No description available.

Students in academic environments today use ever evolving technology to complete their academic tasks. This paper describes the software engineering techniques and practices that were used to develop a web application capable of assisting students with the common academic task of exam preparation. Despite the common utilization of technology tools to gather and assemble information for academic courses, technology applications that support students in preparing to use their assimilated information, namely on an exam, are currently lacking. The web application described in this report strives to provide a technical solution that supports student efforts towards preparing for an exam. The results have leveraged sound software engineering practices with new and innovating development toolsets to yield a stable software application.

Dr. Joel Henry was tasked under a research and development contract with the DPRA to create a computer program that, when given a list of cargo items and a list of containers, would approximate the number of containers required to pack the items and determine a packing order for the cargo items. When commercially used, a computer trained employee runs the program and prints off the resulting packing list. This list is then given to another employee, who works in a different location and performs the physical packing. This is a tedious and time consuming process. This project will examine a possible software solution that utilizes barcode scanning and wireless LAN technology to streamline the packing process.

The Simulink environment allows rapid prototyping of complex software systems. Because many of these systems are mission-critical, it is of utmost importance to determine their input and output constraints. Determining input constraints is a trivial matter, but the constraint determination of a system's output values is a serious and challenging problem that historically has entailed an exhaustive exploration of the system's input states. The work presented in this thesis recounts and extends a research project supported by NASA whose focus was to develop a strategy to constrain the outputs of a Simulink model. Simulink models are quite similar to mathematical functions and therefore optimization algorithms can be applied to constrain the outputs. Optimizations of simple mathematical functions paved the way for random functions and finally led to the development of two optimization algorithms. During the exploration of potential optimization algorithms, strategies such as Monte Carlo, the simplex method, simulated annealing, and evolution strategy were explored. In the end, a combined approach utilizing both simulated annealing and the simplex method was compared with evolution strategy for relative strengths and weaknesses. It was determined that the evolution strategy algorithm was more suited to optimization of Simulink models due to its more effective usage of model calls and to its higher success rate.

This paper describes how user-centered design methodologies were employed to create a software prototype used for outcomes assessment of e-portfolios. Designed to be used by faculty, university administrators and accreditation users, the prototype standardizes, aggregates and displays e-portfolio data in novel ways. Results of evaluating the prototype indicate that e-portfolios can be an effective means of outcomes assessment at the program level.

The Intergovernmental Panel on Climate Change (IPCC) has estimated between 9 and 88 cm of sea level rise over the next hundred years. Of this, only negative 19 to 11 cm is attributed to the largest ice masses on the planet, the Antarctic and Greenland ice sheets. Over the last decade, dramatic activity in the outlet glaciers of Greenland and the Antarctic Peninsula raise the possibility that these large ice sheets will have a much greater contribution to sea level rise over the next century than was predicted by the IPCC. Recent studies have shown these areas are exhibiting decadal scale changes in response to climate forcings, whereas IPCC models show that ice is not responsive to climate change over such short periods of time. Many believe the IPCC type models fail to show short term climate responses due to the simplifications they make to ice sheet mechanics. Here, we develop a higher-order model -- a new ice sheet model which contains all relevant flow physics. In order to gauge our progress, we perform a verification of our model around a structured set of experiments. The analysis reveals our model is performing well over a range of different scenarios.

(none)

A web-based version of CO2SYS program has been implemented to replace the current DOS based version system. The user does not have to download anything to a local computer, instead they can run the calculations online freely. For this new designed program, all the user inputs and options are displayed in one single window instead of several small black and white DOS screens. All the calculation results are listed in a single page, as well. The user can change any inputs and constants before and after the data calculation, i.e., recalculation. Much more powerful error checking has been built into this web-based system. It also provides useful directions and guidance for the user. The user can get access to the helpful information for each input and constant. Typographical error information, which is listed separately from their individual reference paper, is incorporated with the reference through the hyperlinks. Moreover, this new system presents an attractive and dynamic appearance to users.

The scientific domain presents unique challenges to software developers. This thesis describes the application of design patterns to the problem of dynamically changing interfaces to scientific application software (GLIMMER, which performs ice sheet modeling). In its present form, GLIMMER uses a text configuration file to define model behavior, set parameters, and structure model input/output (I/O). The creation of the configuration file presents a significant problem to users due to its format and complexity. GLIMMER is still under development, and the number of changes to configuration parameters, parameter types, and parameter dependencies makes devel-opment of any single interface of use only for a short term. The application of design patterns described here resulted in an interface specification tool that then generates multiple versions of a user interface usable across a wide variety of configuration pa-rameter types, values, and dependencies. The resulting products have leveraged de-sign patterns and solved problems associated with design pattern usage not found in the specialized software engineering literature.

This thesis examines the difference between accepted theoretical and real world return on investment of requirements, design, and code reviews. The differences have a significant impact on the cost of quality. The goal of this thesis is to examine the differences between two data sets (one derived from widely accepted software principles and one derived from real-world data) and draw conclusions about the effectiveness (cost vs. increase defect detection) of reviews based upon these analyses. This thesis will compare accepted relationships pertaining to cost per defect and overall project cost against actual data from a real world project in order to discover any significant differences. This research will also develop a cost estimation tool than may be used in future research to further develop the ideas and conclusions from this thesis. This author speculates that the cost benefit of reviews will decrease as the amount of time devoted to the review increases. This is contradictory to the accepted project management literature currently in wide use today.