International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 10 | October 2024
p-ISSN: 2395-0072
www.irjet.net
AI and Machine Learning in Urban Design: Analyzing Predictive Modeling for Pedestrian Flow Optimization Kunal Chhatlani, Sakshi Nanda 1Kunal Chhatlani, Independent Affiliation, New Jersey, USA 2Sakshi Nanda, Independent Affiliation, Oregon, USA
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Urban design is starting to apply Artificial
of pedestrian movements, we aim to enhance the precision and relevance of predicting pedestrian flow
Intelligence (AI) and Machine Learning (ML) in an attempt to optimize complex challenges with regard to pedestrian flow. This paper provides a comprehensive analysis of the existing literature on predictive modeling of pedestrian movement, with particular focus on integrating data-driven approaches and agent-based modeling. This paper highlights the prospect of improving urban planning by using the potential of data-driven pedestrian dynamics models—supported by multi-source data for building evacuation—coupled with research on agent-based modeling of pedestrian movements. The given analysis underlines the importance of utilizing multi-source data and advanced modeling techniques in a proper way to give accurate predictions of pedestrian behavior for making the urban environment more livable and efficient.
2. LITERATURE REVIEW 2.1 Traditional Pedestrian Flow Models Pedestrian flow modeling has progressed from basic mathematical depictions to more intricate simulations. Initial models, such as the social force model proposed in the mid-1990s, provided a macroscopic view of pedestrian dynamics but lacked the granularity to capture individual behaviors and interactions (Helbing & Molnár, 1995). These models frequently viewed pedestrians as particles impacted by social forces, a perspective beneficial for analyzing overall crowd movement but lacking consideration for individual decision-making processes.
Key Words: Urban Design, Artificial Intelligence (AI), Pedestrian Flow Optimization, Data Integration, Agent-Based Modeling, Data-Driven Modeling, Urban Mobility, Machine Learning (ML).
2.2 Agent-Based Movements
The rapid growth of urban populations has increased the necessity for efficient management of pedestrian flow in cities around the globe. Crowded pathways, ineffective public areas, and insufficient evacuation protocols have a negative impact on safety, accessibility, and the general quality of urban life. Conventional approaches to modeling pedestrian movement frequently fail to accurately represent the intricate and ever-changing behavior of people in urban environments. Progress in AI and ML field offers encouraging opportunities in improving predictive models of pedestrian flow. Through the utilization of large quantities of data and advanced algorithms, these tools offer more in-depth understanding of pedestrian behaviors, allowing city planners to create environments that are both more effective and adaptable. This paper offers a detailed analysis of existing research on predictive modeling for pedestrian flow optimization, with a specific emphasis on combining data-driven methods with agent-based models. By combining insights from studies on data-driven pedestrian dynamics using multi-source data and research on agent-based modeling
|
Impact Factor value: 8.315
of
Pedestrian
Agent-based models (ABMs) simulate how individual pedestrians, acting autonomously, interact and impact the overall system. Research in this area has highlighted the potential of ABMs in simulating pedestrian movements by modeling pedestrians with specific characteristics and behaviors (Kerridge et al., 2001). Key aspects of this methodology include: Individual Agent Representation: Each pedestrian is modeled as an individual agent with parameters such as walking speed, desired destinations, and personal preferences. Behavioral Rules: Agents follow predefined rules that dictate their movement and interactions, including speed preferences, route choices, and avoidance of obstacles and other pedestrians. Environment Interaction: Agents interact with the environment, encompassing physical infrastructure, spatial layouts, and other agents. Critical questions addressed in this modeling approach include: How do the actions of individuals impact group behavior on a larger scale? What are the appropriate behavioral rules for agents?
1.INTRODUCTION
© 2024, IRJET
Modeling
|
ISO 9001:2008 Certified Journal
|
Page 837