The need to anticipate the consequences of policy decisions becomes ever more important as the magnitude of the potential consequences grows. The multiplicity of connections between the components of society and the economy makes intuitive assessments extremely unreliable. Agent-based modeling has the potential to be a powerful tool in modeling policy impacts. The direct mapping between agents and elements of society and the economy simplify the mapping of real world functions into the world of computation assessment. Our modeling initiative is motivated by the desire to facilitate informed public debate on alternative policies for how we, as a nation, provide healthcare to our population. We explore the implications of this motivation on the design and implementation of a model. We discuss the choice of an agent-based modeling approach and contrast it to micro-simulation and systems dynamics approaches.
The present paper explores group dynamics and electronic communication, two components of wicked problem solving that are inherent to the national security environment (as well as many other business environments). First, because there can be no ''right'' answer or solution without first having agreement about the definition of the problem and the social meaning of a ''right solution'', these problems (often) fundamentally relate to the social aspects of groups, an area with much empirical research and application still needed. Second, as computer networks have been increasingly used to conduct business with decreased costs, increased information accessibility, and rapid document, database, and message exchange, electronic communication enables a new form of problem solving group that has yet to be well understood, especially as it relates to solving wicked problems.
An experiment is proposed which will compare the effectiveness of individual versus group brainstorming in addressing difficult, real world challenges. Previous research into electronic brainstorming has largely been limited to laboratory experiments using small groups of students answering questions irrelevant to an industrial setting. The proposed experiment attempts to extend current findings to real-world employees and organization-relevant challenges. Our employees will brainstorm ideas over the course of several days, echoing the real-world scenario in an industrial setting. The methodology and hypotheses to be tested are presented along with two questions for the experimental brainstorming sessions. One question has been used in prior work and will allow calibration of the new results with existing work. The second question qualifies as a complicated, perhaps even wickedly hard, question, with relevance to modern management practices.
Discrete models of large, complex systems like national infrastructures and complex logistics frameworks naturally incorporate many modeling uncertainties. Consequently, there is a clear need for optimization techniques that can robustly account for risks associated with modeling uncertainties. This report summarizes the progress of the Late-Start LDRD 'Robust Analysis of Largescale Combinatorial Applications'. This project developed new heuristics for solving robust optimization models, and developed new robust optimization models for describing uncertainty scenarios.
The 9/30/2007 ASC Level 2 Post-Processing V&V Milestone (Milestone 2360) contains functionality required by the user community for certain verification and validation tasks. These capabilities include loading of edge and face data on an Exodus mesh, run-time computation of an exact solution to a verification problem, delivery of results data from the server to the client, computation of an integral-based error metric, simultaneous loading of simulation and test data, and comparison of that data using visual and quantitative methods. The capabilities were tested extensively by performing a typical ALEGRA HEDP verification task. In addition, a number of stretch criteria were met including completion of a verification task on a 13 million element mesh.
An experiment was conducted comparing the effectiveness of individual versus group electronic brainstorming in order to address difficult, real world challenges. While industrial reliance on electronic communications has become ubiquitous, empirical and theoretical understanding of the bounds of its effectiveness have been limited. Previous research using short-term, laboratory experiments have engaged small groups of students in answering questions irrelevant to an industrial setting. The current experiment extends current findings beyond the laboratory to larger groups of real-world employees addressing organization-relevant challenges over the course of four days. Findings are twofold. First, the data demonstrate that (for this design) individuals perform at least as well as groups in producing quantity of electronic ideas, regardless of brainstorming duration. However, when judged with respect to quality along three dimensions (originality, feasibility, and effectiveness), the individuals significantly (p<0.05) out performed the group working together. The theoretical and applied (e.g., cost effectiveness) implications of this finding are discussed. Second, the current experiment yielded several viable solutions to the wickedly difficult problem that was posed.
Techniques for high throughput determinations of interactomes, together with high resolution protein collocalizations maps within organelles and through membranes will soon create a vast resource. With these data, biological descriptions, akin to the high dimensional phase spaces familiar to physicists, will become possible. These descriptions will capture sufficient information to make possible realistic, system-level models of cells. The descriptions and the computational models they enable will require powerful computing techniques. This report is offered as a call to the computational biology community to begin thinking at this scale and as a challenge to develop the required algorithms and codes to make use of the new data.3
This LDRD Final report describes work that Stephen W. Thomas performed in 2006. The initial problem was to develop a modeling, simulation, and optimization strategy for the design of a high speed microsystem switch. The challenge was to model the right phenomena at the right level of fidelity, and capture the right design parameters. This effort focused on the design context, in contrast to other Sandia efforts focus on high-fidelity assessment. This report contains the initial proposal and the annual progress report. This report also describes exploratory work on micromaching using femtosecond lasers. Steve's time developing a proposal and collaboration on this topic was partly funded by this LDRD.