Spatiotemporal Particle Swarm Optimization for Future Cost Allocation in Large-Scale Transportation Infrastructure Maintenance

When employing the spatial diagnostics examined in Chapter 4, social scientists will often find evidence of spatial autocorrelation (see, e.g., Eff 2004). As discussed in Chapter 1, this predisposition of social science data toward spatial autocorrelation often results from interdependence between the units studied by social scientists. In other cases, social science data exhibit spatial dependence not as a result of behavioral interdependence but as a consequence of spatial clustering in the sources of behaviors of interest to social scientists. The spatial dependence, in short, may be consistent with either a spatial lag model or a spatial error model. Substantive theory will often lead scholars to believe that a spatial lag specification or a spatial error specification is more appropriate for their particular substantive application. Scholars may, for example, expect that a spatial diffusion process is at work and thus believe that a spatial lag model is warranted. Although such a specification may seem appropriate, such a theoretical expectation should not go untested. It would be inappropriate to estimate a diffusion model with a spatially lagged dependent variable if the spatial dependence diagnosed via, for example, the univariate Moran's I , is instead produced by spatial clustering in the sources of otherwise independent behaviors. This model misspecification will lead the researcher to inappropriate substantive inferences about the nature of the spatial dependence in her data. Inappropriate spatial model specification is all the more problematic because of the close mathematical relationship between a spatial lag model and a spatial error model with spatial autoregressive error dependence. As this chapter will discuss, a spatial autoregressive error model can be rewritten as a spatial Durbin model with both spatially lagged dependent and independent variables if a set of nonlinear common factor constraints are valid. Because of this close relationship between spatial autoregressive dependence in a spatial lag model and spatial autoregressive dependence in a spatial error model, a significant spatial parameter in a spatial lag model may reflect spatial clustering in omitted sources of the behavior of interest rather than true spatial lag dependence consistent with a diffusion process.

Traffic forecasting is a challenging task due to complex spatial and temporal dependencies across sensor locations and time. Interest in solving this task has increased, but current research focuses on manually constructing neural network architectures without the aid of neural architecture search (NAS). In our work, we explore evolutionary neural architecture search (ENAS) by deploying a genetic algorithm (GA) to find optimal neural network architectures for predicting traffic conditions. The search space for the GA consists of arbitrary combinations of dilated convolutions and graph convolutions for modelling temporal and spatial dependencies respectively, limited in complexity only by technical constraints. Experimental results show that model architectures obtained via GA are able to match the current state-of-the-art on traffic prediction benchmarks.

Human Activity Recognition (HAR) is essential for monitoring and analyzing human behavior, particularly in health applications such as fall detection and chronic disease management. Traditional methods, even those incorporating attention mechanisms, often oversimplify the complex temporal and spatial dependencies in sensor data by processing features uniformly, leading to inadequate modeling of high-dimensional interactions. To address these limitations, we propose a novel framework: the Temporal-Spatial Feature Decoupling Unit with Layer-wise Training Convolutional Neural Network (CNN-TSFDU-LW). Our model enhances HAR accuracy by decoupling temporal and spatial dependencies, facilitating more precise feature extraction and reducing computational overhead. The TSFDU mechanism enables parallel processing of temporal and spatial features, thereby enriching the learned representations. Furthermore, layer-wise training with a local error function allows for independent updates of each CNN layer, reducing the number of parameters and improving memory efficiency without compromising performance. Experiments on four benchmark datasets (UCI-HAR, PAMAP2, UNIMIB-SHAR, and USC-HAD) demonstrate accuracy improvements ranging from 0.9% to 4.19% over state-of-the-art methods while simultaneously reducing computational complexity. Specifically, our framework achieves accuracy rates of 97.90% on UCI-HAR, 94.34% on PAMAP2, 78.90% on UNIMIB-SHAR, and 94.71% on USC-HAD, underscoring its effectiveness in complex HAR tasks. In conclusion, the CNN-TSFDU-LW framework represents a significant advancement in sensor-based HAR, delivering both improved accuracy and computational efficiency, with promising potential for enhancing health monitoring applications.

This Long-Term Surveillance and Maintenance Plan describes how the U.S. Department of Energy (DOE) intends to fulfill its mission to maintain protection of human health and the environment at the Amchitka, Alaska, Site1. Three underground nuclear tests were conducted on Amchitka Island. The U.S. Department of Defense, in conjunction with the U.S. Atomic Energy Commission (AEC), conducted the first nuclear test (Long Shot) to provide data that would improve the United States' capability of detecting underground nuclear explosions. The second nuclear test (Milrow) was a weapons-related test conducted by AEC as a means to study the feasibility of detonating a much larger device. The final nuclear test (Cannikin), the largest United States underground test, was a weapons-related test. Surface disturbances associated with these tests have been remediated. However, radioactivity remains deep below the surface, contained in and around the test cavities, for which no feasible remediation technology has been identified. In 2006, the groundwater model (Hassan et al. 2002) was updated using 2005 data collected by the Consortium for Risk Evaluation with Stakeholder Participation. Model simulation results indicate there is no breakthrough or seepage of radionuclides into the marine environment within 2,000 years. The Amchitka conceptual model is reasonable; the flow and transport simulation is based on the best available information and data. The simulation results are a quantitative prediction supported by the best available science and technology. This Long-Term Surveillance and Maintenance Plan is an additional step intended for the protection of human health and the environment. This plan may be modified from time to time in the future consistent with the mission to protect human health

In general, the long-term maintenance planning of residential buildings is performed based on uniform repair times. However, in fact, various factors, such as the quality and user patterns, affect the performance of residential building components in the Operation and Maintenance (O&M) phase. Hence, various residential building components are repaired at uncertain times, acting as a risk for the residential building maintenance plan. Therefore, an efficient maintenance plan should be established considering maintenance uncertainty. In this regard, this study aims to analyze the uncertainty of repair times for various finishing works in residential buildings based on a probabilistic methodology and outline the implications for the establishment of efficient maintenance strategies in these buildings. Hence, 47,344 repair data for 63 buildings in 12 public residential building complexes completed between 1991 and 2001 in the Republic of Korea were used for analysis. Before the analysis, a repair time matrix was constructed by classifying the finishing works in 25 types and setting service life times to 6–26 years. The repair time distribution for each finishing work was then derived. Results confirmed that basic repair time setting can be performed and various information for reasonable maintenance decision making regarding each finishing work can be provided through a probabilistic approach. The probabilistic approach can be used as a critical decision-making method because there is uncertainty associated with the repair time of each finishing work owing to the performance degradations of various finishing works due to complex causes. Although this study focused on repair time owing to data collection limitations, maintenance strategies with strategic flexibility can be established by developing probabilistic methods that simultaneously consider frequency and cost by securing additional high-quality cost data.

Although there is already a great amount of scientific literature dealing with the use of building information modeling (BIM) in engineering activities, the majority refer to successful case studies using the usual methods and technology of building construction but rarely bring up the real problems for implementing BIM methodology to the field of transportation infrastructure. It must be also considered that the construction activity is only a part of the infrastructure life and the stakeholder must consider the works of enlargement, renewal, and maintenance of the infrastructure. The purpose of this paper is not only to show a mere review of the existing literature but also present a rational analysis for the use of BIM in different areas of civil engineering. For that purpose, the gathered experience in the use of BIM in civil engineering projects in the final course of Civil Engineering Master Studies in the Civil Engineering School (ETSICCP) at Universidad Politécnica de Madrid were compared with the reported literature. This way, a complete and updated information regarding tendencies, applications, and practice along with limitations and benefits can be presented. The significance of this research relies on the original insight of BIM for civil engineering applications through four case studies. Two of them were focused on construction possibilities and the other two on the possibilities in the exploitation, rehabilitation, and maintenance. The results showed that despite the lack of previous experiences, the use of BIM methodology is possible for activities such as maintenance, managing, or expansion of infrastructure by applying different specific software packages. Among the main problems needing to be addressed are the following: handling of big data files, the integration of new data non-related with the modeled object, and interchange of data without losing information. That proves the need of new more efficient techniques to overcome the challenge of the full use of BIM in the civil engineering field and obtain the mutual advantage of the co-operation of the academic and industrial worlds.

Dynamic brain networks are more effective than static networks in characterizing the evolving patterns of brain functional connectivity, making them a more promising tool for diagnosing neurodegenerative diseases. However, existing classification methods for dynamic brain networks often rely on sliding windows to extract multi-window features, leading to suboptimal performance due to the spatio-temporal coupling on these windows and limited ability to effectively integrate complex topological features. To address these limitations, we propose a novel method called Confidence-Driven Dynamic Spatio-Temporal Convolutional Network (CD-DSTCN). First, our proposed method employs a spatio-temporal convolutional network integrated with a temporal attention mechanism to extract spatio-temporal features within each window. By propagating information across temporal windows during spatial convolution, the method effectively captures and integrates complex temporal and spatial dependencies. Second, each window generates an output probability, which quantifies prediction confidence based on the true class probability (TCP). This confidence score serves as a weight to assess the relative importance of different time windows. Finally, the confidence-weighted fused features are passed through a multilayer perceptron (MLP) for final classification. Extensive experiments on Alzheimer's and Parkinson's datasets show that the proposed method out-performs the state-of-the-art algorithms and can provide valuable biomarkers for brain disease diagnosis. Our code is publicly available at: https://github.com/YNingCode/CDDSTCN.

As one of the most widely used applications of the Internet of Things (IoT), intelligent transportation system (ITS) is of great significance for urban traffic planning, traffic control, and traffic guidance. However, widespread traffic congestion occurs with the increased number of vehicles. The traffic flow prediction is a good idea for traffic congestion. Therefore, many schemes have been proposed for accurate and real-time traffic flow prediction, but there still exist many issues, including low accuracy, weak adaptability and inferior real-time. Meanwhile, the complex spatial and temporal dependencies in traffic flow are still challenging. To address the above issues, we propose a novel spatial-temporal Chebyshev graph neural network model (ST-ChebNet) for traffic flow prediction to capture the spatial-temporal features, which can ensure accurate traffic flow prediction. Concretely, we first add a fully connected layer to fuse the features of traffic data into a new feature to generate a matrix, and then the long short-term memory (LSTM) model is adopted to learn traffic state changes for capturing the temporal dependencies. Then, we use the Chebyshev graph neural network (ChebNet) to learn the complex topological structures in the traffic network for capturing the spatial dependencies. Eventually, the spatial features and the temporal features are fused to guarantee the traffic flow prediction. The experiments show that ST-ChebNet can make accurate and real-time traffic flow prediction compared with other eight baseline methods on real-world traffic data sets PeMS.

CPNet: Conditionally parameterized graph convolutional network for traffic forecasting

In this paper, the focus is on the role of the municipality, as an enabler of a collaboration between freight forwarders and the municipality in which the consolidation of goods is considered as a means for goods flow improvement in urban freight transportation. We present a cost allocation model that is based on solution concepts from cooperative game theory, for allocating the operational costs associated with the collaboration. It is assumed that the municipality is willing to carry some cost to ensure a stable collaboration for the potential benefits received, e.g., reduced traffic congestion in the city. The model is applied to some illustrative examples, and the cost allocation results are discussed. It is shown that the role of the municipality may be decisive in achieving a stable collaboration between the freight forwarders, and further that the municipality does not necessarily need to contribute to covering the costs.

A novel evaluation of heat-electricity cost allocation in cogenerations based on entropy change method

The growing number of interconnected systems in consumer IoT environments and the increasing intelligence of AI-enabled cyber threats have introduced significant security challenges. Intrusion detection systems must be not only accurate but also resilient to evolving threats and robust against adversarial manipulation. A major limitation is the lack of labeled data for evolving adversarial attacks, which limits the generalization of detection models, making them less practical for the dynamic and heterogeneous nature of IoT networks. To overcome these limitations, we propose, a hierarchical graph neural network designed for accurate and resilient network intrusion detection using few labeled examples. utilizes a packet-level graph attention network for adaptive feature filtering, which enhances resilience to subtle data perturbations, coupled with a flow-level temporal model to capture complex spatial and temporal dependencies in traffic patterns. By hierarchically abstracting these features, it learns robust representations of normal and malicious network behavior from limited data. Experimental results on multiple widely used network intrusion datasets demonstrate the ’s remarkable data efficiency and resilience. With as little as 5% of labeled training data, achieves an F1-score of 0.97 and an accuracy of 0.98 across all the benchmarks. Even in extreme data scarcity, using just 0.1% of labeled data, it maintains a high F1-score of 0.94+ and an accuracy of 0.96+. Furthermore, we demonstrate the model’s resilience by evaluating it against common adversarial evasion attacks, such as the Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD). Under these attacks, ś accuracy only degrades minimally to 0.95, outperforming standard deep learning baselines whose performance drops below 90% under similar conditions. These results validate as a scalable, resilient, and practical method for efficient network intrusion detection with high accuracy and low false positives, even in data-scarce and adversarial IoT environments.

This paper builds a generic modeling framework for analyzing the edge-creation process in dynamic random graphs in which nodes continuously alternate between active and inactive states, which represent churn behavior of modern distributed systems. We prove that despite heterogeneity of node lifetimes, different initial out-degree, non-Poisson arrival/failure dynamics, and complex spatial and temporal dependency among creation of both initial and replacement edges, a superposition of edge-arrival processes to a live node under uniform selection converges to a Poisson process when system size becomes sufficiently large. Due to the convoluted dependency and non-renewal nature of various point processes, this result significantly advances classical Poisson convergence analysis and offers a simple analytical platform for future modeling of networks under churn in a wide range of degree-regular and -irregular graphs with arbitrary node lifetime distributions.

This paper focuses on analysis of spatiotemporal binary data with absorbing states. The research was motivated by a clinical study on amyotrophic lateral sclerosis (ALS), a neurological disease marked by gradual loss of muscle strength over time in multiple body regions. We propose an autologistic regression model to capture complex spatial and temporal dependencies in muscle strength among different muscles. As it is not clear how the disease spreads from one muscle to another, it may not be reasonable to define a neighborhood structure based on spatial proximity. Relaxing the requirement for prespecification of spatial neighborhoods as in existing models, our method identifies an underlying network structure empirically to describe the pattern of spreading disease. The model also allows the network autoregressive effects to vary depending on the muscles' previous status. Based on the joint distribution derived from this autologistic model, the joint transition probabilities of responses among locations can be estimated and the disease status can be predicted in the next time interval. Model parameters are estimated through maximization of penalized pseudo-likelihood. Postmodel selection inference was conducted via a bias-correction method, for which the asymptotic distributions were derived. Simulation studies were conducted to evaluate the performance of the proposed method. The method was applied to the analysis of muscle strength loss from the ALS clinical study.

A dynamic network model for imbibition based on a physically realistic and mathematically rigorous treatment of the complex dynamics of frontal displacements, film swelling, snap-off and snap-off nucleated displacements is described. Many of the previously reported observations in micro-model displacements, X-ray CT measured saturation profiles in core floods and the competition between frontal displacements and snap-off all arise as a natural consequence of the dynamics of the model. The model is used to study the effect of displacement rate on imbibition patterns, relative permeabilities and residual saturations. The model shows that at very low displacement rates the displacement is dominated by film swelling and snap-off with a displacement pattern similar to bond percolation. Trapping is extensive and relative permeabilities are low. At high displacement rates snap-off is completely suppressed and the displacement pattern is similar to invasion percolation. Trapping is low and relative permeabilities are high. Introduction Quasi-static network models based on realistic rock topology, geometry and pore-scale physics have been used to predict relative permeabilities and residual saturations for displacements in simple porous media1,2,3,4,5. A major limitation of these models in imbibition displacements is the treatment of wetting film or corner flow and snap-off. In imbibition the wetting fluid advances by piston- like displacements of pores and throats from a connected invasion front and by film swelling and snap-off ahead of the connected front. Snap-off is the dominant trapping mechanism and the competition between snap-off and frontal piston-like pore displacements determines the displacement pattern, residual saturation and relative permeability. Quasi-static models assume that the displacement is dominated by capillary forces and neglect viscous pressure gradients in bulk fluid and in films. Neglecting viscous losses in bulk fluid is a reasonable assumption for capillary numbers Ca<10−3-10−4 (Dullien6), where the capillary number is defined asEquation 1 and µ is the viscosity of the displacing fluid, v is the displacement velocity and s is the interfacial tension. However, this is not true for film flow where the small conductivity of films can result in significant rate effects at capillary numbers as low as 10−6-10−8 (Blunt and Scher7 and Hughes and Blunt8). Displacement rate has a profound effect on the competition between piston-like displacements and snap-off and therefore on the displacement pattern, residual saturation and relative permeability. A number of dynamic network models have been proposed to account for rate effects in imbibition7,8,9,10,11,12,13,14,15. All of these models involve the computation of pressure fields based on simplifying assumptions regarding the nature of wetting films and film flow (constant film conductivity, steady-state flow, lumped hydraulic resistances for film and bulk flow). The computed pressure fields are used to modify the order of pore and throat displacements. None of the models fully capture the complex spatial and temporal dependence of film swelling, snap-off and snap-off nucleated pore and throat displacements ahead of the connected front.