Limit state earned value analysis impact on network security project management decisions

Abstract

Limit State Earned Value Management (LSEVM) has increased project managers' abilities to identify network security project performance issues and select appropriate corrective actions. Limit State Earned Value Management (LSEVM) is an evolution of traditional Earned Value Management (EVM) that incorporates active management through prescriptive interpretation of performance indicators that provides insight to the development of corrective actions. Traditional EVM-analysis does not have the capacity to identify all possible performance states. Our study indicates that project managers struggle to identify project performance and select appropriate corrective actions from project analysis alone. Limit State EVM solves these issues by re imagining the role of EVM-analysis around the identification of project performance and not simply the computation of mathematical variances. The LSEVM solution is built on: a) Inspection of project S-curves, revealing that traditional EVM-analysis does not capture all project performance states. b) Enhancement of EVM-analysis methodology by adding a Budget Variance (BV) to capture missing performance states. This variance is defined as the difference between planned value and actual cost. From this, we can create a mathematical relationship between the EVM variances CV, BV, and SV. c) Demonstrating that the number of total potential project performance states captured by the addition of this BV is 27: three variances with three results each (+,0,−). d) Reducing the number of performance states to only those 13 combinations of (SV, CV, BV) that satisfy the mathematical relationship above. e) Grouping the 13 performance states by common failure modalities to define 7 Limit States. f) Mapping the failure modalities to a suite of appropriate corrective actions. In the Fall of 2012, an introduction to Limit State EVM was added to a PMP® Exam prep course. A statistical analysis was performed to determine if the learning gains were statistically significant or a matter of chance. Eleven (11) assessment topics were presented to 258 corporate project managers as part of a pre-test and post-test assessment instrument using a paired, two tailed t-test with a confidence interval of 95% (P=.025). The p-values indicated that the mean learning gains were statistically significant in every category. Much larger gains were observed in the applied EVM questions and met or marginally met the threshold of competency applying EVM to real world problems. LSEVM significantly enhances a project manager's ability to identify network security project performance and select appropriate corrective actions. The project managers studied were able to make better data-driven decisions.