National Academies Press: OpenBook
« Previous: Appendix A. Interactive Demonstration Titles and Abstracts
Page 68
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 68
Page 69
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 69
Page 70
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 70
Page 71
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 71
Page 72
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 72
Page 73
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 73
Page 74
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 74
Page 75
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 75
Page 76
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 76
Page 77
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 77
Page 78
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 78
Page 79
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 79
Page 80
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 80
Page 81
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 81
Page 82
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 82
Page 83
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 83
Page 84
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 84
Page 85
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 85
Page 86
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 86
Page 87
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 87
Page 88
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 88
Page 89
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 89
Page 90
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 90
Page 91
Suggested Citation:"Appendix B. Breakout Session Abstracts." National Academies of Sciences, Engineering, and Medicine. 2018. Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics. Washington, DC: The National Academies Press. doi: 10.17226/25336.
×
Page 91

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

68 APPENDIX B Breakout Session Abstracts Session 1A: Data Analytics: Maritime and Freight 1 ............................................. 71 Deriving Value from Waybill Data, Traffic Volumes, and the Integration of Both ..................... 71 Charles Edwards Freight Production Modeling Using Census Microdata ......................................................................... 71 Shama Campbell Ohio Maritime Study Information Visualization Techniques .............................................................. 72 Donald Ludlow Session 1B: Data Analytics: Inland Waterways ...................................................... 72 Inland Marine Transportation Data Integration: Extracting Additional Value from Publicly Available Data ................................................................................................................ 72 James Dobbins Measuring the Network Impacts of Local Disruptions: An Inland Waterways Case Study ... 73 Craig Philip Modeling Dynamic Behavior of Navigable Inland Waterways ........................................................... 74 Heather Nachtmann Quantifying the Impacts of Disruptions to the Inland Marine Transportation System ........... 74 Patricia DiJoseph Session 1C: Decision Support: Resilience ............................................................... 75 2017 Hurricanes: A Resilient Path Forward for Marine Transportation System Federal Agencies ..................................................................................................................................... 75 Katherine Touzinsky A GIS Inventory and Exposure Assessment for Critical Coastal Transport Infrastructure Land Use in the Caribbean Small Island Developing States .................................. 76 Gerald Bove Barriers to Climate and Extreme Weather Adaptations for Seaports: A Cultural Consensus Model for North Atlantic Medium and High-Use Port Decision-Makers ................. 76 Elizabeth Mclean Using Geographic Information Science to Evaluate Legal Restrictions on Freight Transportation Routing in Disruptive Scenarios ..................................................................... 77 Steven Peterson Session 2A: Data Analytics: Maritime and Freight 2 .............................................. 77 Container Ship Routing and Scheduling with Multiple Time Windows ........................................ 77 Anastasios Charisis

69 Modeling Agricultural Commodity Flows on the U.S. Railroads and Inland Waterways System Using Waybill and Waterborne Commerce Statistics Data ................................................. 78 Steven Peterson Short Sea Shipping Versus Trucking: A Cost–Benefit Analysis Using Mathematical Modeling ...................................................................................................................................... 78 Evangelos Kaisar Session 2B: Decision Support: Managing Flows ..................................................... 79 Applying Multimodal Freight Network Optimization to Public-Sector Investment Decisions .............................................................................................................. 79 Mark Berndt Data Science Approach to Bottleneck Identification: Freight Network Analytics .................... 79 Catherine Lawson Multiagency Data Fusion Informs Waterway Management ............................................................... 80 Brandon Scully The Voyage Plan: The Missing Link ............................................................................................................... 80 Brian Tetreault Session 2C: Data Analytics: Port Performance ...................................................... 81 Container Ship Bay Time and Crane Productivity: Are They on the Path of Convergence? ........................................................................................................ 81 Shmuel Yahalom Container Ship Dwell Times Through the Automatic Identification System Lens .................... 82 Daniel Smith and Daniel Hackett Ports of the Future: Deploying Emulation and Real-Time Simulation for Identifying Technologies for Improved Port Supply Chain Performance ............................................................ 82 Lawrence Henesey The Application of Freight Fluidity Metrics to the Port Environment ........................................... 82 Kenneth Mitchell Session 2D: Decision Support: Safety .................................................................... 83 An Exploratory Study of Near-Miss Events in Maritime Freight Systems Using the Marine Information for Safety and Law Enforcement Database ............... 83 Robin Dillon-Merrill Examining Maritime Casualties from Vessel Groundings Using Ordered Probit Modeling ................................................................................................................................... 84 Fatima Zouhair From Text to Data: How the U.S. Coast Guard Used Accident Reports for Benefits Analysis ........................................................................................................................... 84 Douglas Scheffler Session 3A: Data Analytics: Maritime and Freight 3 .............................................. 85 Cyber–Physical Applications for Maritime Freight Transportation Systems ............................. 85 Amirhassan Kermanshah

70 How Information Systems and Data Are Used to Control Operations and Influence Management Decisions at the Panama Canal ............................................................................................ 86 Parsa Safa Transport Analysis Framework of the King Abdullah Petroleum Studies and Research Center: Building a Global Freight Network Model with Satellite and Automatic Identification System Data .......................................................................................................... 86 Hector Guillermo Lopez-Ruiz Session 3B: Big Data and Machine Learning: Maritime Applications ..................... 87 Enhancing Database User Experience with Natural Language Processing .................................. 87 Dan Seedah Man Versus Machine: Comparing Traditional Data Collection and Statistical Models with Machine Learning Big Data Analytics ........................................................................................................... 88 Jolene Hayes The Practical and Unified Use of Blockchain, Cyber-Security, the Internet of Things, Machine Learning, Artificial Intelligence, and Big Data for the Marine Transportation Industry: Details on Building a Multimodal Freight Analytics Platform ....................................... 88 Dean Shoultz Validating Automatic Identification System Data Using Machine Learning Algorithms ........ 89 Edward Carr Session 3C: Decision Support: Environmental ....................................................... 90 Automatic Identification System Data Improvements to the 2014 Gulfwide Emissions Inventory Study ................................................................................................ 90 Heather Perez Merging Automatic Identification System and U.S. Coast Guard Data for Modeling Maritime Air Emissions .................................................................................................................. 90 Diane Rusanowsky Multiattribute Performance Analysis for Intermodal Maritime Cargos ........................................ 91 Jim Corbett

71 Session 1A Data Analytics: Maritime and Freight 1 Deriving Value from Waybill Data, Traffic Volumes, and the Integration of Both Charles Edwards, North Carolina Department of Transportation Major growth in freight transportation activity and rapid population growth in the state of North Carolina have brought new opportunities and challenges for the North Carolina DOT freight divisions. Access to additional data sources, the ability to integrate multiple data sources, and the ability to derive insight from the available freight data sources are some key factors that improve freight planning. To provide insights into freight activity, SAS was used to integrate data including traffic counts (average annual daily traffic), waybill data, and weigh-in-motion data and to perform advanced analytics, such as clustering, to study how freight activity around major freight centers has changed over time. The work resulted in multiple dashboards containing interactive visualizations of waybill data available to the North Carolina DOT to identify information about freight activity, such as type and value of the commodities that were shipped through the state and origin and destination information in the state by rail provider and year. Linking the waybill data with heavy vehicle traffic volume enables the North Carolina DOT to gain insight into increased truck activity around the state, including where future vehicle count stations should be located; this link provides valuable information for future data collection activities. The ultimate purpose of the project is to provide improved data for planning models and provide project scores that quantify impacts associated with freight movement that assist with making optimal, objective decisions when Shipboard Tactical Intelligence Processing (STIP) projects are being prioritized. Freight Production Modeling Using Census Microdata Shama Campbell, Rensselaer Polytechnic Institute The findings of the research published in NCFRP Research Report 37: Using Commodity Flow Survey Microdata and Other Establishment Data to Estimate the Generation of Freight, Freight Trips, and Service Trips: Guidebook are summarized. As a part of the research, the authors estimated 1,409 FP and amount of cargo produced/sent with models using the CFS confidential microdata for 37 sectors at two-digit and three-digit NAICS levels. The models, because of their solid economic foundation and their reliance on publicly available data, provide a unique opportunity to identify the specific locations, at the zip code level, which produce amounts of cargo suitable for large-capacity freight modes. The models quantify the relationship between FP at the establishment level with corresponding employment. Models were estimated for the United States and five states (California, New York, Ohio, Texas, and Wyoming). The team estimated FP models at two-

72 digit and three-digit NAICS levels for two sets of aggregations by mode: all modes and only road modes. The authors presented the final FP models, highlighted the differences between states and the overall United States, and discussed potential applications of the models in multimodal freight policy and planning. Ohio Maritime Study Information Visualization Techniques Donald Ludlow, CPCS Transcom Maritime data and analysis are rich in numbers, trends, and graphs. But visualization techniques help to convey important messages quickly, succinctly, and comprehensively. The overarching objective of the Ohio Maritime Study was to provide a foundation of information for the Ohio DOT so that the agency could best leverage Ohio’s MTS (both Great Lakes and inland waterways) to the benefit of Ohio’s economy. In addition to providing this base-level information to the Ohio DOT, the study engaged public- and private-sector stakeholders and shared this information to achieve common understanding. This action was important because of Ohio’s limited role in the maritime system. Informing partners is best achieved through the visualization of data. CPCS Transcom, in conducting the analysis for the Ohio DOT, developed techniques to convey information quickly and succinctly to maritime experts, economic development officials, shippers, port owners/operators, decision makers, and the general public through the use of mapping and simple but powerful visualization techniques. This presentation provided the context for the study, the stakeholders, the analysis methods used, and the visualization techniques developed to inform stakeholders, to solicit feedback and buy-in, and to convey important details and messages for decision-making. The presentation provided innovative approaches that CPCS took to convey information related to maritime inventory/facilities, traffic, commodity flows, bottlenecks, and other physical constraints. Session 1B Data Analytics: Inland Waterways Inland Marine Transportation Data Integration: Extracting Additional Value from Publicly Available Data James Dobbins, FACTOR, Inc. The U.S. inland marine industry has a limited number of publicly available data sets and data services. Agencies that provide such data include USACE, USDA, USCG, and NOAA. In the towing industry, publicly available data are typically considered a lower priority than private, internal data and/or proprietary data (e.g., Port Import/Export Reporting Service).

73 Modern BI platforms and ETL procedures enable companies to extract additional insights and value from a wide variety of data sets and data services with a much lower level of effort than previously required. An important advancement is the combination of robust data transformation and data visualization features within a single software application. These applications can simultaneously combine data from enterprise relational databases, spreadsheets, and websites, among many other disparate sources. The work presented focused on data modeling techniques and not on any specific software package functionality. This presentation demonstrated some of these advanced techniques for combining and leveraging these data sets to gain valuable insights related to fleet performance and business opportunities. In particular, the presenter demonstrated the results of combining USACE Corpslocks XML web services with static vessel operator data (USACE and USCG) and USDA grain rate and barge movement data to illustrate industry and specific business insights as well as fleet performance. The presenter discussed how inclusion of additional data sets (e.g., public rail waybill data) could be leveraged to understand better the competitive pressure posed by a competing mode of transportation. Measuring the Network Impacts of Local Disruptions: An Inland Waterways Case Study Craig Philip, Vanderbilt University America’s mature and often invisible inland waterway system is essential to the nation’s economic health and is unfortunately becoming more visible for all the wrong reasons. Projects more than 75 years old, persistent underinvestment, and disruptions by chronic extreme weather magnify system vulnerabilities, as illustrated on the Ohio River in 2017. The National Wildlife Federation and MARAD supported this study to demonstrate a robust modeling methodology to explore the impacts of extended unscheduled outages at a number of important lock and dam (L&D) projects. L&D PROJECT SCREENING AND SELECTION Initial screening carefully reconciled cross-sectional data describing the characteristics and performance of approximately 170 navigation L&D projects. New metrics include systemwide lockages and above/below pool traffic. Four locks were evaluated: Markland, Calcasieu, LaGrange, and L&D 25. MODELING DIRECT SHIPPER SUPPLY CHAIN COST BURDENS FROM AN UNSCHEDULED CLOSURE Three models were employed. They allowed a comparison of the costs associated with the use of barge service against the projected cost to make such a movement by rail and/or truck. For each of the four locks analyzed, the models estimate the direct shipper supply chain cost burden if barge service becomes unavailable.

74 GEOGRAPHIC DISTRIBUTION OF THE DIRECT AND REGIONAL IMPACTS GIS tools were employed to illustrate the network-wide distribution of impacts of individual lock closures. Finally, the broader anticipated regional impact—lost incomes and lost jobs—were modeled. IMPLICATIONS The connectivity of the inland system is its strength and its vulnerability. This research explored the use of models and GIS methods that illustrate the economic and networkwide impacts of project disruption, an impact that is estimated to exceed $1 billion across several dozen states at each of the four locks analyzed. Modeling Dynamic Behavior of Navigable Inland Waterways Heather Nachtmann, University of Arkansas The inland waterway transportation system, an integrated part of society, the economy, and the environment, provides a variety of ancillary benefits, including flood protection, power generation, recreation, water supply, and habitats for fish and wildlife. This system, however, is vulnerable to natural disruptions, system component failures, and manufactured attacks. Consequently, it is important to understand inland waterway transportation system behaviors to reduce associated risks and mitigate economic losses. Studying the behavior and economic impacts of the inland waterway transportation system is challenging because of high degrees of complexity and uncertainty. Therefore, comprehensive modeling techniques are required to represent accurately the complex relationships between system components and how these relationships influence economic impacts. The maritime transportation simulation decision support tool (MarTranS) integrates agent-based modeling, discrete-event simulation, and system dynamics with a multiregional input–output model to understand better the relationships between inland waterway transportation system components and economic impact factors. To demonstrate these relationships through this model, the McClellan–Kerr Arkansas River Navigation System was presented as the case study region. MarTranS is generalizable to any inland waterway transportation system to enable maritime transportation stakeholders to better allocate investment budgets and increase economic benefits. Quantifying the Impacts of Disruptions to the Inland Marine Transportation System Patricia DiJoseph, U.S. Army Corps of Engineers The U.S. inland MTS provides a critical means of importing and exporting goods and commodities to and from the interior states. The movements of these goods rely on navigation channels and infrastructure, such as locks, maintained and operated by USACE.

75 Availability of the navigation channels and infrastructure can be impacted by natural events or infrastructure failures with minimal prior notice. These disruptions can affect the supply chain and often make news as shipments are delayed behind ice jams, lock failures, or obstructed navigation channels. Each of these disruptions requires prior planning so that a response framework can be activated and traffic flow can be restored. Quantification of the disruption can provide insight to the true cost of the events and information for operations and maintenance decisions. This presentation demonstrated the use of AIS ship-tracking data to estimate the impacts in terms of increased travel times to shippers during a river closure on the Ohio River during the soybean harvest in October 2017. Understanding not only the impacts at the site of the disruption, but also the cascading impacts up and down the river system, allows for a better quantification of the cost savings when infrastructure functions properly and response is executed quickly. Session 1C Decision Support: Resilience 2017 Hurricanes: A Resilient Path Forward for Marine Transportation System Federal Agencies Katherine Touzinsky, U.S. Army Corps of Engineers, Engineer Research and Development Center The MTS plays a critical role in U.S. commerce by facilitating the movement of more than 2 billion tons of goods annually. The MTS is also particularly susceptible to the impacts of weather and climate hazards. This vulnerability has been illustrated by the impacts of major storms during the 2017 hurricane season. The Committee for Marine Transportation Systems recognized the critical impacts of these storms and requested that the Resilience IAT review some findings to provide recommendations for a more resilient MTS. This presentation discussed their interagency effort to understand better the impacts, federal response, and recovery actions across the MTS after three major hurricanes: Harvey, Irma, and Maria. These findings were gathered and presented during a workshop in late spring 2018 to allow attendees to review and identify some critical actions, challenges, and successes during the response and recovery missions. After identifying the major actions that inform the MTS for resilience to storms, researchers and data professionals embarked on an effort to tie those actions with data-driven metrics that allow a retrospective analysis of port resilience. This presentation provided preliminary results on the impacts, recommendations, and port resilience metrics derived from the 2017 hurricane season.

76 A GIS Inventory and Exposure Assessment for Critical Coastal Transport Infrastructure Land Use in the Caribbean Small Island Developing States Gerald Bove, University of Rhode Island This presentation shared preliminary work from the creation of a database and study of Caribbean port and transport network exposure to climate change and storm damage. Climate change has emerged as one of the most significant challenges to global sustainable development, particularly for small island developing states. In the Caribbean region, one of the most natural disaster-prone regions worldwide, no comprehensive data source exists to identify at-risk transport infrastructure lands. This project is developing a baseline inventory and analyses of coastal transport infrastructure and hazard risk assessment of coastal transport facilities. In collaboration with the Joint European Research Centre dynamic storm modeling team, the project identifies exposure risk and incorporates it into a weighted network model to rank critical assets in the network of ports in the Caribbean region under different scenarios of climate change on the basis of representative concentration pathways and storm surge. Modeling the exposure of individual ports and incorporating information about the structure, flow, and capacity of transport networks allows for prioritization of ports for reinforcement. Barriers to Climate and Extreme Weather Adaptations for Seaports: A Cultural Consensus Model for North Atlantic Medium and High-Use Port Decision Makers Elizabeth Mclean, University of Rhode Island This study addressed the question, what are the perceived decision-making barriers to climate and extreme weather adaptation among seaport decision makers? Decision-making barriers impede, prevent, or delay the implementation of climate and extreme weather adaptations for many aspects of society. Barriers include conflicting timescales and priorities, limited financial resources or training, uncertainty of societal costs and benefits, and fragmentation between scales of governance. Climate and extreme weather impacts, such as heavy rains, storms, sea level rise and extreme heat, particularly threaten coastal critical infrastructure. This study identified common decision-making barriers to ultimately improve seaport resilience and thereby protect the broader economy, environment, and port community. Researchers interviewed port directors/managers, safety officers, and environmental risk officers from the 22 medium and high-use ports of the USACE North Atlantic Division. Researchers then used a cultural consensus model to assess trends and patterns across the region and across subsets of decision makers. Results can be used to target appropriate interventions that assist ports to overcome the identified barriers. For example, decision makers may need data to help with planning, financial assistance to offset costs, or educational materials to help them understand their risk levels.

77 Using Geographic Information Science to Evaluate Legal Restrictions on Freight Transportation Routing in Disruptive Scenarios Steven Peterson, Oak Ridge National Laboratory Disasters have consequences and freight transportation is not immune to these consequences. In the aftermath of disasters, planners and policymakers use scarce resources and work within legal frameworks to ensure that inoperable infrastructure assets return to normal operations. For freight transportation, the challenges associated with freight rerouting due to inoperable infrastructure assets extend beyond the physical dimension. Challenges include overcoming legal barriers involved with intermodal freight transportation. This presentation summarized an application of transportation routing analysis to evaluate options for freight rerouting during disasters. The paper also evaluated the legal implications of the Merchant Marine Act of 1920, commonly known as the Jones act, on short sea shipping between coastal points in U.S. territorial waters. With the closure of the Port of New York and New Jersey during Hurricane Sandy as a case study, modal studies were performed to highlight route options and provide insight into the challenges of modal restrictions imposed by the Jones act. Session 2A Data Analytics: Maritime and Freight 2 Container Ship Routing and Scheduling with Multiple Time Windows Anastasios Charisis, Florida Atlantic University Containerized transportation is an integral part of the world’s economy, as it is the major system that supports supply chain and logistics. Two components play a key role in the design and successful operation of container ship networks: the delivery of products on time, especially food products, and the minimization of costs associated with the operation of the systems. The objective of the study was to define the optimal routes, number of vessels required, and arrival and departure times for a liner container shipping service performing deliveries in a multiport network on a weekly basis. A mathematical framework for the capacitated vehicle routing problem with time windows was formulated as a (nondeterministic polynomial time (NP) hard optimization problem, which enables network operators to set more than one appropriate arrival time interval at each node of the network to accommodate berth availability times as well as the product arrival deadlines. Because of the computational complexity of the problem, a metaheuristic method was selected to find optimal routes, number of vessels required to satisfy demand, and departure and arrival times in each port to minimize operational costs and best serve the network. Numerical experiments showed the ability of the solution algorithm to find good-quality solutions in short computational times. Finally, a sensitivity analysis was performed to show the robustness of the model using different problem variables and parameter values.

78 Modeling Agricultural Commodity Flows on the U.S. Railroads and Inland Waterways System Using Waybill and Waterborne Commerce Statistics Data Steven Peterson, Oak Ridge National Laboratory The Oak Ridge National Laboratory (ORNL) Geographic Information Science and Technology Group created an operational routing tool called the Web-Based Transportation Routing Analysis Geographic Information System (WebTRAGIS) model. This model provides routing analysis capabilities for highway, rail, and barge transportation modes within the continental United States. ORNL was tasked by the USDA Agricultural Marketing Service in to use WebTRAGIS to model the transportation of agricultural products in the United States. The objective of the project was to describe the current multimodal freight transportation system for agricultural commodities through a series maps illustrating the O-D flows of these commodities by rail and by barge/waterway using rail waybill data and Waterborne Commerce Statistics (WCS) data. This presentation focused on the methodologies for transforming the waybill and WCS data into O-D matrices for input into WebTRAGIS, the associated route output, and some preliminary results examining agricultural commodity flows in the United States. Short Sea Shipping Versus Trucking: A Cost–Benefit Analysis Using Mathematical Modeling Evangelos Kaisar, Florida Atlantic University Throughout history, mobility and transportation have played a key role in shaping the world. Because of the expansion of communities, land transportation has been the preferred method for transporting people and goods. Many areas experiencing critical congestion conditions need to implement further measures to ease the strain placed across the transportation system. One solution relies on the redistribution of traffic from the current highway systems to other modes of transportation, such as mass transit for the movement of people, and underutilized modes for the movement of freight. With congestion, environmental impacts, and the price of oil influencing businesses and individuals, measures are required to accommodate the growing demand for freight transportation. With many trucks traveling along the National Highway System redirected to the Marine Highway Corridors developed by the USMA, many problems can be addressed in the short and medium terms. To do so, short sea shipping, through the use of roll-on/roll-off vessels, can be implemented. Although the benefits of reducing environmental problems and congestion are considerable, the profitability of this transportation mode is an important consideration. A cost–benefit analysis can determine the margin of profit and attract investors and businesses. By developing a mathematical model that accounts for the costs associated with transporting trucks along a particular corridor, the competitiveness of short sea shipping can be determined.

79 Session 2B Decision Support: Managing Flows Applying Multimodal Freight Network Optimization to Public-Sector Investment Decisions Mark Berndt, Quetica This presentation discussed the formulation of a methodology to examine network investments that reduce shipping costs and improve network efficiency. The Public-Sector Supply Chain Network Optimization Model begins with building a multimodal network module for key freight modes and nodes serving the area to be studied. The demand module is then created by integrating public- and private-sector freight data. Freight demand is measured by commodity origin/destinations, volume, and the service needs of customers. County-level commodity flows are derived from the FAF, Surface Transportation Board Waybill data, and private-sector bill of lading data. Quetica also draws from supply chain price and performance benchmarks that it has assembled from more than 2 decades as a freight auditor. Stakeholders are engaged in the study process to explain project goals and request access to company shipping records under nondisclosure agreements. Stakeholder outreach also helps identify the possible demand changes and prioritize scenarios and contributes to service-level metrics under each network scenario to be tested by the model. Once the data and model are assembled, a baseline run simulates how freight currently moves into, out of, and through a region or port. Baseline optimization identifies opportunities to improve the existing network without changing the underlining network. User-defined scenarios are run to identify alternative designs to meet objectives such as cost reduction or increased speed to market. Design alternatives are evaluated by using return on investment metrics to prioritize investment strategies. Data Science Approach to Bottleneck Identification: Freight Network Analytics Catherine Lawson, University at Albany Freight networks are plagued with bottlenecks that can affect freight stakeholders differently. The New York State DOT has addressed the challenge of high granularity analysis of bottlenecks in collaboration with AVAIL at the University at Albany. AVAIL has developed an open-source analytics platform, leveraging the NPMRDS at both the state and local levels, for transportation planners in state and metropolitan planning organizations. In addition to providing the federally mandated system performance measures (PM3; see FHWA Transportation Performance Management Rules), the tool suite highlights several bottleneck algorithmic approaches. Specifically, the tool provides rankings for bottlenecks most likely to impact travel for single-unit trucks (urban freight) separately from tractor

80 trailer combination units (long-haul operations), in addition to generating a series of metrics, visualizations, and instantaneous mapping capabilities. Multiagency Data Fusion Informs Waterway Management Brandon Scully, U.S. Army Corps of Engineers USACE maintains a nationwide portfolio of coastal port projects. This portfolio overlays AIS data from USCG and water level measurements resulting from NOAA missions. By integrating data from these three agencies, USACE has developed approaches to understand waterway user behaviors better. Ongoing research by USACE has identified methods for mining the data to inform management and funding of project maintenance. By analyzing the arrival patterns of vessels with respect to water level conditions, USACE is able to infer the extent to which users are taking advantage of additional channel depth provided through tidally driven water levels. By incorporating channel shoaling and maintenance dredging data, USACE can infer the extent to which users may take advantage of additional depth provided through dredging. With quantitative characterization of the level of service provided to varying classes of operators, reliability of navigation projects can be determined. Ultimately, this analytical approach may contribute to improvements in waterway management by better aligning maintenance practices with user profiles. The Voyage Plan: The Missing Link Brian Tetreault, U.S. Army Corps of Engineers For a transportation system to operate efficiently and to enable optimization, information is needed to assist users and managers of the transportation mode. This information includes the current status of the applicable infrastructure, short-notice changes to infrastructure status, real-time monitoring of users of the system, and data to help predict near- and long-term system status. Most transportation modes make use of (or mandate) advance notification of the movement of network users, for example, flight plans and rail schedules. Even for surface transportation, while not having broad requirements for advance notice of movement, deviation from normal traffic flow and density can be determined through observation of multitudes of vehicle movements, that is, crowd-sourcing anticipated vehicle movements, such as prediction of rush hour congestion. In maritime transportation, there is no equivalent to the flight plan or rail schedule, and the number of vessel movements is too few and widely variable to provide any high-resolution congestion prediction. If vessels provided voyage plan information, much more accurate information on anticipated waterway usage could be determined, and information needed

81 by vessel operators to transit safely, efficiently, and reliably could be provided on an as-needed basis. This presentation examined the need for planned vessel movement information, that is, a voyage plan. The potential benefits to waterway managers, operators, vessels, and intermodal connections were discussed, as were issues associated with collecting voyage plan information and potential models for submission or creation of voyage plans, along with initial information needs, opportunities, and barriers. Session 2C Data Analytics: Port Performance Container Ship Bay Time and Crane Productivity: Are They on the Path of Convergence? Shmuel Yahalom, SUNY Maritime College Container ships are becoming larger, and their time at the pier for the D&L of containers is increasing because of larger bays. A key factor in reducing bay time is the gantry crane productivity (lifts per hour) of the D&L. This function requires a match between the growth in container ship bay time and the growth of gantry crane output to maintain a constant container ship bay time. This presentation addressed the relationship between the growth in container ship bay time and the growth in gantry crane productivity to determine the long-term relationship between the two, with the use of the container ship bay time factor model developed by the authors. The relationship between crane productivity and bay time was quantified and indicated no convergence. However, after redefining gantry crane output, the research indicated that convergence is possible. A model that was presented captured these two issues and addressed their behavior for different container ship classes. The model tested the relationship between the two variables to determine long-term trends. On the basis of the results, slow growth of gantry crane productivity indicated that other D&L alternatives were necessary to keep D&L time of large container ships efficient. The alternatives included alternate and partial stowing of bays, new D&L technologies, such as new spreaders and Fastnet, and an increased number of ports of call.

82 Container Ship Dwell Times Through the Automatic Identification System Lens Daniel Smith, Tiago Group, and Daniel Hackett, Hackett Associates Container ship dwell times are a vital concern of U.S. ports, ocean carriers, marine terminals, and importers/exporters. Dwell times are at the core of port productivity and capacity and are a key factor in worldwide container shipping reliability. Availability of complete and consistent vessel position data through the AIS creates a new lens through which dwell times can be analyzed. This presentation began by exploring dwell time issues and their implications for port and vessel productivity. Drawing on an extensive analysis of container ship dwell times in the Port Performance Freight Statistics Program of BTS, the presentation detailed the processing steps necessary to compile usable dwell time data. U.S. container port and vessel dwell time summaries were presented for more than 18,000 vessel calls in 2016, and the relationship of dwell time to vessel size, cargo volumes, deployments, weather events, and other key variables was analyzed. AIS coverage of complete U.S. vessel activity provides both a richer data set and greater confidence in findings than is possible with sample data. The ability to follow vessels and vessel class dwell times over multiple calls, multiple ports, and multiple coasts provides new insights into the dynamics of vessel and container port performance. Finally, the presentation described what might be accomplished as AIS data are linked to trade data and resources. Ports of the Future: Deploying Emulation and Real-Time Simulation for Identifying Technologies for Improved Port Supply Chain Performance Lawrence Henesey, Blekinge Institute of Technology Techniques and technologies for improving port and terminal performances while assisting in the transformation to ports of the future were presented. Current and future technologies disrupt various aspects of conducting port operations within supply chains. An underlying premise is that hyperconnected ports and terminals will creatively deploy smart technologies to deliver sustained value to customers and stakeholders in the port, shipping, and broader logistics communities. The use of emulation and real-time simulation offers advantages in analyzing technologies and understanding relationships with various actors that are interconnected in port supply chains. The results from this research will assist ports and terminals in determining which automation and digital technologies to adopt. The Application of Freight Fluidity Metrics to the Port Environment Kenneth Mitchell, U.S. Army Corps of Engineers USACE is responsible for the maintenance of federally authorized navigation channels and associated infrastructure. As such, USACE requires objective performance measures for determining the level of service being provided by the hundreds of maintained navigation projects nationwide. The U.S. Army Engineer Research and Development Center has

83 partnered with the Texas A&M Transportation Institute to develop a freight fluidity assessment framework for coastal ports. The goal is to use archival AIS data to develop and demonstrate how ports can be compared objectively in terms of fluidity, or the turnaround time reliability of oceangoing vessels. The framework allows USACE to evaluate maintained navigation project conditions in conjunction with port system performance indices and thereby provide insight into questions associated with maintaining required channel dimensions. This presentation covered AIS data inputs, quality control, and performance measures development for U.S. ports with a demonstration of the methodology at the Port of Mobile, Alabama. The performance measures were analyzed under different operational and environmental conditions such as vessel type, vessel size, pilot safety guidelines, and weather impact. The case study showed that vessel type and size have a substantial influence on the performance indicators. The pilot safety guideline restricted vessels movements on the basis of their dimensions to navigate the channel. A restriction could be more disruptive than others to channel traffic. The analysis cascaded down from the more restrictive to the less. This work provides foundational knowledge to practitioners and port stakeholders considering improving supply chain performance. These results are also valuable for researchers interested in the development and application of multimodal freight fluidity performance measures. Session 2D Decision Support: Safety An Exploratory Study of Near-Miss Events in Maritime Freight Systems Using the Marine Information for Safety and Law Enforcement Database Robin Dillon-Merrill, National Science Foundation Most transportation industries have seen decreases in accidents over time. One possible factor, learning from near-miss events, was presented. Multiple sources of data exist to examine the role of near-miss events, but more needs to be analyzed. This presentation specifically addressed that gap, considering near misses that are currently captured in the MISLE data system. Designing new near-miss reporting systems for maritime incidents relevant to the USCG was outside the scope of this study. Three data sets in MISLE were used and are defined as follows: • Notifications that do not become preliminary investigations, • Preliminary investigations that do not become casualties, and • Casualties where no person was injured or killed and no vessel damage occurred.

84 Serious casualty incidents are also present in the data set, and the goal was to understand the relationship between the occurrence of near misses and the occurrence of serious casualty incidents. Three serious casualty incident metrics were identified: (1) fatalities, serious casualty incidents in which at least one person has been killed; (2) total vessel losses, serious casualty incidents in which at least one vessel in the incident has been declared a total loss; and (3) severity score, counts of severity scores associated with incidents. The analysis used MISLE data from 2007 to 2016 for ports identified as principal ports by USACE. This presentation focused on data modeling techniques to examine the relationship between near misses in MISLE and future serious casualty incidents related to the maritime freight system. Examining Maritime Casualties from Vessel Groundings Using Ordered Probit Modeling Fatima Zouhair, U.S. Coast Guard Injuries in the maritime industry are common. Accidents are occurring from incidents such as groundings and collision or resulting from other factors that periodically occur. USCG is responsible for promulgating regulations that prevent and mitigate maritime accidents. Because of loss of life, property damage, and adverse environmental effects, these incidents are critical to both decision makers and policymakers. The objective was to assess the injuries by establishing a relationship between injuries and vessel groundings. The research was based on a zero-truncated model for the response variable, injuries; the values cannot be zero. Other competitive techniques were used for comparison purposes. Last, the marginal effects were examined to determine the contribution of the explanatory variables on injuries. Accident data were obtained from USCG MISLE database. The marine casualty data covers 1991 through 2016 for vessel accidents that occurred in U.S. waterways and within the exclusive economic zone. The research is intended to help decision makers develop safety policies. From Text to Data: How the U.S. Coast Guard Used Accident Reports for Benefits Analysis Douglas Scheffler, U.S. Coast Guard USCG issues rules relating to maritime safety, security, and environmental protection. Before these rules are issued, USCG, along with other rulemaking agencies, conducts cost–benefit analyses that show how the benefits justify the costs. For rules relating to the maritime system, the cost–benefit analysis should show how the rule would lead to a reduction of accidents and their consequences. However, data on the causes of maritime

85 accidents frequently are not available; this unavailability leads to crucial data gaps for cost–benefit analyses. A text review methodology used by USCG to resolve this data gap for the Final Rule for the Inspection of Towing Vessels, which was issued in 2016, was presented. The rule covered a wide range of topics, including safety management systems, training, propulsion systems, and electrical systems. To generate data needed for the cost–benefit analysis, USCG used a panel of experts to review relevant accident reports and record needed information. To ensure consistency across the panel and compile the data in a format for the benefits analysis, USCG developed a recording tool in Access that consisted of a menu front end and a recording form. Quality control checks embedded in the review process consisted of training of the panel members, a statistical analysis of the collected data, and an expert review. The reviewed output consisted of a series of qualitative risk reduction scores that were input into the benefits analysis calculations. For the maritime research community, this case study demonstrates that a carefully constructed expert review of text material can bridge existing data gaps. Session 3A Data Analytics: Maritime and Freight 3 Cyber–Physical Applications for Maritime Freight Transportation Systems Amirhassan Kermanshah, Vanderbilt University Maritime freight transportation systems play a crucial role in the U.S. economy. Major challenges include environmental sustainability issues and natural hazards that can significantly affect this industry. To achieve more efficient, safe, secure, and sustainable transportation, the maritime freight transportation industry is moving toward greater use of CP applications, which involves using computing software/hardware to control and to monitor physical components in real time (e.g., real-time tracking bulk cargos using GPS). While CP technologies present opportunities for freight management and operations in both the public and private sectors, concerns remain about the potential limitations to CP adoption because of issues involving information fidelity, application scalability, and acquisition/operating costs. Moreover, excessive dependency on CP systems can introduce vulnerability to accidental and intentional security breaches, which is a growing concern driving many freight operators away from investing in backup systems. A comprehensive review of existing and anticipated CP technologies and applications was presented, along with a focus on their role in improving maritime freight transportation management and operations. These technologies were evaluated according to their performance in achieving system efficiency, safety, security, and sustainability and on the basis of a cost–benefit analysis using simulation of potential disruptions to freight systems

86 for the middle Tennessee region. General recommendations on how to make the most appropriate use of this rapidly growing technology were provided for the public and private sectors of freight transportation. How Information Systems and Data Are Used to Control Operations and Influence Management Decisions at the Panama Canal Parsa Safa, Lamar University For more than 100 years, the Panama Canal has been considered one of the seven wonders of the modern world and has long held the reputation as an engineering marvel and innovative leader in maritime logistics technology. Use of legacy systems, in conjunction with the integration of the newest state-of-the-art technologies, forms a conglomerate of individual systems that monitor and control every aspect of the canal’s operations. Management relies heavily on the historical, real-time, and forecasted data from these systems to make decisions that affect current operations and tolls, as well as growth of the Panama Canal. Over the past century, numerous changes and upgrades have been implemented and expansions have been added that make the Panama Canal one of the world’s most high-tech operations. This presentation outlined how information systems and data are used to control operations and influence management decisions at the Panama Canal. The Panama Canal combines basic technology and complex systems to keep canal operations running safely and efficiently while continuing to grow, ultimately keeping the global economy in motion and profitable. From this perspective, the presentation showed how ports around the world might look to the Panama Canal for solutions that they can incorporate to remain relevant and competitive. The session also discussed and analyzed how the Panama Canal Authority uses legacy systems along with a wide variety of technology systems and equipment to collect data and translate the information into information that is used in planning and running day-to-day operations. Transport Analysis Framework of the King Abdullah Petroleum Studies and Research Center: Building a Global Freight Network Model with Satellite and Automatic Identification System Data Hector Guillermo Lopez-Ruiz, King Abdullah Petroleum Studies and Research Center KAPSARC has been working on building an open source approach to use freely available AIS data and open source satellite nightlight images to build a world freight network model to better understand freight mobility related to Saudi Arabia's economic objectives. This model is open source and is being used to develop an open app that anybody can use. The approach relies on satellite images in three ways: • Identification of economic activity hubs as nodes in the global freight network, • As a measure of the intensity of human and economic activity, and

87 • As a proxy to estimate transportation flows over main shipping routes. This open source approach, coupled with traditional transportation modeling, provides an understanding of the configuration of global freight movement for types of commodities according to their market value. This understanding, in turn, provides the ability to model different futures for global freight transportation and to build different scenarios for different infrastructure and policy options that concern the Saudi Arabian economy. This presentation explained how this approach is used to assess transportation policies to establish Saudi Arabia as a global logistical hub by developing its transportation infrastructure to increase economic diversification. The following underlying questions were addressed: • What is the methodology used to transform satellite data and AIS data into a global freight model? • How does the land-based freight in the Kingdom of Saudi Arabia tie into the global network and its current efficiency? • Why are certain developments in infrastructure (land, sea, and air) and freight policy needed to meet the Kingdom of Saudi Arabia’s goals for economic diversification and ascension to a global hub with respect to global economic changes? Session 3B Big Data and Machine Learning: Maritime Applications Enhancing Database User Experience with Natural Language Processing Dan Seedah, Texas A&M Transportation Institute The ability to retrieve accurate information from databases without an extensive knowledge of the contents and the organization of each database is extremely beneficial to the dissemination and utilization of freight data. Natural language processing (NLP) is an area of research that explores how computers can be used to understand and to manipulate natural language text or speech to perform tasks. This is an active and growing research field; its theories and technologies power products such as Amazon’s Alexa, Google Assistant, Apple’s Siri, and Microsoft’s Cortana. The potential in NLP applications lies in the ability of users to ask questions in conversational language and to receive answers rather than trying to formulate a query into sometimes unfriendly unnatural formats that machines can use to query a database. The challenges of NLP are (1) correctly identifying only the relevant information and key words from questions when dealing with multiple sentence structures and (2) automatically retrieving, preprocessing, and understanding multiple data sources to determine the answer to a user’s query. This presentation illustrated how to overcome the challenges of NLP use in the freight transportation domain

88 and provided a demonstration of how FHWA and other related freight data sets can be queried with the use of natural language. Man Versus Machine: Comparing Traditional Data Collection and Statistical Models with Machine Learning Big Data Analytics Jolene Hayes, Fehr & Peers More and more data sets continue to enter the market, each with a promise of telling transportation planners exactly what is happening in the movement of goods and people. But how good are the data? Who is compiling it, and, more important, how are the data packaged for end users? As new and larger data sets become available, how is AI being used to inform what the data mean? Are machines getting it right, or more appropriate, are the data miners and programmers getting the machine learning right? And are the data statistically representative, especially with trucks and freight moves? This presentation shared recent case studies describing the following common errors and biases with these new big data analytics: • Data collection issues: systemic errors (sample bias/underrepresentation of segments of truck population) and data accuracy/validity; • Sample issues: size of data set, outliers and what they tell us, and selection bias; • Statistical models and analytics tools (AI, machine learning process, data visualization, data collection): wrong assumptions, over-/underfitting, nonnormality outliers, unknown impacts of confounding variables, and uncertainty of future trends relative to historic patterns. The Practical and Unified Use of Blockchain, Cybersecurity, Internet of Things, Machine Learning, Artificial Intelligence, and Big Data for the Marine Transportation Industry: Details on Building a Multimodal Freight Analytics Platform Dean Shoultz, MarineCFO Advancements in technology, including mobility, communications, the cloud, data storage, ingest, and AI, have created revolutionary opportunities for the maritime industry. Technology buzzwords are everywhere: Blockchain, big data, machine learning, public cloud, mobility, AI, and more. There are steep challenges in binding these technologies together to solve maritime problems. The process is more art than science. Machine learning is meaningless without big data to learn from. Big data do not accumulate without IoT and productive applications. AI and machine learning are invaluable in analyzing a supply chain, but without a Blockchain to secure contributed transactions, few will contribute. Mobile apps without a structured, secured, and scalable ingest methodology are islands to themselves.

89 Each of these technologies is subject to the growing threat of cybersecurity vulnerabilities. Each has a unique securable surface, and each is vulnerable to different methods of attack. Another challenge involves the sheer pace of technological innovation. New technologies appear and disappear daily. Navigating this ever-changing landscape is difficult. Classic apps were monolithic silos built on technologies destined to become antiquated. Modern systems need to be service-oriented for future-proofing and to facilitate increased innovation, accessibility, security, reliability, and consumption. The goal of this presentation was to show, in simple, nontechnical terms, how these pieces can be stitched together to solve a specific maritime problem. In short, the presentation illustrated the following: (1) a blueprint and methodology that can be used as a guideline for any technology initiative and that provides step-by-step guidance on building a multimodal freight analytics platform using these technologies; (2) the interrelationships between them in the solution, the cybersecurity challenges of each, and how to mitigate them; and (3) how various business entities play a part in the process. Validating AIS Data with Machine Learning Algorithms Edward Carr, Energy and Environmental Research Associates Machine learning techniques are being developed and big data visualization tools are being used to identify patterns better in maritime AIS vessel tracking data. AIS data contain a combination of automatically generated data fields, such as vessel position, speed, and course, in addition to hand-entered data fields that include operating status and vessel type. While big data such as AIS are important to modelers, AIS data are prone to errors. When errors are encountered, it is common to trim or to omit erroneous data or in some cases to apply manual corrections. While correcting is generally preferred, this is often not possible without simultaneous corroborating data or resampling over a separate time period. Trimming and omitting are often chosen in current work, but these choices raise concerns about issues such as representative sampling and precision at small scales. We are working on reclassifying data with the use of advanced algorithms from other big data tracking efforts. Insights are gained from the animal behavior literature to build models and to help identify vessel behavior better, providing a method to validate or update the vessel status field and better inform vessel emissions models. In addition, visualization is an important step for validating model results for large data. This model, its background, and results were presented along with big data visualization tools, which are invaluable to researchers working in AIS-related fields.

90 Session 3C Decision Support: Environmental Automatic Identification System Data Improvements to the 2014 Gulfwide Emissions Inventory Study Heather Perez, ERG Because the quality of AIS data has improved drastically over the past several years, AIS has become increasingly useful as the primary data source for assessing vessel traffic and marine freight movements. This presentation summarized the evaluation of the quality of hourly AIS observations that formed the basis of the commercial marine vessel component of the 2014 Gulfwide Emissions Inventory Study for BOEM. Tens of millions of observations associated with nearly 9,000 vessels were mapped to perform quality checks on the data, identify activity and emissions hot spots by vessel type, and investigate seasonal and hourly traffic patterns. The compiled AIS data underwent extensive quality checks and vetting, including comparison of vessel operation to other independent data sources (e.g., BOEM pipelaying GIS data files), validating both spatial and temporal data elements. The ability to identify specific vessels, document their location, and calculate true engine load factors and hours of operation provided a more complete inventory of vessel operations. For example, AIS identified twice as many support vessels in the 2014 inventory than in the 2011 inventory, which was based on a trade association census. Many of these additional vessels had half the power that support vessels reported in the 2011 inventory. But, more important, AIS documented that these vessels spent more time idling at sea (28%) than assumed in the 2011 inventory (15%). Use of AIS data provides significant improvement over approaches that rely heavily on surrogates for vessel operations. These enhancements are a crucial step toward developing more accurate activity and emissions calculations. Merging Automatic Identification System and U.S. Coast Guard Data for Modeling Maritime Air Emissions Diane Rusanowsky, U.S. Coast Guard AIS was originally developed to enhance maritime safety and security. AIS is an automatic tracking system used on ships and by vessel tracking services to determine the unique identification, position, course, and speed of a given vessel, which allows maritime authorities to monitor and track movements of vessels fitted with an AIS transceiver. As AIS signals are not encrypted, they may be captured and stored to facilitate conducting maritime research. From 2016 to 2017, USCG successfully developed a vessel emissions model that combined AIS data signals vessel characteristics with data from the USCG’s databases. As a proof of concept, the model was run to produce emissions estimates for the Delaware River Valley, Pennsylvania; San Pedro Bay, California; and the western Gulf of Mexico.

91 This air emissions model is based on an estimate of engine load factor and is capable of producing a granular estimate of air emissions. This presentation described the model’s underlying equations, data sources, and outputs; discussed significant challenges that were encountered as discrepancies in the vessel information and position signals; and summarized the protocol developed to prepare the data and to exclude unreliable information. Key findings from the model runs were also presented. In particular, the emissions profiles in all three geographic areas are different and reflect the maritime uses in each area. The presentation concluded with a discussion of potential improvements in the model and problems that USCG has identified related to scaling the model to larger regions or to the national level. Multiattribute Performance Analysis for Intermodal Maritime Cargos Jim Corbett, University of Delaware This presentation outlined the importance of integrating environmental performance metrics, such as vessel and port inventories of emissions, with intermodal performance metrics. A framework in which mode cooperation can be considered was presented. Economic factors that may be important to supporting environmental decision-making were introduced for discussion. How public good outcomes, such as health benefits, can contribute to decision-making using benefit–cost frameworks was also presented, as was the importance of understanding goals in terms of the shared performance goals, location-specific tradeoffs, and firm opportunities.

Next: Appendix C. Conference Participants »
Conference Proceedings on the Web 22: Transforming the Marine Transportation System Through Multimodal Freight Analytics Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB has released Transforming the Marine Transportation System Through Multimodal Freight Analysis: Proceedings of the Fifth Biennial Marine Transportation System Research and Development Conference that summarizes the discussion about multimodal freight transportation that took place on June 19-21, 2018.The conference considered potential research to address issues associated with transforming the marine transportation system (MTS) and explored opportunities to harness robust multimodal freight transportation data and analytics. Sessions at the conference focused on multimodal freight operations, planning, and policy; challenges associated with the corresponding analytics; and using these analytics for strategic MTS planning. This publication summarizes the presentations and discussions from the workshop.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!