Employer Highlight: Dana Corp
Dana Corp is a leading supplier of drivetrain, chassis, structural, and engine technologies. Our people design and manufacture products for every major vehicle and engine producer in the world. Based in Toledo, Ohio, with operations throughout the world, we are focused on being an essential partner to automotive, commercial, and off-highway vehicle customers, which collectively produce more than 60 million vehicles annually. Our continuing operations reported sales of $8.6 billion in 2005.
Our undergraduates are required to submit a senior thesis. It is a document describing the culmination
of their co-op work. Below are the most recent non-private theses.
Analyzing Software Quality Using Historical Data
In the software industry, data that is collected in the form of defect reports or support calls can be used to analyze the quality of a product. By comparing the data from successive releases, the improvement or decline in quality can be determined. Having hard data to support claims of quality allows the software company to make better decisions by not having to rely on educated guesses.
Rockwell Software has begun to collect defect data weekly to track the defect trends. The only data that is considered is any data after the defect tracking was implemented which contains data only on the current release. All of the existing historical data is ignored by this process. Without historical data, it isn't possible to compare the current release to previous releases to determine whether or not improvements have been made or need to be made. This historical data exists in several databases, it is just a matter of acquiring the data and analyzing it. A process is needed that establishes what data is needed and how to analyze it. It is also necessary to illustrate the benefit this data provides by implementing the process for a product.
The purpose of this thesis is to document the process established to analyze software quality using historical data and the outcomes of that process for the RSLinx Enterprise product. There are several secondary goals to the project: predict future data based on historical data and trends, compare RSLinx Enterprise historical data to industry standard data, reduce support calls and assess the overall quality of RSLinx Enterprise.
There are four main sources of historical data that were used: defect data from Clearquest, customer support data from Clarify, registration from the RSSS database, and the lines of code data. The process created consists of eight steps:
1. Select the product
2. Select the Historical Releases that will provide the data
3. Obtain the version numbers and release dates
4. Obtain the data from the appropriate databases
5. Analyze the data and identify trends
6. Predict data for the next release
7. Compare data to industry standards or internal benchmarks
8. Implement findings and improvements.
As a result of the process, it was determined that the major weakness of RSLinx Enterprise is its usability. The data points to the fact that RSLE is a stable product with good documentation and good support. From the analysis, several ways that the usability can be improved were identified. Improved usability will improve quality and decrease the number of support calls. Decreasing support calls results in savings for the company.
Analysis and Correction of the Calculation Method of the Masterson Index
It was recently discovered in the Masterson Index that some of the index calculations were not producing the expected results. This thesis explains and develops a full analysis of the problem. It offers as a solution the derivation of the Second Generation Chain Formula, which is used to ensure correct calculations.
