skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


This content will become publicly available on February 1, 2026

Title: Interpretable State-Space Model of Urban Dynamics for Human-Machine Collaborative Transportation Planning
Strategic Long-Range Transportation Planning (SLRTP) is pivotal in shaping prosperous, sustainable, and resilient urban futures. Existing SLRTP decision support tools predominantly serve forecasting and evaluative functions, leaving a gap in directly recommending optimal planning decisions. To bridge this gap, we propose an Interpretable State-Space Model (ISSM) that considers the dynamic interactions between transportation infrastructure and the broader urban system. The ISSM directly facilitates the development of optimal controllers and reinforcement learning (RL) agents for optimizing infrastructure investments and urban policies while still allowing human-user comprehension. We carefully examine the mathematical properties of our ISSM; specifically, we present the conditions under which our proposed ISSM is Markovian, and a unique and stable solution exists. Then, we apply an ISSM instance to a case study of the San Diego region of California, where a partially observable ISSM represents the urban environment. We also propose and train a Deep RL agent using the ISSM instance representing San Diego. The results show that the proposed ISSM approach, along with the well-trained RL agent, captures the impacts of coordinating the timing of infrastructure investments, environmental impact fees for new land development, and congestion pricing fees. The results also show that the proposed approach facilitates the development of prescriptive capabilities in SLRTP to foster economic growth and limit induced vehicle travel. We view the proposed ISSM approach as a substantial contribution that supports the use of artificial intelligence in urban planning, a domain where planning agencies need rigorous, transparent, and explainable models to justify their actions.  more » « less
Award ID(s):
2125560
PAR ID:
10618796
Author(s) / Creator(s):
;
Publisher / Repository:
Elsevier
Date Published:
Journal Name:
Transportation Research Part B: Methodological
Volume:
192
Issue:
C
ISSN:
0191-2615
Page Range / eLocation ID:
103134
Subject(s) / Keyword(s):
Smart Planning Human-AI Teaming Markov Decision Process Sketch Planning Machine Learning Multicriteria Land Use Travel Demand Regional Growth Reinforcement Learning
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, PLOS ONE)
    Lenormand, Maxime (Ed.)
    Neighborhoods are the building blocks of cities, and thus significantly impact urban planning from infrastructure deployment to service provisioning. However, existing definitions of neighborhoods are often ill suited for planning in both scale and pattern of aggregation. Here, we propose a generalized, scalable approach using topological data analysis to identify barrier-enclosed neighborhoods on multiple scales with implications for understanding social mixing within cities and the design of urban infrastructure. Our method requires no prior domain knowledge and uses only readily available building parcel information. Results from three American cities (Houston, New York, San Francisco) indicate that our method identifies neighborhoods consistent with historical approaches. Additionally, we uncover a consistent scale in all three cities at which physical isolation drives neighborhood emergence. However, our methods also reveal differences between these cities: Houston, although more disconnected on larger spatial scales than New York and San Francisco, is less disconnected at smaller scales. 
    more » « less
  2. (, ASCE International Conference on Computing in Civil Engineering 2019)
    This paper proposed and tested a multilayer framework for modeling network dynamics of inter-organizational coordination in resilience planning among interdependent infrastructure sectors. Each layer in the network represents one infrastructure sector such as flood control, transportation, and emergency response. Coordination probability was introduced to approximate the inconsistent coordination between organizations, based on which the intra-layer or inter-layer link removal was conducted and inter-organizational coordination efficiency within and across infrastructure sectors was hereby unveiled. To test the proposed framework, a multilayer collaboration network of 35 organizations from five infrastructure sectors in Harris County, Texas, was mapped based on a survey of Hurricane Harvey. The analysis results showed that before Hurricane Harvey, coordination among flood control, transportation, and infrastructure development sectors lacked essential integration to foster robust resilience plans. The proposed framework enables an assessment of coordination efficiency among organizations involving in resilience planning and provides an indicator for urban resilience measurement. 
    more » « less
  3. ; ; ; ; ; ; ; ; (, ACM Journal on Computing and Sustainable Societies)
    Urban population growth has significantly complicated the management of mobility systems, demanding innovative tools for planning. Generative Crowd-Flow  (GCF) models, which leverage machine learning to simulate urban movement patterns, offer a promising solution but lack sufficient evaluation of their fairness–a critical factor for equitable urban planning. We present an approach to measure and benchmark the fairness of GCF  models by developing a first-of-its-kind set of fairness metrics specifically tailored for this purpose. Using observed flow data, we employ a stochastic biased sampling approach to generate multiple permutations of Origin-Destination  datasets, each demonstrating intentional bias. Our proposed framework allows for the comparison of multiple GCF  models to evaluate how models introduce bias in outputs. Preliminary results indicate a tradeoff between model accuracy and fairness, underscoring the need for careful consideration in the deployment of these technologies. To this end, this study bridges the gap between human mobility literature and fairness in machine learning, with potential to help urban planners and policymakers leverage GCF  models for more equitable urban infrastructure development. 
    more » « less
  4. ; ; ; ; ; (, Proceedings of the 2021 ACM SIGCOMM 2021 Conference)
    Network planning is critical to the performance, reliability and cost of web services. This problem is typically formulated as an Integer Linear Programming (ILP) problem. Today's practice relies on hand-tuned heuristics from human experts to address the scalability challenge of ILP solvers. In this paper, we propose NeuroPlan, a deep reinforcement learning (RL) approach to solve the network planning problem. This problem involves multi-step decision making and cost minimization, which can be naturally cast as a deep RL problem. We develop two important domain-specific techniques. First, we use a graph neural network (GNN) and a novel domain-specific node-link transformation for state encoding, in order to handle the dynamic nature of the evolving network topology during planning decision making. Second, we leverage a two-stage hybrid approach that first uses deep RL to prune the search space and then uses an ILP solver to find the optimal solution. This approach resembles today's practice, but avoids human experts with an RL agent in the first stage. Evaluation on real topologies and setups from large production networks demonstrates that NeuroPlan scales to large topologies beyond the capability of ILP solvers, and reduces the cost by up to 17% compared to hand-tuned heuristics. 
    more » « less
  5. ; (, Advances in neural information processing systems)
    Koyejo, S.; Mohamed, S.; Agarwal, A.; Belgrave, D.; Cho, K.; Oh, A. (Ed.)
    While much progress has been made in understanding the minimax sample complexity of reinforcement learning (RL)—the complexity of learning on the “worst-case” instance—such measures of complexity often do not capture the true difficulty of learning. In practice, on an “easy” instance, we might hope to achieve a complexity far better than that achievable on the worst-case instance. In this work we seek to understand the “instance-dependent” complexity of learning near-optimal policies (PAC RL) in the setting of RL with linear function approximation. We propose an algorithm, Pedel, which achieves a fine-grained instance-dependent measure of complexity, the first of its kind in the RL with function approximation setting, thereby capturing the difficulty of learning on each particular problem instance. Through an explicit example, we show that Pedel yields provable gains over low-regret, minimax-optimal algorithms and that such algorithms are unable to hit the instance-optimal rate. Our approach relies on a novel online experiment design-based procedure which focuses the exploration budget on the “directions” most relevant to learning a near-optimal policy, and may be of independent interest. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email