Design and modeling challenges for high density packaging

Today most of the IC and board designs are undertaken using two-dimensional graphics tools and rule checks. System-in-package is driving three-dimensional design concepts and this is posing a number of challenges for electronic design automation (EDA) software vendors. System-in-package requires three-dimensional EDA tools and design collaboration systems with appropriate manufacturing and assembly rules for these expanding technologies. Simulation and Analysis tools today focus on one aspect of the design requirement, for example, thermal, electrical or mechanical. System-in-Package requires analysis and simulation tools that can easily capture the complex three dimensional structures and provided integrated fast solutions to issues such as thermal management, reliability, electromagnetic interference, etc. This paper discusses some of the challenges faced by the design and analysis community in providing appropriate tools to engineers for System-in-Package design

Upgrading automation for nuclear fuel in-core management: from the symbolic generation of configurations to the neural adaptation of heuristics

FUELCON is an expert system in nuclear engineering. Its task is optimized refueling-design, which is crucial to keep down operation costs at a plant. FUELCON proposes sets of alternative configurations of fuel-allocation; the fuel is positioned in a grid representing the core of a reactor. The practitioner of in-core fuel management uses FUELCON to generate a reasonably good configuration for the situation at hand. The domain expert, on the other hand, resorts to the system to test heuristics and discover new ones, for the task described above. Expert use involves a manual phase of revising the ruleset, based on performance during previous iterations in the same session. This paper is concerned with a new phase: the design of a neural component to carry out the revision automatically. Such an automated revision considers previous performance of the system and uses it for adaptation and learning better rules. The neural component is based on a particular schema for a symbolic to recurrent-analogue bridge, called NIPPL, and on the reinforcement learning of neural networks for the adaptation.