Journal of Infrastructure Systems

Forum papers are thought-provoking opinion pieces or essays founded in fact, sometimes containing speculation, on a civil engineering topic of general interest and relevance to the readership of the journal. The views expressed in this Forum article do not necessarily reflect the views of ASCE or the Editorial Board of the journal.

Efficient and safe rail transit (metro) operations require regularly inspected and properly maintained infrastructures. While preventive maintenance is usually undertaken following a predetermined schedule, decisions on repairs and corrective maintenance are usually scheduled ad hoc, following inspections. This paper presents a model for optimally scheduling corrective maintenance activities in a rail transit network, following their identification as part of an inspection process. An integer programming model is formulated for that purpose, aiming at minimizing passenger delay caused by reduced operational speeds on degraded rail segments and prioritizing activities in segments with intensive operations; higher train traffic and associated train loads; and under budget, staff, and other practical constraints. The model is applied for the case of the Athens, Greece, metro system and results are discussed.

Equity:debt ratio and concession period length are the most important financial variables to consider when analyzing a build-operate-transfer (BOT) agreement. Because the involved parties have conflicting interests regarding these variables, this research introduces a new trilateral bargaining model which helps the BOT participants to model the BOT negotiation process (especially in a procurement process based on direct negotiation) and reach an efficient agreement on the concession period length and equity:debt ratio considering the conflicting interests and different bargaining powers of the sponsor, government, and lender. First, a mathematical interpretation of conflicting financial interests of the parties is introduced. Then the noncooperative and cooperative bilateral bargaining games are combined as a novelty of this research to develop a trilateral bargaining game model enabling consideration of the negotiation process in BOT projects. The proposed model can be used to predict the concession period length and capital structure simultaneously. Finally, the proposed model is implemented in a real-world BOT project and the application and performance of the proposed model are illustrated.

In maintenance and rehabilitation (M&R) programming, most budget allocation models in literature often assume one principal budget that can be used for funding all the different projects undertaken by the decision-making entity. For most agencies, this is rarely the case. This study explores the impact of formulating the M&R budget allocation problem with multiple funding categories (or budgets) having stringent project eligibility constraints. For state highway agencies (SHAs), there are often several federal, state, and local funding sources for highway projects. There are also agency norms and regulations that restrain the eligible projects under each funding category. This study presents an integer–linear programming model that accounts for funding restrictions in M&R budgets to assess changes in network performance and M&R decisions for pavement assets. The model is implemented in a numerical case study representing a subset network consisting of 500 pavement sections. Findings from this study suggest that the projected performance of the M&R program is overestimated when there is the assumption of one “central” budget with no project eligibility constraints. This may lead to an increase in unplanned expenditures toward reactive M&R projects to meet performance targets. Furthermore, the findings also suggest that increasingly restrictive budgets lead to a lower network performance for the same aggregate budget. The sensitivity analyses conducted confirm that the results obtained are robust against variations in key input variables used in the model. These findings contribute to ongoing efforts toward incorporating pragmatic constraints in optimization models to enhance their utility for effective decision making.

Because proprietary image data formats are used in the pavement industry, there is a critical need to develop a standard interchangeable format for pavement surface conditions and transverse profiles such that pavement image data from various highway agencies and technology suppliers can be shared across different analyses and software platforms. A comprehensive literature review and a survey of current practices found that two-dimensional (2D) image data compression is primarily based on JPEG and JPEG2000, whereas the three-dimensional (3D) laser scanning and imaging technology has rapidly gained popularity using various data formats and dynamic ranges. In this study, a standard data format for 2D and 3D pavement images is developed to meet different data requirements. The standard contains two parts. The first part is the core section including file header, 2D intensity data, and 3D range data, whose definitions shall not be changed as the data format standard evolves; the second part is primarily user-defined metadata, which can be extended and modified as needed to accommodate individual data collection practices and equipment. The expected benefits include facilitating workable protocols for condition surveys, improving implementation of new technologies, and accelerating the development of analysis tools for pavement condition evaluation.

Low natural gas prices offer an opportunity to expand the use of this fuel in new sectors, such as powering long-haul trucks. The non-return-to-base nature of long-haul trucks requires a large natural gas refueling infrastructure, for which the costs and benefits have not yet been assessed. This paper assesses the spatial distribution, capacity, and economic feasibility of a national natural gas refueling infrastructure for long-haul trucks in the United States. For a small share of natural gas-powered trucks in the total fleet (1%–5%), a national liquefied natural gas (LNG) refueling infrastructure would require from 105 to 193 refueling stations at a capital investment of $263–$483 million (assuming a truck range of 320–960 km or 200–600 mi). The share of LNG trucks has to be greater than 15.2% (assuming the minimum truck range is 320 km or 200 mi), 12.4% (480 km or 300 mi), or 9.5% (960 km or 600 mi), so that the national infrastructure achieves net profits for baseline economic assumptions [$\$0.05/\mathrm{diesel}$ gal. equivalent (DGE) profit margin, 7% discount rate, and 20 years]. Building a national compressed natural gas (CNG) refueling infrastructure requires more investment because a fast-fill CNG refueling station costs more and has lower capacity compared to an LNG counterpart. However, the CNG refueling infrastructure achieves better economic returns due to a higher assumed profit margin ($\$0.1/\mathrm{DGE}$). As the diesel price has the largest effect on the adoption of natural gas trucks, plans to build natural gas refueling infrastructures face significant challenges during times of low diesel prices. A practical strategy is to first target regional highway networks (such as those in California and Texas) to take advantage of high truck activities within regions, favorable market conditions, and supportive government policies.

Any engineering structure is designed and built to operate in a dynamic and changing environment that cannot be fully predicted at design. Thus, successful designs ought to account for provisions that can handle uncertain but predictable futures and that allow the system to cope with new and unplanned events. These needs have attracted interest in concepts such as resilience, robustness, flexibility and adaptability, which are all part of a more general idea: changeability. This paper presents a critical discussion of the limitations of existing life-cycle approaches, discusses the need for and importance of changeability in engineering design, and presents a roadmap that may open new spaces for a new engineering design paradigm that contributes to the efficient development of high-value infrastructure. The paper discusses critically current practices in infrastructure design and management and proposes an alternative approach centered around the system’s ability to change over time.

This paper presents a new method to develop stochastic deterioration models using a combination of methods including Markov chains, logistic regression, and classification trees. It is computationally more efficient to use logistic regression with the Markov chain process than it is to use optimization-based approaches, and the former is shown to marginally improve the prediction of condition ratings for small data sets. Annually inspected bridge data are split into groups using a classification tree, and logistic regression is used to determine transition probabilities for a Markov chain process. A case study was conducted to determine the effectiveness of using the proposed logistic regression and Markov chain approach for the small data sets created by the classification tree. Wyoming bridge inspection data were split into 15 subsets based on 5 explanatory variables, and deterioration models were developed for each subset. Error analysis showed that logistic regression performed marginally better than traditional methods when estimating the transition probability matrix when limited data are accessible.

The successful prediction of civil infrastructure’s deterioration process is crucial for making optimal maintenance, rehabilitation, and replacement (MR&R) decisions under financial constraints. The majority of current deterioration models simulate the deterioration process of a single structure element of civil infrastructure; such models thus ignore the interaction between dependent elements. However, the interaction between structure elements often plays an important role in the deterioration of the overall structure. Therefore, the primary objective of this paper is to address the interaction of these structure elements by developing a method to simulate the deterioration process of civil infrastructure on a system level. The proposed method will also provide a measure of the uncertainty of the simulation using Markov chain Monte Carlo (MCMC) to estimate the optimal parameters of the Markov chain and the probability distribution of those parameters. The Monte Carlo simulation is then used to generate a large number of deterioration process samples, which serve as the base of the uncertainty analysis of the simulation. The model was applied to simulate the deterioration process of a bridge element subsystem as an example application. In this example application, the model was calibrated and evaluated by the bridge inspection record collected in the Commonwealth of Virginia. The results demonstrate that including the interaction between elements in the model improves the accuracy of deterioration simulation while also reducing the uncertainty of the results. Furthermore, the proposed model is relatively easy to implement in current infrastructure management systems (IMSs) compared with other methods such as neural networks and fuzzy logical models.

In the coming decades, Rijkswaterstaat (RWS), the executive branch of the Dutch Ministry of Infrastructure and the Environment, has to replace or thoroughly renovate more than fifty navigation locks. Given the substantial number of locks that needs to be renewed in a relatively short time frame, Rijkswaterstaat is exploring the development of a lock family platform from which future locks can be designed and realized. This study aims to find basic lock component modules, optional lock component modules, and their interfaces, together shaping the platform. Its aim is to identify modules and components that are candidates for standardization. For this purpose, 127 locks in the Netherlands are analyzed using a similarity matrix to find lock variants. Dependency structure matrices (DSM) are built to find commonalities among the variants. By projecting impact scores upon the DSM, component modules are sought that are candidates for (partial) standardization. The results provide clear insight in how to design the lock family platform.

Natural or human-made incidents in urban areas cause surges in demand for hospitals while often simultaneously limiting their capacity to serve patients due to direct physical damage or service interruptions in supporting lifelines. A hierarchical modeling concept is proposed to quantify the resilience of regional hospital response under disaster. To facilitate the optimization of pre- and postevent resilience-enhancing actions for a range of potential hazard-demand-damage scenarios, a multistage stochastic, mixed-integer decision problem (SMIP) is proposed. Dependencies on interdependent, potentially damaged critical lifelines are explicitly modeled. Resilience is estimated in terms of total patient waiting time and unserved patients. These quantities are captured in the objective of the SMIP through the adoption of a simulation-based metamodel of a detailed hospital model. The SMIP is demonstrated in a numerical example developed to represent the main hospitals of the Johns Hopkins Health System Corporation. The proposed structure is broadly applicable to other societal functions that take place in interrelated buildings connected by transportation and communications links and are dependent on interdependent lifeline networks.

Given that decision making for highway infrastructure development is highly complex, it is highly desirable to access expertise and tools, such as information about the environmental and social cost components of highway investments, that can aid in evaluating the investment options. This research aims to develop an integration model to leverage the fuzzy analytical hierarchy process (fuzzy AHP) and life-cycle cost analysis (LCCA) in evaluating highway infrastructure investments. The proposed model assesses the qualitative and quantitative factors of cost components. A case study in Western Australia was adopted to apply and illustrate the proposed model. By closely monitoring the implementation of the sustainability measure together with engineering economics, this research contributes to the systematic and informative assessment approaches to highway infrastructure investments. The case study has also demonstrated a quantified and objective decision-making process for financial viability and sustainability delivery.

Given the advancements of wireless communications and sensor technologies, numerous studies were conducted to investigate the feasibility of using probe vehicle data (i.e., vehicle floor acceleration) for pavement condition assessment. Challenges of this approach are that there are a huge variety of vehicle operating speeds and vehicle dynamic systems, which can directly impact the measurements. To address this issue, the University of Virginia Center for Transportation team previously developed an improved acceleration-based metric that incorporates vehicle speeds. As a follow-up, the study presented in this paper proposes a methodology that can incorporate the impact of vehicle dynamic systems for better pavement roughness measurements. Profile and probe data were collected on a 17.38-km (10.8-mi) segment using three different vehicles. The analysis found that different vehicle dynamic parameters can result in significantly different vibration responses, which necessitates a calibration process in terms of vehicle dynamic parameters. Furthermore, the vibration responses are found to be approximately linearly correlated between different vehicle systems and thus a linear regression model is suitable for a calibration model. Finally, assuming a network screening system using agency-owned vehicles, a calibration procedure in terms of vehicle dynamic systems was developed. Case studies using real probe data collected from different vehicles are also presented to demonstrate that the proposed calibration procedure can improve pavement roughness estimates.

Current economic models specifically designed for infrastructure outage and disaster situations are attempting to incorporate the ability of businesses to prepare for, adapt to, and recover from, disruptions. Using business impact and recovery data from the 2010–2011 Canterbury earthquakes, and qualitative validation, this paper presents an empirically derived, transferable model for estimating business recovery following infrastructure and noninfrastructure disruptions. The business behaviors model (BBM) is a logarithmic function that calculates operability (ability to meet demand) over time based on 15 industry sectors and the level of experienced infrastructure and noninfrastructure disruption. The BBM can be applied as a temporal adjustment factor to an economic model, or it can be used as a tool to explore the impact of different disruption types and magnitudes on organization recovery.

The aim of this work is to provide road agencies with a new strategic approach for integrating risk assessment into pavement management systems (PMS). An application to a small portion of the Italian road network is described, to demonstrate the practicality of the proposed procedure. Innovative concepts of target and actual maintenance levels of service were developed; risk mitigation was studied and optimization techniques were used to define a risk/cost ratio to select the best preservation strategy under different budget scenarios. The work helped managers to determine a proper functional classification of the existing road network, knowledge of the maintenance level on the managed roads, effective communication to road users, and rational use of the available economic resources. The case study indicates that a risk-based management approach is more effective than a performance-based approach. The tool is a valuable contribution to engineers and practitioners, when an actual PMS has not been developed yet. The proposed methodology can also allow researchers to define the relationship between safety and pavement conditions.