Project Number:
08-13
Research Project:
Emergency Traffic Management Tool for the Los Angeles and Long Beach Harbor Area
P.I. Name & Address:
Darin Goldstein
California State University at Long Beach
Department of Computer Science
Long Beach, CA 90840
Email: dgoldste@csulb.edu
Website: http://www.csulb.edu/colleges/coe/cecs/views/personnel/fulltime/goldstein.shtml
Phone: (562) 985-1507
Fax: (562) 985-7823
Co-P.I.
Han Dam Le-Griffin
University of Southern California
Department of Civil and Environmental Engineering
Kaprielian Hall, Room 234A
Los Angeles, CA 90089-2531
Email: hdle@usc.edu
Website: http://www.usc.edu/dept/civil_eng/dept/faculty-staff/faculty-directory/le-griffin-hanh-dam.htm
Phone: (213) 281-2905
Fax: (213) 744-1426
Project Objective:
Since September 11, 2001, transportation agencies and the maritime industry have had to concentrate on safety and security. General awareness has increased, and many protective measures are in place.
The United States is a maritime nation. More than 90% of U.S. imports and exports are delivered by ship. Located in one of America's largest and fastest growing urban centers, the ports of Los Angeles and Long Beach, with their vibrant roadways and intermodal transportation links, are major gateways of Sino-American trade. The Los Angeles and Long Beach ports combined are the busiest container ports in North America and are a microcosm of the nation's maritime security challenges including security considerations for containers, cruise terminals, power generation, critical transportation links, and population and financial centers. Keeping the gateways to America secure and guaranteeing the continued flow of commerce is a top priority of government and industry.
As container volumes continue to increase at the nation's major gateway ports of Los Angeles and Long Beach, state agencies, port authorities, terminal operators, and local and regional landside jurisdictions are grappling to form new institutional arrangements to identify project priorities and funding to improve both safety and security in the goods movement system. In responding to the mandate set by the Maritime Transportation Security Act of 2002, the Central California Area Maritime Committee (AMSC) was established in February 2004. One of the committee's goals is to advise on the development of a continuous overall port security evaluation process that includes contingency plan development and dissemination of maritime security related material. Moreover, programs in the operative federal highway legislation, SAFETEA-LU, provide incentives for the development and implementation of real-time traffic management systems in all states to enhance security with the capacity to provide pertinent information to the traveling public and emergency responders in the case of an incident, natural or manmade. The Secretary of the Department of Transportation has issued a directive specifically for establishing real-time traffic management capabilities and a standard platform for exchanging transportation system performance information with local and state jurisdictions, as well as with the general public.
Currently, standard methods of evacuating vehicles from a potential hazard area at the port involve directing traffic flow towards preset evacuation routes. These evacuation routes are chosen beforehand based on various criteria and made very public. In terms of security, this method works out fine if an "accidental" emergency occurs away from the evacuation routes. However, post 9-11, the maritime community is nowadays quite concerned with man-made disasters instigated by terrorists. In the case of a terrorist attack, a great deal of damage to life and property can be undertaken simply by sabotaging these evacuation routes. Generally, these evacuation routes can quickly become gridlocked in the case of a real disaster; making these preset routes public just allows the enemy to plan more effectively. The key objective of this research will be to deny the enemy the chance to do such damage by researching the possibility of an artificially intelligent traffic management tool. Assuming vehicle drivers follow the instructions given by this tool, preset evacuation routes should become unnecessary and gridlock will be maximally avoided in the transportation network in the case of a disaster. This proposed study will contribute to and support the national efforts underway to enhance port security across the country.
This proposed work builds upon METRANS-funded research that has been conducted and is presently under way by the authors. These studies include research currently being completed relating to convoy routing in an emergency environment (METRANS AR 06-03), managing empty container flow in urban commercial corridors using short sea shipping and regional port systems (METRANS 04-05), and a number of other research efforts relating to port operations, including assessing container terminal/port productivity at LA/Long Beach in comparison with other US and international ports (METRANS AR 05-06) and understanding the logistics of empty container flow in Southern California region (METRANS 02-03). The convoy routing study considers a novel mathematical model for routing convoys in an emergency situation. The model as stated contains many fewer variables than any previously presented version of the problem and yet still retains enough complexity to be NP-complete[1]. This reduction in the number of variables makes routing convoys through a disaster environment a potentially tractable problem in real-time using modern computer systems and various artificial intelligence techniques. The short sea shipping and port operations and productivity studies provide a basic understanding of commercial port operations and the characteristics of landside distribution networks in detail and propose a method to manage empty container flows in urban congested commercial corridors using short sea shipping and the regional port systems. The proposed project takes advantage of the knowledge base developed as part of these past and existing efforts, considering the nature of port operations and activities and port-related commercial flows as inputs to the proposed algorithm.
Qualitative findings obtained through interviews with study participants and quantitative data arising from the analysis will be presented in a final report and be submitted for publication with relevant academic and industry journals.
Task Descriptions
The study will be conducted through several tasks:
Task 1. Investigate current status of port security and emergency evacuation plans and polices: Knowledge gained through conducting previous METRANS studies on the logistics of containers, terminal activity and productivity, theoretical studies, and other goods movement-related issues will allow us to conduct a basic literature review on disaster evacuation plans. This task will review the current state-of-the-art in disaster route planning: problems and issues discovered in other recent studies will be documented, if they exist. We plan to pay particular attention to the recently completed METRANS study (METRANS AR 06-03) that details algorithm design and analysis for convoy routing for emergency evacuation situations in addition to interviews with port authority personnel.
Task 2. Building the Model: Based on our understanding of the vehicle activity and truck movements within the port and harbor area obtained from Task 1, a fair-sized section of the Long Beach harbor area will be selected and a real-life network model for this particular area will be developed and computerized along with speed limits along each leg of each roadway. Additionally, using established research on car and truck movement, an elementary traffic simulator will be programmed in the C/C++ programming language.
Task 3. Building the Algorithm: An algorithm that respects the 4 issues/principles discussed in section 4 will be devised and added to the model built in Task 2.
Task 4. Simulations: A number of simulations of a disaster environment in the selected area of the port (created in Task 2) will be run using (1) the "perfectly selfish" driver scenario; (2) the "new and improved" driver scenario using the algorithm devised in Task 3, and (3) various combinations of the two; some drivers "perfectly selfish" and some "new and improved." Results of these simulations will be presented in comparison to assess the advantage of the proposed system.
Task 5. Institutional Analysis and Benefit Assessment: To evaluate the potential benefits the implementation of this traffic guidance system could provide, a number of in-depth interviews will be conducted with key stakeholders and local authorities. These interviews will be designed to profile possible problems and issues associated with the implementation of the proposed traffic guidance system, particularly in case of an emergency evacuation incident. The knowledge gained through these interviews will help to further improve the practicality of the proposed system, and will also help to identify the changes in institutional practices that will be necessary for the successful implementation of the proposed system. These findings will also likely highlight the need for coordinating the institution of an area-wide evacuation plan for the Los Angeles and Long Beach ports district.
Task 6. Final Report: A final report will be prepared detailing the changes that occur, through time, with the reestablishment of equilibrium traffic flows around the disaster area. The report will include the lit review, a detailed outline of the algorithm (along with justification of its steps), and a quantitative explanation of the benefits of using this kind of adaptive, real-time system.
Milestones, Dates:
September 1, 2007 – August 31, 2008
Total Budget:
$179,081
Student Involvement:
One graduate student at 50% for 11.5 months
Relationship to Other Research Projects:
Related to 06-03 and 04-05; part of the mobility focus area
Technology Transfer Activities:
Project report will be posted soon
Potential Benefits of the Project:
This research studies the possibility of an artificially intelligent traffic management tool that would reduce gridlock on evacuation routes from the Ports of Los Angeles and Long Beach
TRB Keywords:
Evacuation, Artificial Intelligence, Gridlock, Terrorism, Natural Disaster, Port of Los Angeles, Port of Long Beach
Footnotes:
[1] NP-complete problems are thought to be “intractable” by the computer science community. No efficient algorithms are thought to exist for solving any of these problems exactly. The only methods of determining acceptable solutions are via artificial intelligence and/or approximation techniques.
