1.4 Transportation planning and demand modeling
13 min read•august 21, 2024
Transportation planning and demand modeling are crucial for developing efficient and sustainable transportation systems. These processes involve analyzing travel behavior, forecasting future needs, and evaluating various scenarios to inform decision-making.
Key components include the four-step modeling process, activity-based approaches, and emerging data-driven techniques. Planners use these tools to balance mobility, accessibility, and environmental concerns while adapting to new technologies and changing travel patterns.
Transportation planning fundamentals
- Transportation planning involves determining the future needs for the movement of people and goods and developing strategies to meet those needs
- It is a collaborative process that involves various stakeholders and considers multiple modes of transportation
- Planning occurs at different geographic scales, from local to regional to national levels
Goals of transportation planning
Top images from around the web for Goals of transportation planning
Related content View original
Is this image relevant?
Sustainable Transportation: Accessibility, Mobility, and Derived Demand | Sustainability: A ... View original
Is this image relevant?
Discovering Urbanism: From a mobility to an accessibility orientation View original
Is this image relevant?
Related content View original
Is this image relevant?
Sustainable Transportation: Accessibility, Mobility, and Derived Demand | Sustainability: A ... View original
Is this image relevant?
1 of 3
Top images from around the web for Goals of transportation planning
Related content View original
Is this image relevant?
Sustainable Transportation: Accessibility, Mobility, and Derived Demand | Sustainability: A ... View original
Is this image relevant?
Discovering Urbanism: From a mobility to an accessibility orientation View original
Is this image relevant?
Related content View original
Is this image relevant?
Sustainable Transportation: Accessibility, Mobility, and Derived Demand | Sustainability: A ... View original
Is this image relevant?
1 of 3
- Ensuring safe, efficient, and reliable movement of people and goods
- Promoting accessibility and mobility for all users, including those with disabilities
- Supporting economic development and quality of life in communities
- Minimizing negative environmental impacts (air pollution, noise)
- Integrating transportation with land use planning for sustainable development
Key stakeholders in planning
- Government agencies at various levels (local, state, federal)
- Transportation providers (transit agencies, highway departments)
- Private sector entities (developers, businesses)
- Community organizations and advocacy groups
- General public and users of the transportation system
Planning at different levels
- Local planning focuses on specific neighborhoods, corridors, or activity centers
- Includes traffic calming, pedestrian and bicycle infrastructure, parking management
- Regional planning addresses transportation needs across multiple jurisdictions
- Involves metropolitan planning organizations (MPOs) and regional transportation plans
- State and national planning sets broad policies and funding priorities
- Includes statewide transportation improvement programs (STIPs) and national transportation legislation
Travel demand modeling
- Travel demand modeling is a key tool used in transportation planning to estimate and forecast travel behavior
- It involves mathematical models that simulate how people make travel decisions based on various factors
- The modeling process typically follows a four-step sequence: trip generation, trip distribution, mode choice, and route assignment
Four-step modeling process
- Trip generation estimates the number of trips produced by and attracted to each zone in the study area
- Based on socioeconomic data (population, employment) and land use characteristics
- Trip distribution determines the spatial pattern of trips between origin and destination zones
- Uses gravity models or destination choice models
- Mode choice predicts the proportion of trips made by different transportation modes (car, transit, bike, walk)
- Based on factors like travel time, cost, and personal preferences
- Route assignment allocates trips to specific routes on the transportation network
- Considers factors like capacity, congestion, and travel time
Trip generation models
- Regression models relate trip production and attraction to explanatory variables
- Variables may include household size, income, auto ownership, employment density
- Category analysis models estimate trips based on cross-classification of variables
- Example: number of work trips per household by income and auto ownership categories
- Machine learning models (neural networks, decision trees) can capture complex relationships
Trip distribution models
- Gravity models assume that trip distribution is proportional to the attractiveness of destinations and inversely proportional to the travel impedance between zones
- Impedance is typically measured by travel time or generalized cost
- Destination choice models use a logit formulation to predict the probability of choosing each destination
- Based on utility functions that consider attributes of the destination and the travel path
Mode choice models
- Multinomial logit models predict the probability of choosing each mode based on their relative utilities
- Utility functions consider variables like in-vehicle travel time, wait time, cost, number of transfers
- Nested logit models capture correlations among subsets of alternatives
- Example: transit modes (bus, rail) may be more similar to each other than to driving
- More advanced models (mixed logit, latent class) can incorporate heterogeneity in preferences
Route assignment models
- All-or-nothing assignment assigns all trips to the shortest path between each origin-destination pair
- Assumes no congestion effects and unlimited capacity
- User equilibrium assignment achieves a stable state where no traveler can improve their travel time by unilaterally changing routes
- Solved using iterative algorithms (Frank-Wolfe, gradient projection)
- Stochastic user equilibrium models introduce variability in perceived travel times and route choice behavior
- Accounts for imperfect information and differences in driver preferences
Data for transportation planning
- Transportation planning relies on a variety of data sources to understand travel behavior and system performance
- Traditional data sources have been supplemented by emerging technologies and data fusion techniques
- Data quality, coverage, and timeliness are important considerations in the planning process
Traditional data sources
- Household travel surveys collect detailed information on individual and household travel patterns
- Includes trip purposes, modes, origins and destinations, demographics
- Traffic counts measure the volume of vehicles passing a specific point on the roadway network
- Collected using manual counts, pneumatic tubes, or inductive loop detectors
- Transit ridership data tracks the number of passengers using transit services
- Obtained from fareboxes, automatic passenger counters (APCs), or manual counts
Emerging data sources
- GPS data from smartphones or in-vehicle navigation systems can provide detailed travel trajectories
- Enables analysis of route choice, travel time reliability, and origin-destination patterns
- Smart card data from transit fare payment systems captures boarding and alighting locations and times
- Allows for more accurate measurement of transit use and transfer patterns
- Cellular network data provides aggregate information on the movement of mobile devices
- Can be used to estimate population flows and activity patterns over large areas
Data fusion techniques
- Data fusion combines information from multiple sources to create a more comprehensive and accurate picture of travel behavior
- Matching algorithms link data from different sources based on common attributes (location, time)
- Example: combining GPS traces with transit smart card data to infer complete multimodal trips
- Statistical methods (Bayesian inference, machine learning) can estimate missing data or correct biases
- Example: using survey data to adjust biases in passive mobile device data
Land use and transportation interaction
- Land use and transportation systems are closely interrelated and influence each other
- Land use patterns shape the demand for transportation, while transportation investments can stimulate land development
- Integrated land use-transportation models are used to analyze these interactions and inform planning decisions
Land use impacts on travel demand
- The density, diversity, and design of land uses affect the number and types of trips generated
- Higher densities tend to reduce trip lengths and encourage non-motorized modes
- Mixed-use developments can internalize trips and reduce overall travel demand
- The spatial distribution of activities (jobs, housing, services) influences trip patterns
- Job-housing balance can reduce commute distances and peak-period congestion
Transportation impacts on land use
- Transportation accessibility is a key factor in land use location decisions
- Areas with better access to transportation networks are more attractive for development
- Transportation investments can stimulate land use changes and economic development
- New highways or transit lines can open up areas for growth and redevelopment
- Transportation policies (parking requirements, ) can influence land use intensity and form
Integrated land use-transportation models
- Integrated models capture the two-way feedback between land use and transportation
- Land use models simulate the location choices of households, firms, and developers
- Transportation models estimate the travel demand resulting from those location choices
- Equilibrium models iterate between land use and transportation until a stable state is reached
- Example: UrbanSim and MATSim simulate long-term land use changes and daily travel patterns
- Scenario analysis evaluates the impacts of alternative land use and transportation policies
- Example: testing the effects of transit-oriented development or congestion pricing on mode shares and land values
Activity-based travel demand modeling
- Activity-based models represent travel as a derived demand from the need to participate in activities
- They focus on individual and household activity patterns, rather than aggregate trip flows
- Activity-based models can capture complex travel behavior, such as trip chaining and scheduling
Activity-based vs trip-based models
- Trip-based models consider trips as independent entities, defined by origin, destination, mode, and purpose
- Ignore the temporal and spatial relationships between trips
- Activity-based models view travel as a sequence of activities and trips over the course of a day
- Explicitly model the timing, duration, and location of activities
- Capture constraints and interactions between household members
Tour formation and scheduling
- Tours are chains of trips that start and end at the same location (usually home)
- Work tours may include intermediate stops for shopping or picking up children
- Activity scheduling models predict the timing and sequence of activities and trips
- Based on time-space prisms that represent feasible activity locations given time constraints
- Use utility-maximizing or rule-based decision frameworks
Microsimulation in activity-based models
- Microsimulation models simulate the activity and travel patterns of individual agents
- Agents can represent persons, households, vehicles, or firms
- Agents make decisions based on their characteristics, preferences, and constraints
- Example: choosing the mode and departure time for a work tour based on job start time and availability of a car
- Microsimulation allows for detailed representation of heterogeneity and emergent behavior
- Can model the interactions between household members and the effects of transportation policies on specific population segments
Freight demand modeling
- Freight demand modeling focuses on the movement of goods, rather than passengers
- It involves different decision-makers, transportation modes, and spatial patterns compared to passenger travel
- Freight models are important for planning and managing the transportation of raw materials, intermediate goods, and finished products
Freight vs passenger demand modeling
- Freight demand is driven by economic activity and supply chain relationships
- Depends on the production, consumption, and trade of commodities
- Freight movements involve multiple actors (shippers, carriers, receivers) with different objectives
- Cost, reliability, and delivery time are key factors in freight mode and route choices
- Freight trips are less frequent but cover longer distances than passenger trips
- Intermodal transfers and consolidation are common in freight supply chains
Commodity flow models
- Commodity flow models estimate the quantity of goods moved between production and consumption zones
- Based on economic input-output tables or commodity flow surveys
- Gravity models are commonly used to distribute commodity flows between origin-destination pairs
- Attractiveness is measured by economic variables (employment, output)
- Impedance is based on transport costs or travel times
- models allocate commodity flows to different transportation modes (truck, rail, water, air)
- Based on mode choice models that consider shipment size, value, and time sensitivity
Logistics chain models
- Logistics chain models represent the sequence of activities involved in the movement of goods
- Including production, storage, consolidation, distribution, and delivery
- Agent-based models simulate the decisions and interactions of various actors in the supply chain
- Example: shippers choosing carriers based on cost and service attributes
- Network models optimize the flow of goods through the transportation and logistics network
- Consider capacity constraints, transfer costs, and inventory holding costs
- Integrated models combine commodity flow, logistics chain, and network approaches
- Capture the interactions between transportation demand, infrastructure supply, and economic activity
Model validation and calibration
- Model validation and calibration are essential steps in ensuring the accuracy and reliability of transportation models
- Validation involves comparing model outputs with observed data to assess the model's ability to replicate real-world conditions
- Calibration involves adjusting model parameters to improve the fit between model outputs and observed data
Importance of model validation
- Validation is necessary to build confidence in the model's predictive capabilities
- Models are used to forecast future travel demand and evaluate alternative scenarios
- Validation helps identify model limitations and areas for improvement
- Reveals systematic biases or errors in model assumptions or data inputs
- Validation is required for model credibility and acceptance by decision-makers and the public
- Models that are not validated may lead to flawed planning decisions and misallocation of resources
Calibration techniques and metrics
- Calibration involves estimating model parameters to minimize the difference between model outputs and observed data
- Parameters may include trip generation rates, mode choice coefficients, or route choice parameters
- Least squares estimation minimizes the sum of squared differences between modeled and observed values
- Example: calibrating trip generation rates based on household travel survey data
- Maximum likelihood estimation maximizes the probability of observing the data given the model parameters
- Example: calibrating logit mode choice models based on revealed preference data
- Goodness-of-fit measures quantify the agreement between modeled and observed values
- R-squared measures the proportion of variance explained by the model
- Root mean square error (RMSE) measures the average magnitude of errors
Sensitivity analysis methods
- Sensitivity analysis assesses how model outputs change in response to variations in input parameters
- Helps identify the most influential parameters and the robustness of model results
- One-at-a-time (OAT) sensitivity analysis varies one parameter at a time while holding others constant
- Measures the local sensitivity of model outputs to each parameter
- Global sensitivity analysis explores the entire parameter space using techniques like Monte Carlo simulation
- Quantifies the contribution of each parameter to the overall uncertainty in model outputs
- Scenario-based sensitivity analysis tests the model under a range of plausible future conditions
- Example: evaluating the impacts of different population growth or fuel price scenarios on travel demand
Scenario analysis and forecasting
- Scenario analysis involves testing alternative future conditions and their impacts on transportation system performance
- Forecasting predicts future travel demand and system conditions based on assumed scenarios and model parameters
- Scenario analysis and forecasting are used to inform long-range transportation planning and investment decisions
Baseline scenario development
- The baseline scenario represents the most likely future conditions assuming current trends and policies continue
- Includes projections of population, employment, land use, and transportation network changes
- Baseline scenario development involves assembling data from various sources and models
- Demographic forecasts, economic models, land use plans, and transportation improvement programs
- The baseline scenario serves as a reference point for comparing alternative scenarios
- Helps isolate the effects of specific policy interventions or external factors
Alternative scenario generation
- Alternative scenarios represent different possible futures based on varying assumptions or policy choices
- Examples: compact land use scenario, transit investment scenario, automated vehicle scenario
- Scenario generation involves specifying the key drivers of change and their plausible ranges
- Population and employment growth rates, transportation technology adoption, travel behavior shifts
- Scenarios can be defined through and visioning processes
- Reflects community goals, values, and priorities for the future transportation system
- Scenario inputs are translated into model parameters and network changes
- Example: coding new transit lines or adjusting trip generation rates for a transit-oriented development scenario
Performance metrics for evaluation
- Performance metrics quantify the impacts of scenarios on various transportation system outcomes
- Mobility: travel times, delay, congestion levels
- Accessibility: number of jobs or services reachable within a given travel time
- Environmental: air pollutant emissions, greenhouse gas emissions
- Economic: transportation costs, job access, freight movement efficiency
- Equity metrics assess the distribution of impacts across different population groups
- Example: comparing accessibility changes for low-income and minority communities
- Metrics are computed from model outputs and compared across scenarios
- Helps identify the trade-offs and relative benefits of different policy choices
- Visualization tools (maps, charts, dashboards) communicate scenario results to decision-makers and the public
- Facilitates informed discussion and consensus-building around preferred scenarios
Limitations and future directions
- Despite advances in transportation modeling methods and data sources, significant limitations remain
- Emerging technologies and data-driven approaches offer opportunities to address these limitations and improve model capabilities
- Future research directions focus on integrating new data sources, modeling emerging modes and services, and leveraging artificial intelligence techniques
Limitations of current models
- Aggregate nature of traditional four-step models may not capture fine-grained travel behavior
- Limited representation of individual heterogeneity and complex decision-making processes
- Static assignment models do not account for dynamic traffic flow and congestion propagation
- May overestimate travel speeds and underestimate delay in congested conditions
- Lack of data on non-motorized modes (walking, cycling) and multimodal trips
- Makes it difficult to model the full range of travel options and their interactions
- Limited ability to represent the impacts of disruptive technologies and services
- Example: ride-hailing, microtransit, and automated vehicles may have complex effects on travel behavior and land use
Incorporating emerging technologies
- Agent-based models can simulate the adoption and use of new transportation technologies
- Example: modeling the impacts of automated vehicles on traffic flow and parking demand
- Activity-based models can capture the effects of on-demand services on tour formation and scheduling
- Example: incorporating ride-hailing as a mode choice option and modeling its effects on car ownership
- Integrated land use-transportation models can assess the long-term impacts of technology on urban form
- Example: modeling the potential for automated vehicles to enable sprawling development patterns
- Real-time traffic simulation models can evaluate the operational impacts of connected and automated vehicles
- Example: modeling the effects of platooning and cooperative adaptive cruise control on highway capacity
Data-driven and AI-based approaches
- Machine learning models can leverage large-scale data sets to improve travel demand predictions
- Example: using deep neural networks to predict mode choices based on detailed trip and user attributes
- Reinforcement learning can optimize transportation system operations and control strategies
- Example: learning optimal traffic signal timing plans based on real-time traffic conditions
- Natural language processing can extract insights from unstructured text data
- Example: analyzing social media posts to understand public perceptions of transportation services
- Computer vision can automate the collection and analysis of transportation data
- Example: using image recognition to detect and classify vehicles, pedestrians, and infrastructure conditions
- Data fusion and mining techniques can uncover patterns and relationships in multi-source transportation data
- Example: combining GPS, smart card, and cellular data to infer origin-destination matrices and activity patterns
Key Terms to Review (18)
Active transportation: Active transportation refers to any self-propelled mode of transportation that requires physical activity, such as walking, biking, or using a scooter. This mode of transport promotes not only individual health but also helps reduce traffic congestion and environmental impacts. Emphasizing active transportation in urban planning can enhance community connectivity and encourage healthier lifestyles.
Activity-based modeling: Activity-based modeling is a transportation planning approach that focuses on understanding and predicting individual travel behavior by examining the activities that drive travel demand. This method emphasizes the relationships between daily activities, travel choices, and time use, allowing for more nuanced insights into how different factors influence transportation needs. By modeling activities rather than just trips, it provides a more comprehensive view of travel patterns and can lead to better planning decisions.
Congestion Pricing: Congestion pricing is a transportation management strategy that charges drivers a fee to use specific roadways during peak traffic times to reduce congestion and encourage more efficient use of the transportation system. By making drivers pay for using crowded roads, it aims to shift travel behavior, improve traffic flow, and promote alternative modes of transportation. This pricing model can be applied in various settings, including urban centers and busy highways, playing a crucial role in overall transportation planning and demand management.
Equity in Access: Equity in access refers to the principle that all individuals should have fair and just access to transportation systems and services, regardless of their socioeconomic status, geographic location, or physical ability. This concept emphasizes removing barriers that prevent certain populations from fully utilizing transportation options, thereby promoting inclusivity and equal opportunity within the transportation framework.
Four-step model: The four-step model is a systematic approach used in transportation planning and demand modeling that consists of four key phases: trip generation, trip distribution, mode choice, and route assignment. This model helps planners understand travel behavior, predict future travel patterns, and evaluate the impact of transportation projects on traffic flow and accessibility.
Geographic Information Systems (GIS): Geographic Information Systems (GIS) are powerful tools used to collect, manage, analyze, and visualize spatial data linked to geographical locations. GIS enables the integration of various data sources, allowing for better planning, analysis, and decision-making in fields such as transportation, urban planning, and environmental management. By visualizing complex datasets on maps, GIS helps identify patterns and relationships that may not be immediately obvious in raw data.
Intelligent Transportation Systems (ITS): Intelligent Transportation Systems (ITS) refer to the integration of advanced technologies into transportation systems to improve efficiency, safety, and environmental sustainability. These systems utilize real-time data and communication networks to enhance transportation planning, manage traffic flow, and provide valuable information to travelers, all while addressing the increasing demand for mobility.
Level of Service: Level of Service (LOS) is a qualitative measure used to evaluate the performance and efficiency of transportation systems, typically ranging from A (excellent conditions) to F (failing conditions). It reflects the ability of a transportation facility, such as a road or transit system, to accommodate users' demands while considering factors like travel speed, delay, comfort, and convenience. This measure helps in understanding how well infrastructure supports mobility and safety for all types of users, including drivers, cyclists, and pedestrians.
Modal split: Modal split refers to the percentage distribution of different modes of transportation used by individuals or freight to travel from one place to another. This concept is essential in understanding how various transportation options, such as cars, buses, trains, bicycles, and walking, are utilized in a given area. Analyzing modal split helps to identify travel patterns, inform transportation planning decisions, and assess the effectiveness of various transport modes in meeting the needs of users.
Peter Gordon: Peter Gordon is a renowned transportation scholar and economist known for his contributions to transportation planning and demand modeling. His work focuses on understanding how land use, travel behavior, and transportation systems interact, providing insights that are crucial for effective urban planning and policy-making. His theories have significantly influenced how planners approach the challenges of urban mobility and infrastructure development.
Public Outreach: Public outreach refers to the efforts made to engage and inform the community about transportation planning and projects, aiming to gather feedback and foster collaboration. It plays a crucial role in creating awareness about proposed changes, ensuring transparency, and promoting public involvement in the decision-making processes that affect transportation systems. This engagement helps in understanding the community's needs and preferences, ultimately leading to more effective and accepted transportation solutions.
Public Transit: Public transit refers to a system of transportation that is available for use by the general public, typically consisting of buses, trains, subways, and ferries. It plays a crucial role in providing accessible, affordable, and efficient mobility options to individuals in urban and suburban areas. Public transit systems are designed to reduce traffic congestion, decrease reliance on personal vehicles, and promote sustainable transportation solutions.
Stakeholder engagement: Stakeholder engagement is the process of involving individuals, groups, or organizations that have an interest in or are affected by a project or initiative. This interaction fosters collaboration and communication, allowing stakeholders to provide input and feedback, which ultimately shapes decision-making and project outcomes. Engaging stakeholders is essential for building trust and ensuring that diverse perspectives are considered, particularly in areas like transportation planning, infrastructure development, environmental assessments, strategic planning, and legal compliance.
Sustainability: Sustainability refers to the ability to meet the needs of the present without compromising the ability of future generations to meet their own needs. This concept emphasizes a balanced approach to environmental, economic, and social factors, ensuring long-term health for the planet and its inhabitants. By integrating sustainability into planning and modeling, it encourages a vision for transportation systems that reduces negative impacts on the environment while promoting economic vitality and social equity.
Transportation Demand Management: Transportation Demand Management (TDM) refers to strategies and policies aimed at reducing travel demand, particularly during peak periods, by encouraging shifts in travel behavior. TDM encompasses various techniques, such as promoting carpooling, enhancing public transit options, and implementing flexible work schedules, to optimize the use of existing transportation infrastructure. By addressing the root causes of congestion and travel demand, TDM plays a critical role in planning, modeling, and managing traffic flow effectively.
Vehicle Miles Traveled: Vehicle miles traveled (VMT) is a measure of the total distance driven by all vehicles in a specific area over a certain time period, typically expressed in miles. It serves as an essential metric for understanding travel demand, assessing roadway usage, and evaluating the effectiveness of transportation systems and policies.
William Vickrey: William Vickrey was a Canadian economist who made significant contributions to the fields of auction theory and mechanism design, particularly in relation to optimal taxation and public goods allocation. His work emphasizes how economic policies can be designed to align individual incentives with social welfare, which has important implications for transportation planning and demand modeling.
Zoning Laws: Zoning laws are regulations that govern land use and dictate how property can be developed and utilized in specific areas. These laws play a crucial role in urban planning by helping to manage growth, ensuring that land is used efficiently, and promoting safety and environmental protection. By classifying areas for residential, commercial, or industrial purposes, zoning laws help shape transportation infrastructure and influence demand modeling for various transportation modes.