Case Study: Applying Agile Software Methods to Systems Engineering

shane-rounce-1ZZ96uESRJQ-unsplash

Posted: February 10, 2016 | By: Matthew Kennedy

Framework Implementation Example

During the Increment planning phase Juggernaut designated their Sprints as hardware, mechanical, software and / or firmware sprints.  Each Integration Phase required input from a predetermined number of sprints and / or external dependencies (designs, hardware components, COTS products, etc.).  The framework implementation example from the case study requires input from three internally developed Sprints, though other Integration Phases may require both internal and external inputs to achieve the specified objectives.  In this example details have been removed such as the actual specification requirements and additional details required to adequately test the exit criteria for instance required response times and detailed accuracy ranges.  Names of the customer point of contact are also not included.  The example is intended to provide the overall organization Juggernaut used to structure their systems development using the Agile Systems Engineering Framework.  A graphics depiction can be seen in Figure 6.

Integration 1: Wired Functionality

Objectives

  • Decision on the Solar Panel / Internal Battery concept
  • Metrology meets specified tolerance ranges
  • Hardware Interface meets current profile requirements
  • Initial wire testing is complete with list of defects
  • Validation of Test Point access (Test Engineering)

Input Criteria:

Sprint 1 – Hardware

Input Criteria: 

  • Hardware Specification Document

Exit Criteria: 

  • Support multiple energy sources
  • Discrete solution for the register interface

Sprint 2 – Mechanical

Input Criteria: 

  • Design Specification Document

Exit Criteria:

  • Rolled Sealed Assemblies
  • Housings According to Specification

Sprint 3 – Firmware 

Input Criteria: 

  • Firmware Specification Document

Exit Criteria:

  • Wire Communication Standard Conformance
  • Liquid Crystal Display (LCD) Functionality
  • LCD Activation Level
  • Metrology Functionality

Exit Criteria:

  • Current Consumption / Profile
  • Wire Interface HW Testing
  • Primary Power Supply Capabilities
  • LCD Testing
  • E&M Field Testing
  • Electrostatic Discharge Testing
  • LCD Activation Level (multi-source)Wire Response
  • Clock Detection Accuracy
  • ASIC Read
  • Operational Compliance Testing
  • Environmental Testing

Project Results

For case study tracking purposes, the project was divided into three critical milestones: Milestone 1 marked the successful completion of internal QC; Milestone 2 was the release to a limited user base and external testing; and Milestone 3 was mass production and customer sales.  The metrics for this case study focus on the first milestone.

image6

Figure 6 Example Framework Implementation

Completion of Milestone 1 was the focal point of this case study as it included design, development and internal QC of the product.  Juggernaut produced seventy units during this phase to put through internal QC.  After the successful completion of the first milestone, the product specifications were sent to several production plants in various countries for manufacturing resulting in 500 Low Rate Initial Production (LRIP) units.  After receiving the results of the user feedback and QC tests, Juggernaut would perform an assessment to determine if the units were ready for production (thus concluding Milestone 2), then move into Milestone 3, mass production and customer sales.

The completion of Milestone 1 was scheduled for 27.5 weeks and the actual completion took 29 weeks.  This was a 5.5 percent increase in duration from the initial estimate.  When compared to the past performance of Juggernaut, there was a 24.5 percent improvement in predicting their schedule using the Agile Systems Engineering Framework and Practices.

Since Juggernaut had been developing similar systems for years, their cost estimation was typically accurate prior to using the Agile Systems Engineering Framework and Practices.  Throughout the project, the teams were assigned a “cost goal” based on the overall cost estimate.  These cost goals were then used by the teams to make tradeoffs throughout the project.  The teams were able to meet their cost goal by balancing cost factors such as scope, material costs or labor.  On this project, the largest contributor to the under run in budget was that one of Juggernaut’s vendors agreed to decrease their profit margin resulting in a decrease in overall project cost.  Based on Juggernaut’s past performance metrics, Juggernaut typically ran 2.5 percent over budget.  At the completion of Milestone 1, Juggernaut was 5 percent under budget marking a 7.5 percent difference in cost estimation.  Because the cost was due to a vendor renegotiation, the cost fluctuation was not attributed to the Framework or Practices utilized during product development.  Without the vendor renegotiation the cost savings was estimated to be less than 1 percent.

Typically Juggernaut experienced a decrease in 5 percent of the planned functionality in order to better meet their cost and schedule goals.  At the completion of Milestone 1, they delivered 100 percent of the planned functionality showing an overall improvement of 5 percent.

Past Performance New Model Result
Cost Difference from Estimate1 N/A N/A N/A
Schedule Difference from Estimate +30% +5.5% 24.5% Improvement
Functionality difference from planned -5% 0% 5% Improvement

Table 3 Case Results Data

The benefits of the Framework and Practices were seen outside of the engineering division and stretched into marketing.  The increase in predictability of delivery dates allowed the marketing division to better plan for the marketing aspect of the product.

Juggernaut successfully completed Milestone 2 and produced 4000 units.  Approval for mass production was achieved and is currently underway.

Want to find out more about this topic?

Request a FREE Technical Inquiry!