Plan for New Manufacturing Faculty with a Model-based Personalization Focus

“The future ability of the United States to innovate and invent new products and industries, provide high quality jobs to its citizens, and ensure national security depends upon how well we support innovation and the development and use of advanced technologies for our manufacturing sector… The Nation’s historic leadership in manufacturing, however, is at risk” (PCAST Report 2011).

Model-based personalization has emerged as a core unifying theme for next generation manufacturing at Purdue. This truly revolutionary capability has the potential to change how products are designed, manufactured and delivered globally. It involves combining latest advances in tailored materials, on-demand design and manufacturing processes, multi-scale modeling of products and systems, and ubiquitous cyber-infrastructure to sustainably deliver personalized products, anywhere, anytime, with the efficiency of mass production.

The provost of Purdue University has approved a proposal by faculty representing eight departments and four colleges at Purdue to hire six new faculty in the area of model-based personalization of products and services. Six areas are targeted:

• Cyber-enabled Experimentation: Introduction of new materials and processes into next generation products is severely constrained by the cost and latency of physical experimentation. Computationally driven experimentation, measurement and characterization can facilitate rapid product process optimization, visual analytics and decision-making. This hire will focus on identifying faculty with a multi-disciplinary background in computational sciences, statistics and physical processes/materials modeling.
• Personalization: Revolutionary advances in the ubiquitous connectivity and the sheer magnitude of available computational resources, both in terms of cycles and storage, have the potential to fundamentally change manufacturing and delivery of products and services. We can now envision delivering personalized products and services with the efficiency of mass production. The ability to sense-and-respond in real-time, connect with their global supply chains, conduct design, evaluation and analysis of products with distributed partners can allow manufacturing enterprises to personalize products to meet individual customer needs. Moreover, personalized manufacturing has the potential to profoundly change how healthcare is delivered to patients: i) by providing exact replicas of individual organs for implants during surgery or printing drug doses in a home (based on embedded sensor readings), ii) on-demand design and production of products and services, and iii) just-in-place therapeutics, prosthetics, and drug delivery
• Multi-scale Predictive Modeling: Multi-scale modeling and simulation approaches to characterize product properties based on material and process specifications are needed in order to develop effective predictive capabilities. The development of effective multi-scale modeling methods that go beyond a single discipline has been enabled by major breakthroughs in computational mathematics and new thinking on how to model events occurring at multiple scales. The next generation of predictive models offers us the ability to rapidly synthesize design alternatives for systems and processes, by capturing temporal dynamics and spatial variations in a unified fashion, across large-scale heterogeneous media, processes and infrastructures. Verification and 2 validation of models of complex systems presents an extremely challenging task.
• Scaling Principles: Over the past two decades, significant advances have been made in our ability to create material structures from atomistic and genomic levels with desired characteristics. These discoveries have the promise of enabling the creation of macro-scale products with unique functional properties that cannot be achieved with traditional materials. However, most of these materials cannot yet be produced economically in commercial quantities and in openmanufacturing conditions. Moreover, advances in our understanding of living organisms as incredibly complex factories making, processing and storing food and other ingredients to sustain life can lead to radically different production processes and methods in our factories. The primary challenge in transitioning the discoveries in advanced materials and living organisms to products that revolutionize society is in inventing new manufacturing processes and methods. Key emphasis areas of this thrust area would be:
  • identifying scaling principles for nano- and biomaterials processing, and developing commercial scale manufacturing processes
  • developing high-throughput, low-investment manufacturing processes for renewable energy generation devices, such as photovoltaic cells and lithium batteries, and
  • knowledge-based self-organizing enterprises.
• Optimization and Design On-the-fly: Several aspects of next-generation manufacturing require the ability to design materials or products using fast computational algorithms, and to optimize the design in real-time. This requires expertise in algorithms and software for computing sensitivities to design parameters and the ability to design when the parameters have uncertainties in them. Computationally these require the ability to effectively solve stochastic nonlinear programs, which need high performance computing resources. On-the-fly design and optimization also makes use of real-time optimization, which also necessitates the use of high performance computing.
• Social Manufacturing: Supply chains are evolving into social manufacturing, where sourcing will be done over social networks. Developments in the study of large-scale networks have enabled the design of interactions and network topologies that lead to desired system behavior and performance. Approaches to design and analyze social networks are needed to manage next generation of manufacturing enterprises