The study of user-behavior and decision-making dynamics in transportation network are vital in modeling and simulation of user interactions. Different users access transportation network in order to accomplish different activities. Such activities can be regular commuting, transit services, commercial taxicabs, deliveries, long distance trips, logistics or fleet services, etc. While the world is becoming increasingly urbanized reliable and cost effective movement of people and goods is important for the productivity and economic growth at large. Urbanization and population growth have created the shift in how travel activities are tied to the economy. In today's economy, businesses and individuals are looking for ways of making their fiscal resources and workforce more efficient. However, traffic congestion dampens the efficiency and prosperity by imposing additional operating costs, slowing mobility and causing wastage of time and by hindering efficient metropolitan services such as deliveries, public safety and maintenance. Traffic congestion in the United States in 2011 for instance, caused urban commuters to travel 5.5 billion hours more and to purchase an extra 2.9 billion gallons of fuel (enough to fill Superdome in New Orleans, two times) for a congestion cost of $121 billion. In larger cities and in busy expressways, traffic infrastructures are already operating at near or full capacity. With today's shrinking budgets, often no funding is available to rebuild or expand an aging public transportation infrastructure, making it crucial to devise ways to optimize the performance of existing transportation assets. Since the recurring congestions in large metropolitan areas are mainly due to predictable behavioral activity scheduling, traffic management efforts should be geared towards behavior analysis and modeling. Modeling behavior and decisions, pertinent to route choice and activity scheduling dynamics are crucial for capturing microscopic and mesoscopic nature of traffic flow patterns. In this research, the focus is placed on the development of multi-agent transportation demand estimation and simulation framework to be used by the public entities for performance optimization of existing transportation network and scenario evaluation of new investments. The framework employs several mathematical and statistical methods for the derivation of sampling distributions of users' (i.e., agents') behavior and travel characteristics for the initial network demand generation. The processes of deriving sampling distributions of agents' behavior and travel characteristics largely rely on the quantity, quality and resolution of the available data of the region under study. Travel characteristics/travel surveys data from South East Florida Regional Planning Model (SERPM) region and the National Household Travel Survey (NHTS) data contained individuals' travel characteristics such as origin, destination, departure and arrival time, chain of activities and tours within the trip. These are micro-information needed for the derivation of household and individual agent's travel behavior. The data was processed to develop probability distributions for groups of agents with similar travel behavior, given the agents' household characteristics. In a similar fashion, with agents' household characteristics given, the logit models for agents' activity and locations choices were developed. Besides behavior simulation and demand estimation, the developed framework included an ad-on module for lane choice and pricing approaches applicable to dynamic high occupancy toll (HOT) lanes pricing. The reinforcement learning (RL) approach was used for updating the optimal pricing strategy in a given traffic condition. The pricing controller was configured to start with a predefined base price at a given traffic level, and then in the process of learning, it varies the price in accordance with the acceptable price levels at a given level of service (LOS). In this way, the pricing controller learns the states in which a higher price is more beneficial and those in which a lower price is more beneficial, and then adjusts the parameters of the pricing function to minimize the difference between the current computed price and the posted price. The framework was tested and validated for the scenario based on the data from SERPM region. The scenario was simulated in Multi-Agent Transport Simulation (MATSim). In MATSim, the simulation is constructed around the notion of agents that make independent decisions about their actions. Each traveler of the real system is modeled as an individual agent. Generally, the observation of network traffic evolution from the simulation showed the expected traffic patterns for both morning peak and afternoon peak traffic. One of the most important aspects of travel behavior is the characterization of travel activities by trip duration. The distribution of travel activities by trip duration is the reflection of user behavior in the study area. This determines the expected users departing, en-route, stuck, and arriving to their destinations at a particular time interval. In this research, the simulation results show that network users in our case consist mainly of regular commuters (≥ 20%) whose trips take about 15 minutes. As any other research study, there are some limitations with this work. Due to lack of relevant data, transit use and other modes other than personal vehicle were not considered. Future directions for this research include the inclusion of other data sources and optimization of the demand estimation framework in order to scale-down the computation cost. In addition to the reduction of computation cost, focus will be on development and implementation of modules for simulating dynamic toll pricing on high occupancy toll lanes and assessing the effects of social media information exchange among the agents on mobility. / A Dissertation submitted to the Department of Civil and Environmental Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester, 2015. / January 23, 2015. / Activity-based modeling and simulation, Multi-agent simulation, Network demand estimation, Network optimization, Smart mobility, Travel Behavior / Includes bibliographical references. / Ren Moses, Professor Directing Dissertation; Mark W. Horner, University Representative; Eren Erman Ozguven, Committee Member; John O. Sobanjo, Committee Member; Chiwoo Park, Committee Member.
Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_253474 |
Contributors | Mtoi, Enock T. (authoraut), Moses, Ren (professor directing dissertation), Horner, Mark W. (university representative), Ozguven, Eren Erman (committee member), Sobanjo, John Olusegun (committee member), Park, Chiwoo (committee member), Florida State University (degree granting institution), College of Engineering (degree granting college), Department of Civil and Environmental Engineering (degree granting department) |
Publisher | Florida State University, Florida State University |
Source Sets | Florida State University |
Language | English, English |
Detected Language | English |
Type | Text, text |
Format | 1 online resource (133 pages), computer, application/pdf |
Rights | This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). The copyright in theses and dissertations completed at Florida State University is held by the students who author them. |
Page generated in 0.0018 seconds