Adaptive Part Variation: A Near Real-Time Approach to Construction Tolerances

Lauren Vasey, Max Maxwell, Dave Pigram

Research output: A Conference proceeding or a Chapter in BookConference contribution


This chapter introduces the concept of Adaptive Part Variation (APV) as a method where robotically automated fabrication and construction processes employ sensors and feedback to make real-time corrections to material and assembly processes by varying the geometry and location of future parts to respond to deviations between digitally defined and physically accumulating form. The potential disciplinary implications of the method are described followed by a comparison to existing approaches to providing tolerance for dimension error in architecture. As a case study, the material system of cold bending steel rod is utilized to investigate strategies for implementing Adaptive Part Variation within a fabrication workflow that includes the production, handling, and assembly of uniquely bent parts through synchronized robotic tasks and iterative sensor feedback. Two computer vision systems are compared to assess their value for APV processes. Finally, potential shifts in the deployment of procedural design methodologies are discussed in relation to adaptive automated construction processes.
Original languageEnglish
Title of host publicationRobotic Fabrication in Architecture, Art and Design 2014
EditorsWes McGee, Monica Ponce de Leon
Place of PublicationCham, Switzerland
Number of pages13
ISBN (Electronic)9783319046631
ISBN (Print)9783319046624
Publication statusPublished - 2014
Externally publishedYes
EventRobotic Fabrication in Architecture, Art and Design 2014 - University of Michigan, Ann Arbor, Ann Arbor, United States
Duration: 17 May 201418 May 2014


ConferenceRobotic Fabrication in Architecture, Art and Design 2014
Abbreviated titleRob|Arch 2014
Country/TerritoryUnited States
CityAnn Arbor
Internet address


Dive into the research topics of 'Adaptive Part Variation: A Near Real-Time Approach to Construction Tolerances'. Together they form a unique fingerprint.

Cite this