self assembly, thermoelectric, photovoltaic, monte carlo simulation



Additive manufacturing offers substantial flexibility in shape, but much less flexibility in materials and functionality—particularly at small size scales. A system for automatically incorporating microscale components would enable the fabrication of objects with more functionality. This paper considers the potential of self assembly to serve as an automated programmable integration method. In particular, it addresses the ability of random self assembly processes to successfully assemble objects with high performance despite the possibility of assembly errors.


A self-assembled thermoelectric system is taken as a sample system. The performance expectations for these systems are then predicted using modified one-dimensional models that incorporate the effects of random errors. Monte carlo simulation is used to predict the likely performance of self assembled thermoelectric systems and evaluate the impact of key process and system design parameters.


While assembly yield can drop quickly with increasing numbers of assembled parts, large functional assemblies can be constructed by arranging components in parallel to provide redundancy. In some cases, the performance losses are minimal. Alternatively, sensing can be incorporated to identify perfect assemblies. For small assemblies, the probability of perfection may be high enough to achieve an acceptable rate. Small assemblies could then be combined into larger functional systems.


This analysis identifies two strategies that can guide the development of additive manufacturing processes that incorporate miniature components to increase the system functionality. The analysis shows that this may be possible despite significant errors in the self-assembly process because systems may be tolerant of significant assembly errors.

Original Publication Citation

Self Assembly in Additive Manufacturing: Opportunities and Obstacles

Document Type

Peer-Reviewed Article

Publication Date


Permanent URL


Rapid Prototyping Journal




Ira A. Fulton College of Engineering and Technology


Mechanical Engineering

University Standing at Time of Publication

Full Professor