Conference Agenda

1. Introduction

As cities worldwide commit to ambitious decarbonization targets, high-density urban districts face unique challenges in achieving carbon neutrality due to complex land ownership, aging infrastructure, and intense spatial constraints. Japan’s pledge to reach net‑zero greenhouse‑gas emissions by 2050, reinforced by the Tokyo Metropolitan Government’s “Zero Emission Tokyo” roadmap, places unprecedented decarbonization pressure on the capital’s high‑density inner districts. Nihonbashi as Tokyo’s historical mercantile nucleus and a contemporary hub for finance, retail, and tourism epitomizes this challenge. This area faces compounded challenges from aging buildings, outdated energy infrastructures, and inefficient urban layouts, which exacerbate energy consumption and vulnerability during extreme climate events. Nihonbashi’s urban form is a palimpsest of Edo‑period street grids containing post‑war mid‑rise concrete stock and fragmented land ownership, in which critical energy, water and highway infrastructure date largely back to the 1960s–1980s urban renewal wave. These conditions, coupled with rising heat‑island intensities, pluvial flood hazards along the Sumida watershed, and seismic exposure, create a complex context in which carbon‑neutral retrofits must simultaneously deliver climate‑adaptation and heritage‑conservation benefits.

In Tokyo, where bottom-up stakeholder-driven redevelopment dominates, there is a need to explore design decisions of diverse urban form scenarios while accounting for multiple, often competing, performance goals to facilitate comprehensive, data-driven assessments and proactive urban decision-making for enhancing energy resilience and sustainability. This research introduces a framework of integrating Generative Design (GD) and Multi-Objective Optimization (MOO) that aims for transforming Nihonbashi toward a net-zero smart urban district. This framework will build upon systems architecting principles and an urban digital twins platform previously demonstrated by this research team (Ramnarine et al., 2025).

1.1 Literature Review

1.1.1 Data-driven Smart Urban Districts

Smart cities are becoming a new global movement that uses technologies to drive urban development. Test beds are sprouting up in cities and their strategic areas like Sidewalk Toronto, smart city-nation initiative in Singapore, and Kashiwanoha in Tokyo. There are increasing literatures on smart urban districts to explore impacts of emerging technologies including artificial intelligence (AI), urban automation, Internet of Things (IoT), pervasive computing, and data science to cities, urban infra-structures, public spaces, and our daily life spaces for live, work, and play (Yang and Yamagata., 2020). The concept of urban digital twins is also emerging as a new field of study in urban planning, systems engineering and geospatial information science, with a focus on high-fidelity computational models of cities or ‘replicas’ of urban systems over 4D space and time (Bettencourt, 2024, Batty, 2018).

1.1.2 Generative Design (GD)

Generative design methods offer an algorithmic approach to exploring a wide array of spatial configurations and building typologies. Within urban design, generative systems define and manipulate parameters such as building density (FAR), height, orientation, land-use distribution, and façade characteristics like window-to-wall ratio. Rather than relying on singular masterplans or fixed interventions, generative design methods allow for rapid production of hundreds of spatial variants, enabling comparative assessment of form-based performance outcomes. These techniques are essential for revealing unexpected synergies between urban form and performance goals, especially under high uncertainty.

1.1.3 Multi-Objective Optimization (MOO) Framework for Performance Evaluation

The Urban Building Energy Modeling (UBEM) approach is a key component of this study. UBEM is a method for evaluating the energy and environmental performance of urban districts. This bottom-up, physics-based methodology simulates energy use intensity (EUI), carbon emissions, and operational dynamics across multiple buildings and scenarios. Integrated with the broader framework, UBEM supports performance modeling and impact assessments linked to generative design options.

The framework uses simulation-driven performance evaluation and optimization along multiple objective goals: minimizing district-level energy use, carbon emissions, and peak energy load, while maximizing on-site energy generation (solar PV, kinetic floor systems) and energy storage, open space, thermal comfort (microclimate), daylight access, quality of view, walkability, and network connectivity. These objectives are inherently interdependent and occasionally conflicting that necessitate a robust optimization engine capable of multi-dimensional analysis and visualization.

1.1.4 From Planning Support Systems (PSS), Geodesign to Urban Digital Twins

The proposed methodology builds on decades of planning support tools and geodesign practices that emphasize scenario-based decision making. However, it advances conversation by embedding these capabilities into a live and interactive digital twins. Urban digital twins are dynamic representations of the built environment that synchronize real-time data with predictive modeling. These platforms extend beyond visualization to support creation, negotiation, and iterative refinement with stakeholders transforming design from a static outcome into a collaborative and informed process.

1.2 Problem Formulation

Tokyo’s commitment to becoming a carbon-neutral metropolis by 2050 includes sustainable urban regeneration policies and an emphasis on bottom-up participation. Yet, the fine-grained urban form of districts like Nihonbashi poses major challenges to large-scale retrofitting or conventional urban redevelopment. The current revitalization efforts often follow fixed retrofit pathways such as reconstruction, façade renovation, or operational upgrades, but do not explore the broader design space nor adequately support complex multi-criteria optimization.

Moreover, the current energy simulation methods such as UBEM provides a strong technical foundation for energy performance of the built environment, it remains disconnected from generative design and lacks tools to meaningfully engage stakeholders in evaluating and selecting among alternatives.

To address these gaps, we propose a data-driven smart district systems design framework that fuses GD with MOO on a high‑resolution Digital Twins of Nihonbashi. Building upon the systems‑architecting principles, the framework will algorithmically generate data-driven redevelopment scenarios, evaluate their performance across carbon‑neutrality and resilience metrics, and visualize solutions for stakeholder deliberation. By embedding rigorous analytics within an exploratory design environment, the framework seeks to operationalize Tokyo’s carbon‑neutral agenda at the district scale while safeguarding Nihonbashi’s cultural identity and economic vitality.

1.3 Research Questions

1. How can generative design methods systematically broaden the urban design space by altering form variables such as building density, orientation, height, mixed uses, and façade characteristics?
2. What kinds of trade-offs and synergies arise when optimizing across energy, carbon, environmental quality, and urban network indicators?
3. How can interactive urban digital twins be leveraged to guide stakeholders in selecting optimal scenarios and making informed decisions toward carbon neutrality?

2. A Methodology on Urban Digital Twins for a Net Zero Smart District

Stage One – Descriptive and Evaluative Model: Analysis of Existing Conditions

We begin with a detailed descriptive analysis of the existing urban fabric in Nihonbashi, focusing on 154 buildings in District 2. Carbon emissions and energy profiles are applied for performance evaluation using UBEM-based simulations, revealing hotspots of inefficiency and opportunity areas for intervention. The analysis is supported by data provided by the University of Tokyo.

Stage Two – Predictive Model: Generative Design and Multi-Objective Optimization

Generative design scenarios are developed using Rhino/Grasshopper (with DOTS) and CityEngine, allowing procedural and parametric variation of building attributes across the district. The parameters include building density (FAR), building height, orientation, building-use distribution, and window-to-wall ratio (WWR). These scenarios are then evaluated through URBANopt simulations, modeling site-level energy consumption, carbon emissions, and comfort metrics using bottom-up energy archetypes for Tokyo’s climate. A Python-based optimization engine conducts multi-objective optimization based on the following criteria: minimized energy use, carbon emissions and peak demand; maximized renewable energy generation (solar/kinetic), energy storage, open space, thermal comfort, daylight/view, walkability, and network connectivity.

Stage Three – Prescriptive and Interactive Model: Integration into Digital Twin Platform

Finally, the optimized scenarios are integrated into an interactive dashboard inspired by the CANVAS urban digital twins (Ramnarine et al., 2025). This interface allows users, planners, architects, and community members to manipulate variables (e.g., WWR, FAR, PV efficiency) and immediately view the impacts on performance. The dashboard enables stakeholders to visualize future impacts and align design preferences with environmental targets.

3. Expected Outcomes

The application of this integrated workflow in the Nihonbashi district is expected to yield design solutions with substantial EUI and carbon reductions. For example, the previous study estimated that combining reconstruction and operational improvements could yield 99 kWh/m²/year in EUI reductions, while solar PV systems alone could contribute 42.5 kWh/m²/year in on-site energy generation. By embedding these capabilities within a generative and multi-objective framework, the new methodology allows for deeper exploration and more balanced solutions.

References

Bettencourt, L. M, N. 2024, Recent Achievements and Conceptual Challenges for Urban Digital Twins, Nature Computational Science, Vol 4. P. 150-153.

Batty, M, 2018, Inventing Future Cities, MIT Press.

Ramnarine, Ishwar D, Sherif, Tarek A, Alorabi, Abdulrahman H, Helmy, Haya, Yoshida, Takahiro, Murayama, Akito, Yang, Perry P. J., 2025, Urban Revitalization Pathways Toward Zero Carbon Emissions through Systems Architecting of Urban Digital Twins, Environment and Planning B: Urban Analytics and City Science.

Yang, Perry P. J., Yamagata, Yoshiki, 2020, Urban Systems Design: Shaping Smart Cities by Integrating Urban Design and Systems Science, in Urban Systems Design: Creating Sustainable Smart Cities in the Internet of Things Era, Yamagata and Yang eds., Elsevier.

Understanding how young people envision happiness and ideal cities is essential for designing human-centered smart urban futures. This study explores the well-being perceptions of Japanese youth aged 16–24 (N = 2,437), using an online survey that included both closed and open-ended questions.

Topic modeling and cluster analysis revealed three dominant value orientations: Connection and Belonging, Freedom and Autonomy, and Stability and Security. These themes reflect diverse urban preferences grounded in emotional, social, and existential needs. While no urban feature showed strong statistical significance in predicting well-being, nature, safety, and cultural elements appeared consistently relevant.

Narrative responses highlighted symbolic openness and emotional nuance, suggesting the importance of environments that foster flexibility and non-prescriptive engagement—what we refer to as “slack fields.” Though not directly measured, such affective dimensions may complement spatial metrics in digital twin and urban analytics systems.

This study contributes to value-based urban design by integrating youth perspectives and affective indicators into the smart city discourse. Its findings support more inclusive, emotionally resonant planning frameworks that go beyond technical efficiency.

1. Introduction

Location-based storytelling, a key strategy in smart tourism, helps visitors and residents deepen their understanding of local culture and history through urban exploration. By fostering a sense of attachment and pride, such experiences contribute to more comfortable and meaningful mobility within the city. A representative example is the proactive local guide, which delivers guides automatically when users arrive at key points of interest (POIs). These services rely on geofencing technology that triggers content delivery as users enter predefined spatial zones.

Manually designing effective and scalable geofences in urban settings is challenging. It requires consideration of varying GPS accuracy across locations and the complexity of pedestrian movement patterns, making manual implementation labor-intensive and often suboptimal (Garzon et al, 2017). Moreover, the appropriate timing and spatial conditions for triggering content depend on the nature of the storytelling itself (see Section 2). Thus, geofence design must address not only technical and behavioral constraints, but also the need to differentiate spatial triggers based on narrative intent. This study proposes a data-driven approach for identifying geofence zones by leveraging mobile sensor data collected from tourists.

2. Geofence Typology Based on Storytelling Needs

To align geofence design with different storytelling needs, we classify geofences into two types: enclosure-geofences and viewpoint-geofences (Figure 1).Enclosure-geofences associates situations where users physically enter museums. These geofences support immersive content tied to the space users are exploring. In contrast, viewpoint-geofences are placed at outdoor viewpoints from which users observe POIs, such as a building façade, from a distance. This distinction reflects different user behaviours—exploring from within versus observing from outside—and leads to two geofencing models: the ROI-based model for indoor cases and the viewpoint-based model for viewing cases. Clarifying this typology allows appropriate content to be delivered at the right place and moment, enhancing the relevance of location-based storytelling.

3. Classification Framework for Geofence Identification

This study investigates methods for extracting clusters corresponding to enclosure- and viewpoint-geofences from GPS trajectory data. Spatial clustering for detecting meaningful clusters from trajectory data have been proposed. We examined two widely adopted methods: HDBSCAN (Campello et al., 2015) and Stay Point Detection (Li et al., 2008). Figure 2 presents GPS trajectories recorded during approximately two hours of walking tours. HDBSCAN suffers from difficult parameter tuning (Yang et al., 2014), often resulting in overly large clusters including enclosure- and viewpoint-geofences, Stay Point Detection requires users to remain at fixed locations, making it unsuitable for detecting single clusters in large facilities.

This study proposes a classification framework utilizing mobile sensor data to identify geographic zones enabling to associate with each storytelling needs (Figure 3). First, GPS accuracy clustering extracts clusters from segments of the trajectory where the GPS horizontal accuracy falls below a certain threshold, indicating enclosure-geofence candidates. However, some of the extracted clusters did not accurately correspond to indoor stays. For example, low GPS accuracy may also occur outdoors, such as under tree cover or immediately after the GPS sensor is activated. In addition, areas like underground walkways, although technically indoor, often serve as mere transit routes rather than destinations of interest. To address these issues, we implemented a binary classification step for noise detection. This classifier aims to distinguish between meaningful indoor zones and spurious clusters caused by environmental noise or transient movement. Specifically, we examined combinations of features derived from smartphone sensor data, including the mean and variance of acceleration and horizontal accuracy values, as well as the number of GPS points contained within each cluster ().

4. Evaluation of Enclosure-Geofence Identification

To evaluate the proposed approach, we conducted an in-situ data collection study involving 12 participants who each carried an iPhone 11 while engaging in walking-based sightseeing activities for approximately 90 minutes on average. The resulting mobile sensor dataset was used to apply the GPS accuracy clusteringmethod. This process identified 35 clusters in total. Of these, 25 clusters were confirmed to correspond to actual user stays, while the remaining 10 were associated with either underground walkways or erroneous detections in outdoor environments. To effectively classify these clusters, we developed a decision tree classifier and examined various combinations of features through scatter plot analysis. The combination of mean acceleration and acceleration variance yielded the highest class separability, achieving an accuracy of 0.83 in leave-one-out cross-validation (LOOCV). These results suggest that users tend to slow down or pause intermittently around points of interest, leading to distinctive acceleration patterns that help distinguish indoor stays from transient movement.

5. Conclusions

This study introduces a systematic and computational framework for generating story-aware geofences from mobile sensor data. By classifying clusters based on narrative intent, and using data-driven techniques to detect and validate these regions, we provide a foundation for scalable, context-sensitive storytelling in urban environments. Future work includes the automatic delineation of geofence boundaries, improved alignment with POI datasets, and the development of a fully autonomous geofencing service deployable across diverse cities through crowdsourced mobile data collection.