very little interaction. Process alternatives are traded off strictly from the standpoint of engineering considerations, and plans are developed without consideration of the current ability of the shop to implement them in a cost-effective manner. Production scheduling is performed under fixed process assumptions and without regard to the opportunities that process alternatives can provide for acceleration of production flows. This lack of coordination leads to unnecessarily long order lead times and increased production costs and inefficiencies. Increased manufacturing agility depends critically on the development of approaches for integrating these two activities.

Objectives
The IP3S project aims to achieve the following objectives:

1. couple a generative process planner with a finite-capacity scheduler
2. enable rapid response to changes
3. shorten cycle times and increase resource utilization
4. complete a pilot demonstration within a large machine shop at Raytheon's Andover manufacturing plant

Status
Experiments with the IP3S shell show an average performance improvement of 23% in solution quality over a more traditional, decoupled approach to building process planning/production scheduling solutions in a dynamic factory environment such as the Raytheon Andover facility. Our group is currently extending this architecture to support a broader range of coordination protocols in the context of a system called MASCOT ("Multi-Agent Supply Chain COordination Tool").

Sponsors

Defense Advanced Research Projects Agency

Affiliates

Raytheon Company

Personnel

• David Hildum
• Norman Sadeh
• Allen Tseng
• Dag Kjenstad

Collaborators

• Thomas J. Laliberty
• John McA'Nulty

Publications
Please note the links to select Postscript files (some gzipped):

	1998
[Sadeh et al., 98] 23 pages (240 Kbytes)	Norman M. Sadeh, David W. Hildum, Thomas J. Laliberty, John McA'Nulty, Dag Kjenstad, and Allen Tseng. A Blackboard Architecture for Integrating Process Planning and Production Scheduling. Concurrent Engineering: Research & Applications (CERA), 6(2), June 1998, pp. 88-100.
	1997
[Sadeh et al., 97]	Norman M. Sadeh, David W. Hildum, Thomas J. Laliberty, John McA'Nulty, Stephen F. Smith, Dag Kjenstad, and Allen Tseng. Integration of Process Planning and Production Scheduling for Agile Manufacturing: A Case Study. Proceedings, ASME Design for Manufacturing Conference, Davis CA, September 1997. American Society of Manufacturing Engineers (ASME).
[Hildum et al., 97] 6 pages (208 Kbytes)	David W. Hildum, Norman M. Sadeh, Thomas J. Laliberty, John McA'Nulty, Stephen F. Smith, and Dag Kjenstad. Blackboard Agents for Mixed-Initiative Management of Integrated Process-Planning/Production-Scheduling Solutions Across the Supply Chain. Proceedings, Ninth Conference on Innovative Applications of Artificial Intelligence (IAAI-97), Providence RI, July 1997. American Association for Artificial Intelligence (AAAI).
	1996
[Hildum et al., 96]	David W. Hildum, Norman M. Sadeh, Thomas J. Laliberty, Stephen F. Smith, John McA'Nulty, and Dag Kjenstad. Mixed-Initiative Management of Integrated Process-Planning and Production-Scheduling Solutions. AI and Manufacturing Research Planning Workshop, Albuquerque NM, June 1996. American Association for Artificial Intelligence (AAAI).
[Sadeh et al., 96b] 20 pages (464 Kbytes)	Norman M. Sadeh, David W. Hildum, Thomas J. Laliberty, Stephen F. Smith, John McA'Nulty, and Dag Kjenstad. An Integrated Process-Planning/Production-Scheduling Shell for Agile Manufacturing. Technical Report #CMU-RI-TR-96-10, The Robotics Institute, Carnegie Mellon University, Pittsburgh PA, May 1996.
[Sadeh et al., 96a]	Norman M. Sadeh, Thomas J. Laliberty, David W. Hildum, John McA'Nulty, Robert V.E. Bryant, Stephen F. Smith, David Flood, and Ann Gardner. Development of an Integrated Process Planning / Production Scheduling Shell for Agile Manufacturing. Proceedings, Fifth National Agility Conference, Boston MA, March 1996. Agile Manufacturing Enterprise Forum (Agility Forum).

back to top