This paper presents the use of subassembly models instead of the entire assembly model to predict assembly quality defects at an automotive original equipment manufacturer (OEM). Specifically, artificial neural networks (ANNs) were used to predict assembly time and market value from assembly models. These models were converted into bipartite graphs from which 29 graph complexity metrics were extracted to train 18,900 ANN prediction models. The size of the training set, order of the bipartite graph, selection of training set, and defect type were experimentally studied. With a training size of 28 parts, an interpolation focused training set selection with a second-order graph seeding ensured that 70% of all predictions were within 100% of the target value. The study shows that with an increase in training size and careful selection of training sets, assembly defects can be predicted reliably from subassemblies' complexity data.

Experiments that study engineering behavior in design often rely on participants responding to a given design prompt or a problem statement. Moreover, researchers often find themselves testing multiple variables with a relatively small participant pool. In such situations multiple design prompts may be used to boost replication by giving each participant an equivalent problem with a different experimental condition. This paper presents a systematic approach to compare given design prompts using a two-step process that allows an initial comparison of the prompts and a post-experiment verification of the similarity of the given prompts. Comparison metrics are provided which can be used to evaluate a level of similarity of existing prompts as well as develop similar problems. These metrics include complexity (size, coupling, and solvability), familiarity, and prompt structure. Statistical methods are discussed for post-experiment verification. Guidelines are provided for a post-experiment survey which may be used for an additional perspective of prompt similarity. The proposed approach is demonstrated using an experiment where two design prompts were used for within-subject replication.

Engineering systems are often represented in terms of their function or the constituent components. Several representation styles and modalities have been proposed and used in the engineering design literature. This study investigates the effects of three different system representation styles on system understanding. Component graph, function graph, and function structure are compared in term of subsystem clustering. A mixed replication experiment is conducted where undergraduate student participants are asked to group or cluster system elements into subsystems. Data collected is analyzed in terms of distances between elements and occurrence of element pairs in the same parent cluster. Distances are measured in terms of connectivity, physical distance, and semantic distance. Results show that participants tend to cluster elements together when they are separated by smaller distances. Connective and physical distances show clear trends with pair occurrence, whereas trends with semantic distance are unclear. Correlations between physical distance and pair occurrence point to differences between the representation styles. Alternatively, deviations from the general trends indicate that strength of influence for each of the distance measures may be contextually dependent. Findings suggest that visual layout of system elements affects system understanding, and we should be cognizant of the implicit information that we may be communicating by our choice of layout. Finally, limitations of the research are discussed, and future work is identified.

Valves are widely used in numerous industries like Beverage, Food, Dairy, Cosmetic, Pharmaceutical and Biotech to serve various purposes. Hence, it is strongly needed that each valve must be tested thoroughly for proper functioning. The equipment used for testing of valves is known as test-rig. At present, no standard test-rig is available in the market for testing of valves. This study is a part of an attempt to develop a cost-effective customized test-rig for multiple valve testing. In present study, pressure-drop in various cross-sections of the proposed.

Introduced is a new physics-based 3D mathematical model capable of efficiently predicting time histories of the nonlinear structural dynamics in cold rolling mills used to manufacture metal strip and sheet. The described model allows for prediction of transient strip thickness profiles, contact force distributions, and roll-stack deformations due to dynamic disturbances. Formulation of the new 3D model is achieved through combination of the highly-efficient simplified-mixed finite element method with a Newmark-beta direct time integration approach to solve the system of differential equations that governs motion of the roll-stack. In contrast to prior approaches to predict structural dynamics in cold rolling, the presented method abandons several simplifying assumptions and restrictions, including 1D or 2D linear lumped parameter analyses, vertical symmetry, continuous and constant contact between the rolls and strip, as well as inability to model cluster-type mill configurations and accommodate typical profile/flatness control mechanisms used in industry. Following spatial and temporal convergence studies of the undamped step response, and validation of the damped step response, the new model is demonstrated for a 4-high mill equipped with both work-roll bending and work-roll crown, a 6-high mill with continuously-variable-crown (CVC) intermediate rolls, and finally a complex 20-high cluster mill. Solution times on a single computing processor for the damped 4-high and 20-high case studies are just 0.37 seconds and 3.38 seconds per time step, respectively.