Document Type

Project

Lead Author Type

CIS Masters Student

Advisors

Dr. Gregory Schymik, schymikg@gvsu.edu

Embargo Period

6-9-2018

Abstract

One of the most common challenges that businesses face today is the constant pressure of global competition. Technology is the tool of choice for executing excellence in the marketplace. As a result, the dilemma of needing to deliver software change at an accelerated pace is extremely normal and many organization struggle to identify what approach will work best for them.

At the same time, it is not unusual for humans to jump to conclusions before thoroughly examining a problem and taking in all available facts. After viewing a few symptoms, conclusions are made, action plans are formed and then wonderment arrives when the problem remains unsolved. The intention of this project was to partner with a large organization immersed in this struggle and use Lean Six Sigma as the mechanism for learning why their software delivery is tried and true – but late.

The goal of this specific project is to discover and define performance requirements and associated metrics by employing a problem solving method commonly used in Six Sigma initiatives referred to as the DMAIC method. The result of this research project was a surprising mix of outcomes that speak to the need to balance culture, process, and architecture.

  • The project client did not have any internal service level agreements. When outlining the factors critical to quality, they were confused and struggled to define quality of process. The idea of learning how to improve software change management through the use of metrics was new for them and produced some fear mixed with hopeful excitement.

  • The historical data proved incredibly difficult to acquire and convert into usable form. This was due to the architectural choices made when the software change management system was created. A short-term focus on an incomplete set of use cases rendered the system forever hostile for reporting and analysis.

  • The content of the historical data was minimal in value. There was no data to use for comparing due dates with delivery dates. Effort was not being tracked. The only means of measuring work activities was to measure the time duration within work flow steps.

  • The strength of DMAIC is the discipline of following the data. This helps teams avoid pursuing erroneous leads or tempting short-cuts that ultimately leave the problem unresolved.

    Despite obstacles, the goal of defining software performance metrics was met. This was due to the fact that DMAIC works well as a tool for learning because it is a disciplined approach with a wide array of tools that also serve a need for flexibility under certain conditions.

Share

COinS