Unfortunately, limited studies consider human factors in automation techniques for construction field information acquisition. Fully utilization of the automation techniques requires a systematical synthesis of the interactions between human, tasks, and construction workspace to reduce the complexity of information acquisition tasks so that human can finish these tasks with reliability. Overall, such a synthesis of human factors in field data collection and analysis is paving the path towards “Human-Centered Automation” (HCA) in construction management. HCA could form a computational framework that supports resilient field data collection considering human factors and unexpected events on dynamic job sites.
This dissertation presented an HCA framework for resilient construction field information acquisition and results of examining three HCA approaches that support three use cases of construction field data collection and analysis. The first HCA approach is an automated data collection planning method that can assist 3D laser scan planning of construction inspectors to achieve comprehensive and efficient data collection. The second HCA approach is a Bayesian model-based approach that automatically aggregates the common sense of people from the internet to identify job site risks from a large number of job site pictures. The third HCA approach is an automatic communication protocol optimization approach that maximizes the team situation awareness of construction workers and leads to the early detection of workflow delays and critical path changes. Data collection and simulation experiments extensively validate these three HCA approaches.
In this paper, we present a Bayesian analysis for the Weibull proportional hazard (PH) model used in step-stress accelerated life testings. The key mathematical and graphical difference between the Weibull cumulative exposure (CE) model and the PH model is illustrated. Compared with the CE model, the PH model provides more flexibility in fitting step-stress testing data and has the attractive mathematical properties of being desirable in the Bayesian framework. A Markov chain Monte Carlo algorithm with adaptive rejection sampling technique is used for posterior inference. We demonstrate the performance of this method on both simulated and real datasets.
We describe mechanical metamaterials created by folding flat sheets in the tradition of origami, the art of paper folding, and study them in terms of their basic geometric and stiffness properties, as well as load bearing capability. A periodic Miura-ori pattern and a non-periodic Ron Resch pattern were studied. Unexceptional coexistence of positive and negative Poisson's ratio was reported for Miura-ori pattern, which are consistent with the interesting shear behavior and infinity bulk modulus of the same pattern. Unusually strong load bearing capability of the Ron Resch pattern was found and attributed to the unique way of folding. This work paves the way to the study of intriguing properties of origami structures as mechanical metamaterials.
Extreme events, a type of collective behavior in complex networked dynamical systems, often can have catastrophic consequences. To develop effective strategies to control extreme events is of fundamental importance and practical interest. Utilizing transportation dynamics on complex networks as a prototypical setting, we find that making the network “mobile” can effectively suppress extreme events. A striking, resonance-like phenomenon is uncovered, where an optimal degree of mobility exists for which the probability of extreme events is minimized. We derive an analytic theory to understand the mechanism of control at a detailed and quantitative level, and validate the theory numerically. Implications of our finding to current areas such as cybersecurity are discussed.
We develop a completely data-driven approach to reconstructing coupled neuronal networks that contain a small subset of chaotic neurons. Such chaotic elements can be the result of parameter shift in their individual dynamical systems and may lead to abnormal functions of the network. To accurately identify the chaotic neurons may thus be necessary and important, for example, applying appropriate controls to bring the network to a normal state. However, due to couplings among the nodes, the measured time series, even from non-chaotic neurons, would appear random, rendering inapplicable traditional nonlinear time-series analysis, such as the delay-coordinate embedding method, which yields information about the global dynamics of the entire network. Our method is based on compressive sensing. In particular, we demonstrate that identifying chaotic elements can be formulated as a general problem of reconstructing the nodal dynamical systems, network connections and all coupling functions, as well as their weights. The working and efficiency of the method are illustrated by using networks of non-identical FitzHugh–Nagumo neurons with randomly-distributed coupling weights.
X-ray tomography has provided a non-destructive means for microstructure characterization in three and four dimensions. A stochastic procedure to accurately reconstruct material microstructure from limited-angle X-ray tomographic projections is presented and its utility is demonstrated by reconstructing a variety of distinct heterogeneous materials and elucidating the information content of different projection data sets. A small number of projections (e.g. 20–40) are necessary for accurate reconstructions via the stochastic procedure, indicating its high efficiency in using limited structural information.