Institute for Transportation

About the research

This project was intended to create an intelligent prognostics platform for lithium-ion (Li-ion) batteries, which would equip existing battery management systems with the capability to perform predictive maintenance/control for failure prevention. The platform developed in this project consisted of two modules:

• Deep feature learning, which automatically learns the features of (capacity) fade from large volumes of voltage and current measurement data during partial charge cycles and estimates the real-time state of health (SOH) of a battery cell in operation
• Ensemble prognostics, which leverage the current and past SOH estimates in Module 1 to achieve robust prediction of the cell’s remaining useful life

Robust prediction of remaining useful life was achieved by ensemble learning-based prognostics, which synthesized the generalization strengths of multiple prognostic algorithms to ensure high prediction accuracy for an expanded range of battery applications and their operating conditions. The two modules aimed to learn features of fade from partial charge data, assess real-time health of individual battery cells, and predict when and how the cells are likely to fail. A case study involving implantable-grade Li-ion cells was conducted to demonstrate a deep learning approach to online capacity estimation, developed for Module 1.

Funding Sources:
Iowa State University ($50,258.00)
Midwest Transportation Center
USDOT/OST-R ($50,000.00)
Total: $100,258.00

Contract Number: DTRT13-G-UTC37

About the research

Thanks to an array of advanced digital technologies, much of today’s transportation project data are available in digital format. However, due to the fragmented nature of the highway project delivery process, the growing amount of digital data is being archived and managed separately. This makes it difficult for professionals to take full advantage of the efficiencies of digitized data and information. The purpose of this research was to identify current data workflows and areas for improvement for five of the most common types of highway assets—signs, guardrails, culverts, pavements, and bridges—and offer guidance to practitioners on how to better collect, manage, and exchange asset data.

The research team conducted focus group discussions and interviews with highway professionals to capture their knowledge and practices about the data workflows. In addition, the team conducted an extensive review of the literature, manuals, project documents, and software applications regarding the exchanged information. For each type of asset, an information delivery manual (IDM) was developed. Each IDM consists of several process maps (PMs) and one exchange requirement (ER) matrix. A total of 15 PMs and 5 ER matrices were developed.

A set of limitations in current data workflows was identified and a set of recommendations to overcome those limitations was also determined and documented. The conclusion was that current data workflows were designed mostly for contract administration purposes. Thus, more efficient asset-centric data workflows need to be implemented to truly streamline the data workflows throughout an asset’s life cycle and minimize wasted resources in recreating data in the asset maintenance stage.

Funding Sources:
Iowa Department of Transportation
Iowa Highway Research Board ($50,000.00)
Iowa State University ($22,500.00)
Midwest Transportation Center
USDOT/OST-R ($80,000.00)
Total: $152,500.00

Contract Number: DTRT13-G-UTC37

About the research

The transportation systems sector, one of the most critical infrastructure sectors in the US, has a subsector of highway and motor carrier industries that supports daily activities and emergency actions by providing services to other critical infrastructure segments such as healthcare and public health, emergency services, manufacturing, food and agriculture, etc. However, transportation networks face risks from natural and human-made events such as hurricanes, tsunamis, earthquakes, bridge collapse, and terrorist attacks. Thus, to improve the reliability of the components in interconnecting networks, it is necessary to consider these unpredictable failures in the network design. Resilient network design ensures that the network functionality is at an acceptable level of service in the presence of all probabilistic failures.

In this study, the authors addressed uncertainty in a transportation network by proposing a trilevel optimization model, which improves the resiliency of the network against uncertain disruptions. The link capacities are uncertain parameters and the origin-destination demands are deterministic. The goal was to minimize the total travel time under uncertain disruptions by designing a resilient transportation network. The trilevel optimization model has three levels. The lower level determines the network flow, the middle level assesses the resiliency of the network by identifying the worst-case scenario disruptions that could lead to a maximal travel time, and the upper level uses the system perspective to expand the existing transportation network to enhance the network’s resiliency. In addition, the authors propose a new formulation for the network flow problem that will significantly reduce the number the number of variables and constraints.

The results of solving the trilevel optimization model can improve the resiliency of the network. However, this study was subject to some limitations, which suggested future research directions. In reality, transportation demands are not consistent, but the proposed model considers origin-destination demands as deterministic parameters. Relaxing this assumption requires a more complicated model to reflect uncertain demands. Other possible future work would be designing an exact algorithm to find the optimal solution.

Funding Sources:
Iowa State University ($99,999.00)
Midwest Transportation Center
USDOT/OST-R ($124,998.00)
Total: $224,997.00

Contract Number: DTRT13-G-UTC37

About the research

Current technology in traffic control is limited to a centralized approach that has not paid appropriate attention to efficiency of fuel consumption and is subject to the scale of transportation networks. This project proposes a transformative approach to the development of a distributed framework to reduce the balanced fuel consumption and travel time through hybrid control on speed limit and ramp metering rates. It proposes to integrate the roadway infrastructures equipped with sensing, communication, and parallel computation functionalities in the new traffic control paradigm.

The research approach builds on three essential objectives that will jointly lead to a solid theoretical and experimental project to establish energy-efficient traffic control methodology:

• Implementation of distributed control framework in large scale transportation networks
• Simulation of dynamic traffic flow and performance tracking under implemented control signals using real traffic and vehicle data
• Data analysis and sustained strategy improvement

Going beyond the existing distributed architectures where precise dynamic flow models and fuel consumptions have not been considered, the work generated traffic control strategies to realize real-time, macroscopic-level traffic regulation with high precision.

Simulation results demonstrated reduced fuel consumption and alleviated traffic congestion. The feasibility of the proposed optimization method was verified through Vissim simulation that considered different traffic volumes and random seed parameters.

About the research

Current technology in traffic control is limited to a centralized approach that has not paid appropriate attention to efficiency of fuel consumption and is subject to the scale of transportation networks. This project proposes a transformative approach to the development of a distributed framework to reduce fuel consumption and travel time through the management of dynamic speed limit signs. The project proposes to integrate the roadway infrastructures equipped with sensing, communication, and parallel computation functionalities in the new traffic control paradigm.

The research approach was built on three essential objectives to establish an energy-efficient traffic control methodology:

• Implementation of a distributed control framework in large-scale transportation networks
• Simulation of dynamic traffic flow and performance tracking under implemented control signals using real-time traffic and vehicle data
• Data analysis and sustained strategy improvement

Going beyond the existing distributed architectures where precise dynamic flow models and fuel consumptions have not been considered, the work generated traffic control strategies to realize real-time, macroscopic-level traffic regulation with high precision.

Simulation results demonstrated reduced fuel consumption and alleviated traffic congestion. The feasibility of the proposed optimization method was verified through Vissim simulation that considered different traffic volumes and random seed parameters.

Funding Sources:
Iowa State University ($71,102.00)
Midwest Transportation Center
USDOT/OST-R ($70,047.00)
Total: $141,149.00

Contract Number: DTRT13-G-UTC37

About the research

Consistent efforts with dense sensor deployment and data gathering processes for bridge big data have accumulated profound information regarding bridge performance, associated environments, and traffic flows. However, direct applications of bridge big data to long-term decision-making processes are hampered by big data-related challenges, including the immense size and volume of datasets, too many variables, heterogeneous data types, and, most importantly, missing data. The objective of this project was to develop a foundational computational framework that can facilitate data collection, data squashing, data merging, data curing, and, ultimately, data prediction. By using the framework, practitioners and researchers can learn from past data, predict various information regarding long-term bridge performance, and conduct data-driven efficient planning for bridge management and improvement.

This research project developed and validated several computational tools for the aforementioned objectives. The programs include (1) a data-squashing tool that can shrink years-long bridge strain sensor data to manageable datasets, (2) a data-merging tool that can synchronize bridge strain sensor data and traffic flow sensor data, (3) a data-curing framework that can fill in arbitrarily missing data with statistically reliable values, and (4) a data-prediction tool that can accurately predict bridge and traffic data. In tandem, this project performed a foundational investigation into dense surface sensors, which will serve as a new data source in the near future. The resultant hybrid datasets, detailed manuals, and examples of all programs have been developed and are shared via web folders.

The conclusion from this research was that the developed framework will serve practitioners and researchers as a powerful tool for making big data-driven predictions regarding the long-term behavior of bridges and relevant traffic information.

Funding Sources:
Iowa State University ($80,000.00)
Midwest Transportation Center
USDOT/OST-R ($80,000.00)
Total: $160,000.00

Contract Number: DTRT13-G-UTC37

About the research

This project investigated the factors impacting individual vehicle energy consumption, including vehicle characteristics, ambient temperature, season, speed, driving behavior, and traffic flow. A fleet of 18 vehicles with a variety of ownership, size, model, year, and powertrain characteristics was monitored using on-board diagnostics II (OBD-II) loggers to collect each vehicle’s controller area network (CAN) bus data for a one-year period.

Traffic data, including flow rate, space mean speed, and density, were also collected and linked with the vehicle data. Based on vehicle CAN bus data, fuel consumption models for gasoline vehicles were calibrated, and a new electricity consumption model for electric vehicles was proposed. Both models can reliably estimate individual vehicle energy consumption.

Funding Sources:
Iowa State University ($41,261.00)
Midwest Transportation Center
USDOT/OST-R ($40,000.00)
Total: $81,261.00

Contract Number: DTRT13-G-UTC37

About the research

While levee embankments are the first line of defense for urban flooding, recent flooding events have revealed widespread slope instability of embankments around the country that can lead to levee failures. This work aimed to improve the slope stability of earthen levees by strengthening the soil through the use of a biologically-inspired technique called biocementation.
The major objective of this study was to determine the optimal and most practical biocementation method that results in the best performance of levee slopes under various flooding conditions.
A novel biocementation method called bacterial enzyme-induced calcite precipitation (BEICP) was tested. This method differs from the well-studied microbial-induced calcite precipitation (MICP) method mainly in the size of the biological agent (whole bacterial cell vs. enzyme), which influences the agent’s mobility in soils of different grain sizes.
The BEICP methods for soil strengthening were optimized in laboratory-scale column experiments. The optimized methods were then used to construct levees in an experimental flume system.
The results indicated that the BEICP treatment resulted in significant strengthening of the surface of the soil specimens, which were measured as increases in unconfined compressive strength. The strengthened samples were able to resist erosion for a longer time period under overtopping scenarios of water challenges with high water velocity in a flume. Therefore, BEICP offers a sustainable and economical method for treatment of soil surfaces to improve erosion resistance.
The results of this study will help mitigate flooding events that could cause major issues in transportation infrastructure and traffic safety.

Funding Sources:
Iowa State University ($82,233.00)
Midwest Transportation Center
USDOT/OST-R ($80,000.00)
Total: $162,233.00

Contract Number: DTRT13-G-UTC37

About the research

More than a million Iowans and 59 million Americans reside in rural areas. The median rural resident is older, economically poorer, and more ethnically diverse, and enjoys a lower level of accessibility to infrastructures/facilities such as hospitals, grocery shops, social service providers, and governmental facilities (“essential services,” for shorthand hereafter) relative to their urban counterparts. On the one hand, essential services have been centralized and/or reduced due to economic stagnation; on the other hand, the availability and state of repair of our transportation infrastructure significantly influence the accessibility of essential services. However, much of the existing research on the accessibility of essential services has focused on urban residents/areas. There is a critical need to evaluate the accessibility to essential services among rural residents and to propose feasible solutions to enhance it if necessary.

This research has four objectives:

1. Investigate the minimum sufficient essential services (MSES) among rural residents and assess the accessibility to MSES among rural Iowans, taking into account sociodemographic characteristics and the transportation infrastructure’s state of repair.
2. Measure the existing accessibility to various essential services among rural Iowans, accounting for sociodemographic characteristics and the transportation infrastructure’s state of repair.
3. Quantify the accessibility gaps between the existing and minimum sufficient essential services among representative subgroups of rural Iowans.
4. Based on the above, provide policy recommendations for transportation decision-makers regarding how to optimize rural Iowans’ accessibility to MSES in an era of tight budgets.

This research is not merely of interest to transportation professionals and officials. Understanding whether a million rural Iowans have accessibility to MSES and wisely enhancing that accessibility is a community and economic development issue that will also be of interest to a variety of stakeholder groups and community leaders.

Funding Sources:
Iowa State University ($76,019.00)
Midwest Transportation Center
USDOT/OST-R ($73,874.00)
Total: $149,893.00

Contract Number: DTRT13-G-UTC37

About the research

Current bridge management systems predict the condition state of bridge elements primarily based on the extent of continuous structural deterioration. While the existing systems deliver a range of capabilities for the management of bridges under normal operational conditions, they lack the capability to take into account the consequences of sudden extreme events in a systematic way. In addition to extreme events, the consequences of extreme environmental exposure due to climate change are missing from current management systems. Given the uncertainties involved in natural and manmade hazards in addition to the ones associated with environmental exposure conditions, there is a critical need to develop risk-based approaches that not only take into account the site-specific aging mechanisms and extreme events at the same time, but also accommodate the spatial and temporal randomness originating from these factors. Another significant source of randomness and uncertainty is inspectors’ judgment, which directly affects predictions of the condition state of deteriorating bridges. Towards this goal, the current study introduces a risk-based life-cycle cost analysis framework that can be implemented in the current bridge management systems used by transportation agencies.

Funding Sources:
Iowa State University ($81,143.00)
Midwest Transportation Center
USDOT/OST-R ($77,121.00)
Total: $158,264.00

Contract Number: DTRT13-G-UTC37