Grace Orowo Kagho


Last Name

Kagho

First Name

Grace Orowo

ORCID

0000-0002-2671-1375

Organisational unit

03859 - Adey, Bryan T. / Adey, Bryan T.

Filters

2019 - 201912020 - 202529
Reset filters

Search Results

Publications 1 - 10 of 30
  • Framework for analyzing equity-concerns related to mobility on-demand
    Item type: Conference Paper
    Kagho, Grace Orowo; Murthy Gurumurthy, Krishna; Verbas, Ömer; et al. (2024)
    2024 TRB Annual Meeting Online Program Archive
    This study looks at the concept of equity in the context of urban transportation, particularly focusing on Mobility on-Demand (MoD) simulations. MoD services, such as ride-hailing and shared autonomous vehicles, have the potential to transform transportation systems, offering new opportunities to enhance equity while reducing transport externalities. However, a comprehensive assessment of the equity impacts resulting from the integration of these services into various transport scenarios is lacking in most MoD simulations. Even though this is essential in informing policy and decision-making, as MoD impacts on equity are complex and depend on various factors, including service availability, affordability, accessibility, and their interaction with existing transportation modes. Therefore, this study presents a framework for evaluating and quantifying equity impact in MoD simulations. Several metrics are defined, such as demographic equity, spatial equity distribution, spatial regression analysis, hotspots and coldspots cluster identification and social inclusion analysis. The effectiveness of this framework is presented using a first-mile-last-mile subsidy case study for two US regions, Austin and Chicago. The findings from these case studies underscored the importance of equity-focused assessments in simulating MoD policy scenarios. And how it is essential to take into account the varying needs of different sociodemographic groups and ensure that the services are distributed in a manner that improves accessibility and affordability for all, particularly the most vulnerable groups.
  • Effects of population sampling on agent-based transport simulation of on-demand services
    Item type: Conference Paper
    Kagho, Grace Orowo; Meli, Jonas; Walser, Dominique; et al. (2022)
    Procedia Computer Science
    Large-scale agent-based simulations require higher computing resources than are usually available. Consequently, many applications rely on downscaling, that is, simulating with smaller population samples in which the results are then scaled. Existing studies have shown a need to investigate the impact of downscaling on the output statistics of such simulations. Downscaling is a common practice in transport modeling. In this study, we investigate the impacts of population downscaling on a ride-sharing service with a focus on vehicle occupancy and wait time, travel time and detour time. Our findings reveal that if transport modelers want to model on-demand services with ride sharing, it is strongly recommended to use a 100% population, or when using a smaller population sample, to estimate the relative biases of their desired metrics compared to the results of a 100% population in order for their results to be applicable for real-world situations.
  • Potential impacts of integrating an on-demand transport service with public transit system
    Item type: Other Conference Item
    Kagho, Grace Orowo; Balać, Miloš; Axhausen, Kay W. (2022)
    2022 TRB Annual Meeting Online Program Archive
  • An agenda for data-rich, action-oriented, ethical research on infrastructure in informal settlements
    Item type: Journal Article
    Borofsky, Yael; Kersey, Jessica; Caprotti, Federico; et al. (2025)
    Nature Cities
    Infrastructure inequities define modern cities. This Perspective reflects the viewpoint of a transdisciplinary group of co-authors working to advance infrastructural equity in low-income urban contexts. We argue that methodological silos and data fragmentation undermine the creation of a knowledge base to support coordinated action across diverse actors. As technological advances make it possible to ‘see’ informal settlements without engaging residents, our agenda advocates for (1) the integration of diverse methodological and epistemological traditions; (2) a focus on research that informs context-specific action; and (3) a commitment to ethical standards that center affected communities in efforts to improve infrastructure access.
  • A cognitive architecture for intelligent urban planning
    Item type: Conference Paper
    Cárdenas, Miguel; Kagho, Grace Orowo (2025)
    Urban planning requires complex reasoning across interconnected systems such as transportation, demographics, land use, and environmental factors. Artificial intelligence can integrate diverse urban data and models, but current tools often lack domain grounding, memory, or simulation integration needed for proper planning support. We present Echo, a cognitive architecture embedded in digital twin model of a transport system, that combines an urban knowledge graph, multi-memory reasoning (semantic, episodic, working), and a multi-agent controller to answer natural-language planning queries grounded in explicit sources. Echo lets agents traverse an ontology, retrieve spatio-temporal relationships, and incorporate simulation feedback into explanations. In a transport planning case study, Echo produced richer, more reliable answers than standard approaches (RAG, ReAct), achieving an 88% first-attempt accuracy when validated by domain experts, at the cost of higher per-query computation. These results advance practical, source-grounded decision support for urban planners.
  • Predicting behavioral patterns in data-scarce regions using spatial embeddings
    Item type: Conference Paper
    Kagho, Grace Orowo; Pougala, Janody (2025)
    Activity-based models (ActBMs) are increasingly vital in urban planning, transportation, epidemiology, and disaster response due to their capacity to simulate individual-level daily activities and capture complex interactions among transport, land use, social networks, and demographic elements. This study addresses the data scarcity challenge of ActBM development in many regions by exploring the integration of embedding-based geospatial models and transfer learning within ActBM. We propose a probabilistic modeling framework that integrates spatial embeddings with transfer learning to identify and predict “sister regions”: geographical areas exhibiting similar behaviors. A formal protocol for identifying these sister regions based on embedding similarity is presented. To facilitate this, we use NGBoost with the Population Dynamics Foundational Model (PDFM) to quantify spatial analogies between regions. These PDFM embeddings are then combined with socioeconomic features from the American Time Use Survey and the US census to predict state-level distributions of time-use characteristics and activity patterns. Preliminary results indicate that combining PDFM embeddings with socioeconomic characteristics reduces prediction uncertainty compared to models relying solely on traditional socioeconomic features. We also determine the effect of missing socioeconomic data on achieving stable predictions, demonstrating a scalable solution for data-limited contexts.
  • An agent-based MATSim scenario for Lagos, Nigeria
    Item type: Conference Paper
    Kagho, Grace Orowo; Axhausen, Kay W. (2019)
    Presently there is a shift in assessing infrastructure investment decisions in developing regions. Whilst in the past, transport and mobility-related metrics that have worked for developed regions have been used in developing countries, this practice is now changing. Some transport researchers have started to consider inclusive metrics outside of the normal metrics that do not account for issues such as economic inequality among citizens and citizen individual mobility behaviors in developing urban regions which may be very different from that of developed regions. In this same spirit, an agent-based microsimulation framework, MATSim has been applied hereto build a baseline scenario for Lagos, Nigeria. At the initial stageis a model using only a population derived froma household travel survey, which serves as a starting point to a full-scale scenario for the whole of Lagos pending access to more data
  • Understanding of group travel and weekend travel behavior for on-demand mobility services
    Item type: Other Conference Item
    Kagho, Grace Orowo; Balać, Miloš; Axhausen, Kay W. (2023)
  • Sensitivity analyses for ride-hailing and ridesharing modelling strategies
    Item type: Other Conference Item
    Kagho, Grace Orowo; Balać, Miloš; Axhausen, Kay W. (2022)
    Agent-based models are widely used to simulate on-demand mobility services. This enables the capturing of the supply and demand dynamics and the complex interactions between individual travelers and the operational decisions of the on-demand system. Such models usually have nonlinear interactions between input parameters and the model components, which affects the model outputs. Controlling for the effects reduces uncertainty in the model. This can be done through sensitivity analyses, which help to determine to what extent the input parameters contribute to the outcome of a model. Shared automated vehicles (SAVs), a form of on-demand mobility service, are seen as a potential future modal mix in the advent of autonomous vehicles. This is because when coupled with ridepooling, SAVs can reduce the environmental impacts that AVs are purported to bring, as less vehicles would be required to serve the population. With this outlook, the research interest in modelling SAVs has grown, as researchers, planners try to understand what operational scenarios, and transport policies are needed for their successful implementation in reality. With the growing interest in these agent-based SAV models, sensitivity analysis then becomes crucial. SAVs specific model parameters could include, trip density, fleet size, vehicle capacity, operating hours, spatial distribution, fares and service constraints such as maximum acceptable wait time for a pickup or detour time (the extra time added to one’s trip while another passenger is picked up). Model outcomes could be the SAV operational (e.g. average vehicle occupancy, or vehicle kilometers traveled (VKT)), demand, level of service or externalities measures. Therefore, to what extent do each of these model parameters affect the model outcomes or other model parameters? For example, what is the relationship between trip density and the fleet size or the vehicle occupancy of the system? How would modeling a small sample of the demand affect the results? The aim of this study is to provide an understanding of how the different model parameters interact and how they affect the model outcomes. Particularly, this paper would discuss the impact of demand sampling, fleet sizing and spatial distribution of vehicles.
  • Demand responsive transit simulation of Wayne County, Michigan
    Item type: Journal Article
    Kagho, Grace Orowo; Hensle, David; Balać, Miloš; et al. (2021)
    Transportation Research Record
    Demand responsive transit (DRT) can provide an alternative to private cars and complement existing public transport services. However, the successful implementation of DRT services remains a challenge as both researchers and policy makers can struggle to determine what sorts of places or cities are suitable for it. Research into car-dependent cities with poor transit accessibility is sparse. This study addresses this problem, investigating the potential of DRT service in Wayne County, U.S.A., whose dominant travel mode is private car. Using an agent-based approach, DRT is simulated as a new mobility option for this region, thereby providing insights into its impact on operational, user, and system-level performance indicators. DRT scenarios are tested for different fleet sizes, vehicle occupancy, and cost policies. The results show that a DRT service in Wayne County has a certain potential, especially to increase the mobility of lower-income individuals. However, introducing the service may slightly increase the overall vehicle kilometers traveled. Specific changes in service characteristics, like service area, pricing structure, or preemptive relocation of vehicles, might be needed to fully realize the potential of pooling riders in the proposed DRT service. The authors hope that this study serves as a starting point for understanding the impacts and potential benefits of DRT in Wayne County and similar low-density and car-dependent urban areas, as well as the service parameters needed for its successful implementation.
Publications 1 - 10 of 30