SEVinS

A few words about the project

SEVinS is a data analysis and decision making tool created for a leading defense industry contractor designed to vastly improve the Earned Value Management System (EVMS) process on large United States government projects. Being able to correctly track earned value is critical when large projects use government funding. NBI developed SEVinS in conjunction with the defense contractor and played a crucial role in solving a problem where data from multiple sources needed to be centrally processed in order to provide the best EV data and control. This system became a distinguishing factor that allowed the defense contractor to win multiple billion-dollar contracts over other large defense players without this system.

The Federal Government required the use of Delteks’ wInsight Analytics, however, it did not have the all of the data needed to satisfy governmental regulations. Data from the master schedule, time entry system, Delteks’ MPM (EVMS Cost Engine), shop floor inventory systems, supply chain procurement and billing systems, and the NBI custom developed Program Log were missing. Armed with in-depth domain and development knowledge, NBI proactively approached the defense contractor’s management on the development and use of an earned value dashboard system (SEVinS) that would identify outliers and exceptions and could easily extract the information needed to influence not only the success of a project, but to insure adherence to all of the governmental EVMS regulations.

SEVinS used an online analytical processing (OLAP) cube that went beyond the Cost Performance Index (CPI) and Schedule Performance Index (SPI). Using extensive domain, mathematics, programming, and database expertise, NBI provided a direct interface from each of the aforementioned disparate systems into a central database using an Extract, Transform, and Load (ETL) process on a single SQL Server. This ETL process required pulling up to one hundred gigabytes of metadata from each of the sources, de-normalizing the data, making inferences, and merging this data into an OLAP cube of approximately one terabyte. Each run was completed in a mere nine minutes, where as previously the data that the defense contractor had been using was a month old.

In addition, to make all of the extremely complex information easy to consume, NBI used a novel approach, pulling from stock analysis development experience, and applied the Bollinger Bands® technique. This technique had never before used within the EVMS community. The use of Bollinger Bands proved revolutionary to EVMS by allowing users to see the whole EVMS health picture holistically eliminating nearly all the pitfalls of traditional EVMS analysis. This ultimately differentiated the defense contractor from all their competitors, resulting in billions of dollars in new sales.