The phenomenon of vibration-assisted entrapment is analyzed in the context of a vibratory parts feeder - the SONY automated parts orienting system (APOS). Entrapment is treated as an energy minimization process. Mechanisms of energy increase and decrease are described. Conditions that lead to failure of entrapment are discussed for a planar part-relief system.

Bacteriophage viruses spontaneously self-assemble in the presence of their component parts (certain protein molecules). It is believed that conformational switches, interacting chemical bonding sites that allow tentative incorrect bonds, facilitate this randomized assembly process. A one-dimensional conformational switch is proposed and used to study the randomized assembly of mechanical parts. A genetic algorithm is used to search the space of parameterized switch designs to maximize the rate of a desired assembly.

A question is posed on how a particular subassembly sequence is generated in randomized assembly. An extended design of mechanical conformational switches [16] is proposed that can encode several subassembly sequences. A particular subassembly sequence is generated due to conformational changes of parts during one-dimensional randomized assembly. The optimal subassembly sequence that maximizes the yield of a desired assembly can be found via genetic search over a space of parameterized conformational switch designs, rather than a space of subassembly sequences. The resulting switch design encodes the optimal subassembly sequence so that the desired assemblies are put together only in the optimal sequence. The results of genetic search and rate equation analyses reveal that the optimal subassembly sequence depends on the initial concentration of parts and the defect probabilities during randomized assembly. The results indicate that abundant parts and parts with high defect probabilities should be assembled earlier rather than later.

A hierarchical approach for nesting two-dimensional shapes based on genetic algorithms is described. For the higher-level search, a representation that facilitates genetic search based on recombination is developed. An alternatiye to overlap computation based on assembly of polygons is used at the lower level of search. Two implementations to find minimum-area enclosures for polygons, with and without a cohstraint on the width of stock, are discussed. Sample output illustrating the effectiveness of the approach is provided.

A survey/test was developed to determine the capabilities, education, and noneducational experiences that contribute to the success of an engineering design student. The survey/test provided measurements of variables that were used to develop regression equations to predict student performance. To predict performance in a design contest, academic class, previous grade point average, family financial status, and perceived competence level were most important. To predict course grade, only previous grade point average was important.>