Comparing Functional Analysis Methods for Product Dissection Tasks

Joran W. Booth, Tahira N. Reid, Claudia Eckert, Karthik Ramani

Research output: Contribution to journalArticlepeer-review

32 Scopus citations


The purpose of this study is to begin to explore which function identification methods work best for specific tasks. We use a three-level within-subject study (n = 78) to compare three strategies for identifying functions: energy-flow, top-down, and enumeration. These are tested in a product dissection task with student engineers who have minimal prior experience. Participants were asked to dissect a hair dryer, power drill, and toy dart gun and generate function trees to describe how these work. The function trees were evaluated with several metrics including the total number of functions generated, the number of syntactical errors, and the number of unique (relevant and nonredundant) functions. We found no statistical, practical, or qualitative difference between the trees produced by each method. This suggests that the cognitive load for this task for novices is high enough to obscure any real differences between methods. We also found some generalized findings through surveys that the most difficult aspects of using functional decomposition include identifying functions, choosing function verbs, and drawing the diagram. Together, this may also mean that for novice engineers, the method does not matter as much as core concepts such as identifying functions and structuring function diagrams. This also indicates that any function identification method may be used as a baseline for comparison between novices in future studies.

Original languageEnglish (US)
Article number081101
JournalJournal of Mechanical Design
Issue number8
StatePublished - Aug 1 2015

All Science Journal Classification (ASJC) codes

  • Mechanics of Materials
  • Mechanical Engineering
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design


Dive into the research topics of 'Comparing Functional Analysis Methods for Product Dissection Tasks'. Together they form a unique fingerprint.

Cite this