Journal of Infrastructure Systems

With a plethora of major capital and regional centers worldwide being subjected to increasing population levels, there is a need to upgrade and expand the capacity and performance of their urban rail transit (URT) networks and systems. Fiscal constraints, however, have hindered many governments’ ability to fund the construction and operation of URT projects. While there have been numerous studies that have examined the use of public–private partnerships (PPPs) to deliver URT, there has been a paucity of research that has critically reviewed, analyzed, and compared developments in this important and emergent area of transportation procurement. In addressing this void, the authors undertake a state-of-the-art review by performing a systematic scientometric analysis of the extant URT PPP literature to determine gaps in knowledge and identify implications for policy and areas for future research. As a result of performing a state-of-the-art review, the contributions of this paper are twofold: (1) provide the foundations for developing a theoretical framework that can be used by policy-makers to ensure URTs are resilient and adaptable to change; and (2) identify the need for research to examine how land value capture can be used to finance mechanism to ensure the viability of PPPs.

In recent decades, closed circuit television (CCTV) has been the most used tool for visually inspecting the internal condition of pipelines. However, CCTV inspection requires long videos to be observed and analyzed by certified inspectors, which is time-consuming, labor-intensive, and error-prone. This paper proposes a novel approach for automated anomaly detection and localization in sewer CCTV inspection videos. The developed algorithms employ three-dimensional (3D) Scale Invariant Feature Transform (SIFT) to extract spatio-temporal features in sewer CCTV videos. Anomaly detection is performed using a one-class support vector machine (OC-SVM) trained by frames without defects to model states considered normal and to classify outliers to this model as anomalous frames. Then, the identified anomalous frames are located by recognizing included text information in them using an end-to-end text recognition approach. The proposed localization approach is divided into two main steps: text detection using maximally stable extremal regions (MSER) algorithm and text recognition using a deep convolutional neural network (CNN). Extracting and localizing the suspicious frames out of these videos for further analysis can reduce the time and cost of detection because thousands of normal frames would be detached in the inspection process. The proposed model performance showed acceptable viability, because the testing accuracy was 92.3% in anomaly detection and 86.6% for frame localization in sewer inspection video frames.

In the event of a strong earthquake, the motorway administrator will likely have to interrupt network operations to inspect potentially damaged bridges. Although continuing operation without inspection may be dangerous for motorway users, unnecessary interruption may have adverse consequences, especially with respect to rescue operations. This calls for development and implementation of a RApid REsponse (RARE) system, which will facilitate rationalized decisions. Such system requires real-time assessment of the seismic damage of motorway infrastructure, including which bridges are arguably the most vulnerable. This study addresses the issue by combining nonlinear finite element (FE) simulations with advanced econometric modeling. Based on a published simplified modeling approach and classification schemes, simplified FE models of characteristic bridge categories are developed, accounting for key structural components (pier, deck, abutment bearings, abutment stoppers) and soil-structure interaction. Employing the three-stage least squares (3SLS) approach, the data from the FE analyses are used to develop a relationship among the seismic damage (using the maximum and residual drift ratio, and the ratio of maximum ductility demand over ductility capacity as damage indices) of the bridge and the statistically significant intensity measures. The proposed 3SLS approach accounts for (1) both simultaneous equation bias and cross-equation contemporaneous correlation of the disturbances (error terms) caused by shared unobserved effects across the damage indices; (2) endogeneity among the damage indices with the use of instrumental variables; and (3) unobserved heterogeneity and panel effects, through the use of fixed effects. The 3SLS approach is compared to a traditional ordinary least squares (OLS) regression, and the comparison depicts the superiority of 3SLS in terms of explanatory power and forecasting accuracy.

The measurement of road roughness is important for the management of economic road maintenance. Not only is it an indicator of road condition and ride quality, but it also is used to determine road use costs, including travel time, fuel consumption, and vehicle maintenance. Because of the importance of roughness for road asset management decision-making, road agencies spend considerable resources trying to measure road roughness in a repeatable and reproducible manner. However, many road agencies with large road networks are unable to record the condition of the entire network on a sufficiently frequent basis to determine adequately road condition to make informed preventative maintenance decisions. To address this, research has been carried out to develop low cost smartphone based technologies fitted inside vehicles to measure road condition. The trial of these systems has met with varying degrees of success. This paper presents an in-depth parametric study carried out using state-of-the-art vehicle dynamics software, informed by a review of the literature, to appreciate how and to what degree various influencing variables might affect roughness measurements using a smartphone fitted to a moving vehicle. These variables included the type and position of the smartphone; the type, speed, mass, dynamic response, suspension system, and tire pressure of the vehicle in which the smartphone is fitted; and the longitudinal road profile. The results of the parametric analysis were used to build multivariate linear regression and machine learning algorithms which predict road roughness from a measure of a vehicle’s vertical acceleration taking into account the predominant influencing variables. The multivariate linear regression equations can be used to predict road roughness with a similar degree of accuracy that is expected from a visual inspection. On the other hand, the machine learning algorithms, when suitably trained, were able to estimate reliably the road roughness on an integer-based rating scale at a level of detail which is suitable for strategic road asset management, provided that the vehicle type and speed and the type of smartphone are taken into account.

This study compares greenhouse gas (GHG) emissions, embodied energy, and air pollutant emissions of alkali-activated mortars and conventional portland cement (PC)-based mortars. Alkali-activated materials (AAMs) do not require the use of PC to offer cementitious properties; these materials can valorize industrial waste streams and noncementitious natural resources. In this work, several AAMs containing blast-furnace slag and natural pozzolans were examined. Comparisons were drawn both based on the production on $1{\mathrm{m}}^3$ of material and based on ratios of GHG emissions to mortar strength. To facilitate robust assessments, mechanical and material properties were determined. GHG emissions, embodied energy, and nitrogen oxides (${\mathrm{NO}}_x$), sulfur oxides (${\mathrm{SO}}_x$), carbon monoxide (CO), and lead (Pb) emissions for the alkali-activated mortars were lower than their conventional counterparts. However, the AAMs exhibited higher volatile organic compound (VOC) and particulate matter 10 microns or smaller (PM₁₀) emissions. When ratios of GHG emissions to strength were examined, results indicated that the lower environmental impacts of AAMs could be desirable relative to PC mortars, even when the AAMs displayed lower mechanical strength. These findings suggest that, depending on application, AAMs could contribute to environmental impact-mitigation strategies.

Changes in weather patterns pose a threat to the serviceability and long-term performance of roads, and porous asphalt (PA) roads are particularly sensitive to the freezing-thawing (FT) phenomenon. The main objective of this research is to assess the impact of climate change, particularly freezing and thawing cycles, on PA. Climate models predict changes in air temperature, not pavement temperature. To predict the climate change impact on pavements performance, this requires first establishing a relationship between air and road temperature and a correlation between pavement performance and FT cycles. This project focuses on the Netherlands, where PA pavement use has become mandatory, and recent severe winters have increased the discussion about the cold weather performance of porous asphalt and the potential challenges of changing winter weather patterns. When considering long-term changes in climate, the cost impacts of freeze–thaw on PA pavement are predicted to vary regionally and in most areas reach a point in the middle of the century when a reactive wait-and-see approach is more advantageous than proactive adaptation. Further research is suggested to refine the relationship between observed damage and freeze–thaw impacts on PA pavement.

Over the last three decades, public–private partnership projects have emerged as strong alternative service delivery mechanisms to conventional public sector initiatives. The results, however, have been a mix of failures and successes. The determinants of success and failure of such arrangements have generated much interest among stakeholders and the research community, but the combinations of factors that influence the success of public–private partnership projects are relatively less researched. This article investigates the relationships among the complex factors that influence the success of such arrangements, using urban drinking water projects in India as an example. Eighteen projects were selected for the study and 11 factors affecting the project outcome (causal conditions) were identified based on the literature review. Fuzzy set qualitative comparative analysis was conducted to understand which combinations of causal conditions potentially influence the outcome of a public–private partnership project. The analysis identified five causal solutions comprising seven to nine causal conditions each. Defining service standards in a manner that is amenable to measurement and public disclosure emerged as a necessary condition. Two causal solutions presented themselves with capacity of the public sector, and its consequent outputs, as a primary theme. Two other solutions have elements relating to appropriate project structure and baseline information as the key causal conditions. These causal solutions indicate the interrelationship between various factors that influence project performance and provide pointers for project proponents to focus on the causal conditions over which they have relatively better control, when configuring public–private partnership projects.

Designing effective recovery strategies for damaged networks is important across built, human, and natural systems. Postperturbation network recovery has been motivated by two distinct philosophies, specifically, the use of centrality measures in complex networks versus network optimization measures. The hypothesis that hybrid approaches may offer complementary value and improve our understanding of recovery processes while informing real-world restoration strategies has not been systematically examined. This research shows that the two distinct network philosophies can be blended to form a hybrid recovery strategy that is more effective than either. Network centrality–based metrics tend to be intuitive and computationally efficient but remain static irrespective of the desired functionality or damage pattern. Optimization-based approaches, while usually less intuitive and more computationally expensive, can be dynamically adjusted. The proposed approach, based on edge recovery algorithms with edge importance informed by network flow and node attributes, outperforms recovery informed exclusively either by network centrality or network optimization. We find that optimization methods outperform centrality-based approaches for networks that are large enough for the power law to be manifested, but for treelike networks typically found at smaller scale, the two approaches are competitive and scenario specific.

Water distribution and transportation systems are geospatially colocated, forming a network of connections. This network of connections is referred to as an interface network. Investigation of interface network connectivity can help understand and minimize failure propagation from water to transportation systems. Water distribution–transportation interface networks consist of nodes, which can be either pipes or roads, and edges, which represent the geospatial colocation of a pipe and road. The purpose of this study is twofold: to topologically represent geospatial colocation by characterizing the connectivity of water distribution–transportation interface networks for multiple cities, and to identify the nodal attributes that are most predictive of a given connectivity profile. A total of forty interface networks from eight cities of varying geospatial morphology are extracted and analyzed using network analysis and machine learning. Using network analysis, we investigate whether the topological connectivity between water and transportation is consistent across different cities. Then we use a random forest model to ascertain which nodal attributes may have predictive power to identify the connectivity cluster of the city to which a node belongs. The results indicate that cities of different geospatial morphology may vary in their interface network connectivity, and the average shortest path length of a given node is the major nodal feature contributing to a given city’s interface network connectivity. These findings hold implications for urban planning and water distribution design to mitigate potential cascading failures.

Designing infrastructure for a changing climate remains a major challenge for engineers. In popular discourse a narrative has emerged that infrastructures are likely underdesigned for the future. Weather-related hazards are directly embedded in the infrastructure design process. Yet the codes and standards that engineers use for this risk analysis have been changing for decades, sometimes increasing and other times decreasing design values. Further complicating the issue is that climate projections show increasing or decreasing intensities depending on the hazard and region. Thus, it is not clear that infrastructure is universally underdesigned. Here, analyses are developed at both regional and national scales using precipitation and roadway drainage systems to answer this question. First, it is shown that modeling uncertainty can pose challenges for using future projections to update region-specific standards. Second, the results show that depending on the historical design conditions and the direction of projections, roadway drainage infrastructures may be designed appropriately in some regions while in others they are possibly underdesigned. Given these uncertainties, the authors believe that there is a need for alternative design paradigms, and these needs are discussed.

To address the rapidly growing need for asset management of aging water distribution systems (WDS), this study introduces the adaptation of the triple top line (TTL) as a tool for WDS sustainability evaluation. The WDS TTL model is tested using EPANET hydraulic simulations of a hypothetical WDS subjected to multiple aging scenarios. The results of one aging scenario showed the sustainability index (SI) over the long term to be 0.52 for WDS with new pipe versus 0.23 for a WDS with old pipe, confirming the expectation of age influencing the SI. The testing showed pipe breakage rate to be the major contributing factor significantly decreasing economic sustainability. Additional analysis using the TTL model showed that replacing a pipe at an age of 70 years would maximize the sustainability measure of the WDS. Sensitivity analysis further illustrated that pipe aging, demand increase, water price reduction, and pipe breakage all acted to reduce sustainability, confirming the ability of the TTL model to provide useful insight into sustainability for planning, design, and operations comparative analyses.

The current study addresses the issue of vulnerability assessment of transportation systems to climate change, with the purpose of improving current indicator-based approaches in this field. It builds upon other efforts and seeks to enhance rationality in the decision-making process by minimizing arbitrary choices. The study focuses on coastal Massachusetts, which is affected by sea level rise (SLR) and storms and, thus, requires adaptation plans that will prevent interruption of service to communities that are dependent upon the transportation system. A prior step to such planning is identifying the vulnerable parts of the system. An indicator-based approach is commonly employed for that matter, wherein various indicators are integrated to measure vulnerability. Limitations of this approach include arbitrary aggregation and weighting of indicators. These limitations are addressed by prioritizing indicators to create ordered asset lists for adaptation needs. Roads and bridges are analyzed in separate schemes, and the results are presented both in a table and a map format, visualizing the relevant features to users, with the purpose of providing a vulnerability assessment that is transparent, modifiable, and rational.

Investment valuation of underground freight transportation (UFT) systems is subject to a high level of uncertainties, including construction cost fluctuations and shipment revenue variations. Real options analysis is known as a promising approach for investment valuation of transportation projects under uncertainty. However, the extant real options approaches are theoretically deficient to appropriately model the UFT uncertainties with heterogeneous stochastic behaviors. While fluctuations in the construction cost of transportation projects typically follow the geometric Brownian motion (GBM) process, freight rate variation is a mean-reverting process. For the first time, a bivariate real options model is developed for investment valuation of UFT systems considering both these underlying uncertainties. A calibration approach is established to customize the model for capturing both GBM and mean-reverting attributes of the uncertainties. Also, a dynamic programming approach is adopted to identify the optimal time to invest in a UFT system. Finally, an illustrative case example is used to show the step-by-step implementation of the proposed model. Although the net present value (NPV) analysis showed this UFT case example is not financially viable, the proposed real options model demonstrated that this investment becomes quite promising if future contingent outcomes and the value of decision-making flexibilities are taken into account. The proposed model provides transportation capital planners with an optimal policy scheme that offers a timeline for viable investment in UFT systems. The presented model can also be used for investment valuation of other transportation development projects that involve two uncertainties with GBM and mean-reverting stochastic characteristics.

The transportation sector is a major source of traffic congestion and greenhouse gas (GHG) emissions in urban areas. This study aims to design the system-optimal (SO) traffic management policy, i.e., multiperiod tradable credit scheme (MPTCS), for urban transportation infrastructure. Under this scheme, the central authority (CA) seeks to minimize the total system travel time while achieving a predetermined vehicular emissions standard in each period of a long-term horizon with a duration of multiple years. Because travel demand and supply are uncertain in transportation infrastructure over a long-term horizon, the CA cannot provide accurate forecasts of future credit prices (CPs) to travelers a priori, leading to traveler heterogeneity in perceiving these future CPs. It impacts travelers’ decisions about using credits or transferring them to the next period. Further, travelers perceive credit purchases in the market as monetary losses and selling credits as monetary gain. We formulate the SO MPTCS design as a bilevel model. The CA determines the credit charging and allocation schemes in the upper-level model to minimize total system travel time over a horizon constrained by the emissions standard for each period. The MPTCS equilibrium condition is formulated in the lower level where travelers decide about credit consumption and path choice to minimize their travel costs. Numerical experiments suggest that as the difference between travelers’ perception of future CPs and the actual CPs (set by the CA for various periods) increases, the effectiveness of the SO MPTCS design in minimizing total system travel time decreases unless this difference is explicitly factored in the design. Also, if the CA increases emissions standards under the SO MPTCS design, the travel costs increase.

Infrastructure interdependencies have been widely recognized, especially in the postdisaster restoration process. It is essential to develop models to simulate interdependencies and quantify their impact on the functionality recovery of infrastructures. This study presents a generalized simulator to investigate the impact of different types of interdependency on functionality recovery. The proposed simulator considers that there are multiple possible modes to execute a restoration task by framing the restoration process of interconnected systems as a multimode resource-constrained project scheduling problem (MRCPSP). In addition, it considers three sets of uncertainties: restoration duration and resource demand to execute a task, as well as intersystem functionality dependency. By solving the MRCPSP with the objective of minimal restoration completion time, the optimal restoration schedules for different systems are calculated to predict functionality recovery. This simulator implements three types of interdependencies at both the component level and the system level: resource-sharing interdependency, restoration precedence dependency, and functionality dependency. Through a simple example, it is demonstrated how the proposed approach can quantitatively evaluate the impact on system recovery due to different types of interdependency. Research findings from this study can help to identify the interdependencies with the strongest impact and then develop preventive mitigation actions and effective plans of emergency response and disaster recovery for interconnected systems.

Bridge infrastructure asset management system is a prevailing approach toward having an effective and efficient procedure for monitoring bridges through their different development phases including construction, operation, and maintenance. Damage to any structural component of a bridge will negatively affect its safety, integrity, and longevity. Bridge decks are more susceptible to severe deterioration because they are exposed to harsh conditions including heavy traffic, varying temperatures, road salts, and abrasive forces. The ability to forecast the conditions of bridges in an accurate way has been a great challenge to transportation agencies. Many previous research studies highlighted the need to have a data-driven approach in predicting and evaluating the deterioration conditions of bridges. As such, this paper develops a computational data-driven asset management system to evaluate and predict bridge deck deterioration conditions. A multistep interdependent research methodology was utilized. First, the best set of variables affecting the conditions of bridge decks was identified. Second, two computational machine learning models were developed for the prediction of deck conditions using artificial neural networks (ANNs) and $k$-nearest neighbors (KNNs). Third, a comparison between the developed models is conducted to select the ultimate model with the highest accuracy. The result is a framework that is able to evaluate and predict deck conditions with a prediction accuracy of 91.44%. While this research is applied to bridges in Missouri, the technique can be used on any similarly available data set nationwide. This study adds to the body of knowledge by devising a computational data-driven framework that is valuable for transportation agencies as it allows them to evaluate and predict deck conditions with high accuracy. Consequently, this will help in ensuring proper and effective distribution of funds allocated for the maintenance, rehabilitation, and repair of bridges. Ultimately, this will result in minimizing efforts, time, and costs associated with site inspection of bridge decks.

Bicycle paths or even bicycle lanes have not emerged as key priorities in traditional pavement systems analysis. Most cities rely on route preferences (e.g., common school routes) or visual checks to prioritize pavement conditions on bicycle facilities. We used 31 bike path sections with a representative range of pavement surface conditions to collect acceleration data, GPS location data, bicycle steering angle, surface displacement data, and mean texture depth (MTD) data. We also recruited cyclists to complete a post-ride survey on ride quality. Using these data, we specified two ordered logit regression models to separately examine the relationships between bicycle ride quality and traditional pavement roughness measurement (or surface defect density on trajectories) while holding other parameters (e.g., bicycle accelerations and steering angle) constant. Our study shows that a surface defect index can replace the MTD test for bicycle facilities and can produce better performance in predicting ride quality, especially when pavement condition needs moderate repair to avoid becoming much worse. We also examine ride quality, specifically the vertical acceleration effect on ride experience, for different types of bicycles (e.g., a mountain bike with a suspension system versus a touring bike).

Considering the uncertain performance of lifelines with respect to earthquakes, the seismic assessment of critical infrastructures is of the utmost importance. The main purpose of this paper is to develop a model for critical infrastructures, with consideration of the uncertainties in a seismic damage analysis, using the Monte Carlo Simulation (MCS) and Latin hypercube sampling (LHS) methods. Accordingly, a system is designed, based on the geo-spatial information system to analyze the spatial data by the seismic hazard analysis, vulnerability assessment, and seismic damage analysis of critical infrastructures, including water, power, gas, and oil networks. In addition to the probabilistic based methods, the one-loop (1L) method also has been used to evaluate the damage analysis without considering the uncertainty. The estimated results from various methods have been compared, and it is concluded that the estimated peak ground acceleration (PGA) values are high in the central areas of the desired case study. Moreover, the permanent ground displacement (PGD) due to lateral spreading using the 1L method is higher than the estimated values based on MCS and LHS approaches. The outcomes based on the 1L method are within the range of 0–0.762 m and within the range of 0–0.381 m for MCS and LHS approaches. However, damages from MCS and LHS are equivalent, according to the Mann-Whitney test.

The relevance of social networks in engaging communities in planning and project decision making has been established. This case study work analyzes the social and semantic networks for the Twitter accounts of four large light rail transit (LRT) projects in North America. These networks portray typical features of social networks: consistent small-world behavior, formation of stable subcommunities, and dynamic variation in topic and issues discussed with changes to the project conditions. The proposed methodology can be helpful to future projects—mainly because the cases were analyzed with an additional focus on the dynamics of social networks of participants. In early stages of the projects, these networks have a limited group of participants. As time passes and more interest is cumulated in the project, the number of nodes increases but the number of connections (between the nodes) does not increase at a similar speed. This reduces the overall density of the network and, at the same time, creates “hub” nodes. These are influential participants with an impact on the information diffusion and formation of opinions. The subcommunities that form evolve in membership, density, and topics of interest, in accordance with the project progress. Citizens migrate between subcommunities, mainly based on the central topics of discussion and, partially, the addition or migration of influential nodes. The use of the proposed methodology can be helpful to future projects because it showcases a sociosemantic approach for the analysis of stakeholders. This can help decision makers to not only understand what is on the mind of communities, but also track how such thoughts (issues) evolve over time and what the role of key players is in forming or changing the issues. A careful analysis over the project life span can help the decision makers to predict some of these trends and, accordingly, shape the project communication strategy in a more proactive fashion to meet the possible community needs. In the era of unstructured big data, this methodology can help practitioners to study the impact of decisions made or scope deviations in a project on the themes and levels of interest of local communities. Both of these two features will help to transfer public communication in construction mega-projects from a one-way outreach to a two-way interaction: try to understand/predict community issues, gather community interests and concerns, and study how community structure and views change based on the changes made to the project scope.

Emergency medical services (EMS) rely on a robust road network to respond to patient demand as quickly as possible. An important measure of system effectiveness during adverse scenarios, such as heavy traffic and road closures, is ambulance response time. This project used data from the 2010 US National Census and the US Census Bureau’s 2015 Longitudinal Employer-Household Dynamics Origin-Destination Employment Statistics (LODES) to construct an agent-based microsimulation using Multi-Agent Transport Simulation (MATSim version 0.10.1) software. The model included both ambulances and commuter traffic in New Windsor, NY. The average ambulance response time was recorded for an undamaged road network. Then, disaster scenarios were simulated on three sets of links: the set of four bridges in New Windsor, the most mathematically central links, and the most frequently traveled links in the control scenario. After simulating a postdisaster ambulance response, the removal of all three sets of links increased response time similarly, but the removal of the most frequently traveled links had the greatest adverse effect.