From chaos to excellence: The Federal Aviation Administration's use of a business process improvement model

This descriptive case study examined the Federal Aviation Administration's (FAA's) use of the Carnegie Mellon University's Software Engineering Institute's Capability Maturity Model (CUM). The study assessed the FAA's implementation of CMM and measured changes in software quality and workload as business processes were transferred to a Federal government workforce from a private sector workforce.;The hypothesis stated that if CMM-related software business processes were implemented by a government workforce, then software quality and workload outcomes would not diminish. The private sector workforce was performing in a quality manner; therefore, it was considered an accomplishment for the FAA to maintain an equivalent performance level.;A CMM process assessment was used to determine the implementation level of CMM. The assessment data facilitated the development of an improvement plan to increase service quality and, as a by-product, to increase CMM maturity. The dependent variables, software quality and workload, were pre and post tested, then compared to the treatment group. Also, the treatment group's post test data was compared to a control group and the population.;The research verified that CMM, the independent variable, was partially implemented. The dependent variable measures did not diminish below the pre test data of the treatment group or did not diminish below the lowest collective measure of either the control group or the population. Nonetheless, data were insufficient to prove causality. Therefore, the research hypothesis was partially confirmed as CMM key process areas were partially satisfied and the dependent variables did not diminish below the lowest collective measures.;When reviewing the treatment group's performance alone, the dependent variable results were inconclusive and did not confirm a direct correlation between software process improvement and quality improvement. However, the dependent variable results did partially confirm the hypothesis when the treatment group's data was compared to the control group and the population.;Given the descriptive nature of this case study the results were open to alternative explanations. The lack of experimental controls precluded definite conclusions regarding the impact of the FAA's CMM efforts on software quality.