An Investigation into Subject Matter Expert Elicitation in Cost Risk Analysis

Abstract

This research compares the efficacy of subject matter expert (SME) elicitation methods to other cost estimation methods using a development and production dataset provided by AFLCMC/FZC. First, by using descriptive statistics to evaluate low versus high amount of the respective cost estimation methods by analyzing the means of percent cost growth for both groups. Next, this research involved using a statistics-based approach to investigate whether SME based cost estimating methods have an associated relationship to percent change of Program Acquisition Unit Costs (PAUC), which will be our proxy variable to cost growth. Using a pooled cross-sectional OLS regression analysis model with adjusted R2 of 0.298, 144 POEs sample for development have statistical evidence to support SME based cost estimates have a positive association with Program Acquisition Unit Cost (PAUC). Lastly, this research critically examines SME elicitation methods used within DoD and provides best practices used by industry and academia when eliciting SMEs that the cost estimating community should consider implementing.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 26, 2020
Accession Number
AD1101597

Entities

People

  • Christopher W. Miller

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Ground and Sea Platforms
  • Human Systems
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Aircrafts
  • Best Practices
  • Business Administration
  • Cost Analysis
  • Cost Estimates
  • Data Science
  • Databases
  • Delphi Method
  • Descriptive Analytics
  • Electronically Scanned Array
  • Information Science
  • Information Systems
  • Knowledge Management
  • Regression Analysis
  • Statistical Analysis
  • Statistical Inference

Readers

  • Logistics and Supply Chain Management.
  • Regression Analysis.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.