An Operations Cyberinfrastructure: Using Cyberinfrastructure and Operations Research to Improve Productivity in the Enterprise

Download 85.76 Kb.
Size85.76 Kb.
1   ...   5   6   7   8   9   10   11   12   13
Product Lifecycle Informatics

Within the interrelated processes of design, manufacturing, and supply-network management, there are a number of areas where the overall productivity of the enterprise can be transformed. CI components such as network support, computing grids, and data storage and analysis systems are not sufficient by themselves to cope with the complexity of modern production and manufacturing enterprises. Embedding of OR into CI will create new possibilities for handling large complex systems at multiple scales.

Management of product manufacturing enterprises is one class of problems that can benefit enormously from a synthesis of OR and CI. The resulting OCI will allow us to develop mechanisms whereby key issues at the enterprise level, the factory level, and the individual product level can be addressed in new and more effective ways. This end-to-end approach in product lifecycle and manufacturing, as implemented and delivered through an OCI, has the potential to lift productivities to very high levels.

While an OCI cannot directly manufacture products, it can be viewed as the “central nervous system” for an enterprise, providing the framework for connections from concept through design to process planning, factory operations, services, retirement, and recycling/reuse. For example, machine sensors can proactively link to production planning systems to support maintenance and self-assessment, with consequent financial impact at all levels: process, factory, supply network, and customer service. Factory operations can be linked together and therefore be responsive to a more demand-driven network. Supply networks of diverse size and scale can be linked by infrastructures that are able to reconfigure and respond quickly.

The present Internet is data- and information-driven, and information overload has become a significant problem in complex product and manufacturing networks. The OCI will be the key to extracting, synthesizing, and exploiting the information most relevant to the production enterprise; it will manage the enterprise’s networks for design, production, and supply concurrency, configuration management, and decision support.

Hybrid approaches that combine experimental component testing with modeling (including real-time interactions) will allow simulation and testing of complex production systems. OR can provide complex optimization technologies as an adjunct to simulation and modeling components; CI will provide the computational capabilities for solving the large-scale distributed optimization and constraint problems that result from the OR formulations.

There is an urgent need for OR-centric enterprise resource planning (ERP) models to promote integrated decision-making in the enterprise. Existing ERP tools sorely lack this capability, because they lack essential CI and/or OR features such as sophisticated modeling capabilities and computational tools. An understanding of human factors at many levels of the manufacturing and decision-making process — the “human in the loop” — is also lacking in the current generation of tools. The OCI will enable tight integration of human factors (including social science and industrial engineering research) with CI and OR tool sets and methodologies. ERP tools can thus be taken to a new level of usefulness, with significant benefits for productivity across the enterprise.

The computing and communication resources available on a CI platform will enable users to perform modeling and optimization at a scale far beyond their current capacities. Perhaps the most exciting possibility here is the potential to apply modeling and optimization techniques to manufacturing enterprises on a global scale.

Finally, we note that the OCI will allow us to avoid replication of costly resources among several domains such as design, manufacturing, services, and modeling tools in optimization and simulation.

Share with your friends:
1   ...   5   6   7   8   9   10   11   12   13

The database is protected by copyright © 2020
send message

    Main page