Modeling Mobility
Parts of the following lecture is summarized from Activity Based Travel Demand Models and Urbano: A Tool to Promote Active Mobility Modeling and Amenity Analysis in Urban Design
A former version of this lecture — Activity & Trip Based Travel Models — was developed by Professor Violet Whitney.
Why Model Mobility?
Modern planning paradigms promote the design of walkable neighborhoods. To allow urban designers to understand the consequences of design choices regarding the street network and the allocation of density, program, and amenities, it is imperative to develop new modeling capabilities to facilitate the design of healthy and sustainable urban habitats that promote active mobility.
Mobility modeling allows the designer to evaluate the accessibility performance of an existing condition or design proposal, and actively address challenges within the context to create better design. It also help answers daring questions that will shape the urban landscape or tomorrow.
“How will the current built environment and transportation infrastructure perform in the next 30 years?”
“How can future cities be built on a platform that prioritizes equitable accessibility?”
Mobility Modeling
The type of model you choose to use will also depend on the type of question you aim to answer.
Where do you need detail? Different model types allow for varying levels of fidelity in various areas. However there are tradeoffs for higher fidelity models because they take more computing time and can be more costly to develop models that reflect all of the information of interest. At lower fidelities, sketch planning (using GIS and spreadsheets) may be sufficient in early stages of the design before creating higher fidelity models.
Trip vs Tour vs Activity
- Trip (aka four step model) — individual person trips, doesn’t understand that trips are interrelated
- Tour — strings together trips that will be typically done in sequence, but doesn’t account for vehicle capacities or total course of the day
- Activity — accounts for vehicle trips (accounts for vehicle capacity) and how trips would be strung together over the course of a day
Trip-Based Model
A trip-based model is structured in four steps: trip generation, trip distribution, mode choice, and route choice.
1 Trip Generation
ITE: Trip Generation Report — ex: will show number of trips to a proposed shopping mall based on amount of sq ft
2 — Distribution
desire lines between two zones, then attractiveness applied between two zones (if more jobs in one zone = more attractive)
3 — Mode
mode — likelihood of someone to take one transit type versus another based on the amount of time it will take
4 — Route
route — which route is taken
Activity-Based Model
- Distinguished from trip based — represent each person’s activity and travel choices across the entire day, considering the types of activities the individual needs to participate in and setting the priorities for scheduling these activities. As a persons schedule fills up, they no longer have time left to travel. This is particularly important for account for the differences in travel flexibility experienced by various travelers.
- are based on behavioral theories about how people make decisions to participate in activities based on when and where those activities are and how to get to them. Activity based models tend to represent behaviors and decisions at higher fidelity and thus can be more realistic.
Amenity Demand Profiles
The active mobility simulation is based on the concept of activity trips, defined by the shortest path between a trip origin in a building and the destination’s amenity. The activity trips are generated via a data-driven metric of amenity demand profiles (ADPs). ADPs describe the spatiotemporal distribution of human activities according to activeness in urban amenities.
The ADPs drive Urbano’s trip-sending algorithm. In theory, if an ADP can describe human behavior in cities with greater detail, the urban mobility pattern and amenity utilization predictions will be more accurate. However, defining ADPs at a high level of detail is notoriously difficult, and the required data is often not available to support a design process. Hence, the tool uses a simple ADP that lists desired activities over time. The user can create this data to test scenarios for an assumed demographic, or the data can be derived from urban data sources, if available.
Evaluating Mobility
Walk Score
To evaluate the walkability of cities, efforts have been made to rank them based on a shortest-distance analysis between different points of interest (POIs). These walkability ratings, commonly referred to as Walkscores (Brewster et al. 2009; ESRI 2019; Walkscore 2019), are computed on a scale of one to one hundred and include factors such as accessibility to services and amenities such as grocery stores, doctors, parks, schools, hospitals, and public transportation. The Urban Modeling Interface (Reinhart et al. 2013) can compute the Walkscore metric. The main challenge with this tool is to provide the required inputs (e.g., street networks, buildings, and the locations of amenities) that the user must enter manually.
Street Score
This tool measures street utilization via a simple counter called Streetscore, which evaluates how many people use a certain street segment. This may be used as an indicator of how vibrant a street is within the network in general or at a given time of interest.
Amenity Score
This tool is introduced as a counterbalance to the Walkscore metrics. In an urban design process, it is easy to increase the Walkscore by adding more amenities to the neighborhood. However, adding amenities can be costly and can present an economic risk, especially when it is unclear whether sufficient user demand exists to sustain them. Amenityscore (A) aims to measure the difference between the supply and demand for a particular amenity type in the area as specified in equation (5). It employs a simple counter (H) that tallies up the total number of people sent to a specific amenity on all trips. The amenity capacity © represents the maximum occupancy. A desirable Amenityscore is close to zero, which indicates a balance of supply and demand. A higher Amenityscore is a sign for a demand that exceeds the supply, while a negative score is an indicator for underutilization of a specific amenity.
