While many statistical measures have been devised to measure team and player performance in professional baseball, relatively little work has been done to study and improve measures of defensive performance. This project’s goals were to model defensive contribution to team performance and develop related measures of individual player defensive production. Factor analysis was employed to extract positional factors and combine them with pitching and offensive measures into a final logistic regression model to determine effects of defense on team performance.
Preventing the smuggling of radiological and nuclear material into the United States has become a major effort of the DHS-DNDO since the tragic events of September 11, 2001. The implementation of the DNDO's GNDA consists of several layers of opportunity to detect RN material overseas, at the borders, and within the US. One of the most abundant resource DNDO can use is the presence of weigh stations on the borders and throughout the US interior. Installing detection equipment into a system that is already in place will not only save money, but increase security and reduce the threat of RN material. By modeling a weigh station and implementing proposed solutions it is found that the detection equipment has little impact on the current system, yet greatly increases the probability of catching potential threats.
The American Recovery and Reinvestment Act introduced the Transit Security Grant Program. This program allows transit authorities to apply for grants to cover projects protecting against weapons of mass destruction and other terrorist attacks. This paper looks at how to calculate potential risk for two attack scenarios at three Washington Metropolitan Area Transit Authority stations. Risk is measured as the likelihood of the event taking place versus the result of losing the asset. The likelihood is split into the potential threat of an attack at the asset and the likelihood the attack is successful. Using Analytica and influence diagrams, this paper explores which scenario has the most risk.
Utility Theory is common in several decision models, but rarely is Prospect Theory ever considered. This project looks at how Prospect Theory can explain deviances from traditional Utility Theory results.
FEMA is charged with supporting the United States public with food, water, shelter, and other supplies during and after hazards. This paper will discuss a Bayesian Network approach to determining the amount of people to be employed by FEMA for disaster recovery support. The model built to accompany this paper will use information on eight hazard categories to determine the amount of personnel required compared to the amount of personnel available. Netica is used to build the model and display the prioritized options for personnel decisions.
The Northern Virginia Senior Softball (NVSS) League assigns members to teams among three conferences at the beginning of each season. This paper describes the dynamic programming approach to assigning players to each team though use of the NVSS goals.
The teams in the Northern Virginia Senior Softball (NVSS) league are reorganized every year with the desire to make them as even as possible and promote social interactions. This project’s goals were to develop a methodology to produce team assignments for players such that the teams are equal in regards to player ability and other attributes. Multiple regression was employed to obtain offensive, defensive, and team metrics. An integer program (IP) model was developed where constraints were aligned with the goals of the NVSS league to solve the assignment problem. The IP model was linked to a team assignment tool which allows the user to select settings and parameters before running the model.
No comments:
Post a Comment