Strategic Management of Urban Sustainability and Resilience: Navigating the BANI Environment in Ukrainian Context

Abstract

This article proposes a strategic framework for Kyiv’s post-conflict sustainability and resilience under brittle, anxious, non-linear, and incomprehensible (BANI) conditions. We integrate adaptive governance, circular-economy reconstruction, and city-scale digital capabilities, including AI-enabled analytics, IoT sensing, and urban digital twins. Building on recent assessments of Ukraine’s reconstruction needs, we outline a socio-technical model that links sustainability and resilience objectives under shock risk and budget constraints and show how an illustrative five-year optimisation can rebalance investments toward distributed renewables and early-warning infrastructure. The example portfolio achieves an end-horizon composite performance of F_optimized(5) = 0.65 (on a 0–1 normalised index where 1 indicates achieving the policy-defined targets; 0.65 indicates ~65% progress toward those targets at year 5, improving on the baseline allocation under shocks), indicating improved robustness relative to a baseline allocation. We emphasise that effective implementation depends on secure-by-design digital architecture, participatory prioritisation of indicators and weights, and iterative monitoring that supports rapid adaptation as conditions evolve. The framework provides a pragmatic roadmap for Kyiv and similarly vulnerable cities seeking a low-carbon recovery while reducing systemic brittleness and mitigating anxiety-driven decision-making delays.

1. Introduction

Many cities now confront disruption patterns that are abrupt, emotionally taxing, and hard to interpret, which the BANI lens captures more sharply than earlier uncertainty frameworks [1]. For Kyiv, BANI is not an abstract diagnosis but a practical description of repeated shocks to energy, housing, mobility, and municipal service continuity. This context demands recovery strategies that remain functional under cascading failures and shifting constraints, while keeping low-carbon and equity goals visible rather than deferred. The city’s infrastructure, services, and social fabric have been repeatedly stressed by war-related shocks that cascade across sectors and expose system fragilities [2]. In such an environment, classical, long-range master plans and siloed programmes underperform. What is needed is an integrated approach that joins adaptive governance, data-driven technologies, and explicit attention to human well-being [3]. This paper takes Kyiv as a focal case to examine how strategic urban sustainability and resilience can be managed in a BANI world, and to propose a practical, evidence-informed pathway for post-conflict recovery and long-term urban prosperity.

We use BANI, Brittle, Anxious, Non-linear, Incomprehensible, in the sense introduced by Jamais Cascio (2020) as a framing for environments where failure can be abrupt, responses are shaped by anxiety, dynamics are non-linear, and outcomes may remain difficult to interpret even with more data [4]. Cascio’s articulation positions BANI as a successor to VUCA once routine volatility and ambiguity become background conditions.

Brittleness surfaces when single points of failure and over-optimised networks break under stress. Anxiety grows with uncertainty, misinformation, and disrupted routines [5]. Non-linearity means small events can trigger outsized, unexpected outcomes. Moreover, incomprehensibility reflects decision contexts saturated with data but short on clarity [6]. In Kyiv, these dimensions are visible: targeted damage to energy and transport networks, population movements that defy linear projections, and complex, fast-changing operational pictures that challenge traditional planning tools. The city’s urgent tasks, restoring essential services, protecting the vulnerable, stewarding scarce resources, and keeping public trust, must unfold under conditions where causes and effects are irregular and decisions must be revised as evidence changes [7,8].

This local reality is embedded in a national reconstruction challenge of unprecedented scale. Current joint assessments by the Government of Ukraine, the World Bank, the European Commission, and the UN estimate a decade-long recovery and reconstruction bill of roughly US$524 billion (as of 31 December 2024), underscoring the need to allocate resources to interventions that are resilient to shocks and adaptable over time [9]. In parallel, international programmes led by UN-Habitat and partners emphasise “smart and resilient urban recovery,” strengthening utilities and municipal capacities to manage uncertainty [10].

Ukraine’s urban recovery challenge is shaped by a distinctive inherited form of urbanisation. Many cities retain Soviet-era planning logics (large residential microdistricts, high-rise prefabricated housing) and heavy reliance on centralised, networked utilities (district heating, water and wastewater systems, trunk substations and transport corridors). These configurations can be efficient in peacetime but become brittle under repeated shocks because failures propagate across interdependent networks. The current war amplifies this legacy through uneven infrastructure damage, spatially concentrated housing loss, and large-scale displacement that reshapes service demand across districts. For Kyiv, this means that reconstruction and well-being interventions must be geographically targeted, prioritising neighbourhoods where damage, displacement pressure, and service disruption jointly constrain recovery, rather than relying on citywide averages.

Recent syntheses stress that urban resilience is multidimensional (infrastructure, institutions, communities, and households) and must be advanced with adaptive, learning-oriented governance rather than static blueprints [11]. Yet there is limited work that operationalizes these principles under wartime disruption, where shock frequency, information ambiguity, and decision delays are atypically high [12]. Our paper contributes by translating resilience and adaptive governance into an implementable agenda for Kyiv’s post-conflict trajectory.

Reviews of urban digital twins (UDTs), AI-enabled analytics, and IoT show accelerating adoption for monitoring, forecasting, and scenario testing in complex city systems [13]. These tools can convert high-volume, heterogeneous data into actionable signals, precisely the function needed when decision environments feel “incomprehensible” [14]. However, cybersecurity risks to smart grids and connected assets can become brittleness multipliers unless secure-by-design practices are embedded [15]. Our analysis integrates these dual realities: the promise of AI/UDTs/IoT to reduce uncertainty and the need to harden the digital layer that now underpins critical urban services.

Evidence continues to show that access to high-quality green/blue spaces and walkable, mixed-use neighbourhoods protects mental health, especially for disadvantaged groups, and that digital mental health supports can extend reach when in-person care is constrained [16]. Kyiv’s recovery must therefore couple physical reconstruction with equitable investments in restorative public spaces and blended care models that address chronic stress.

A second gap concerns circular economy strategies in post-disaster settings. While circularity is well-established in urban sustainability, its systematic use in rapid rebuilding after large-scale destruction is still emerging. New studies and initiatives support rubble recycling and material loop-closing as viable pathways to reduce costs, carbon, and supply risk, directly countering system brittleness [17]. This paper extends that discussion to Ukraine’s context, aligning circular practices with debris management and local industrial capacity realities.

Against this backdrop, the paper pursues three aims:

• To frame post-invasion urban recovery through a BANI lens by characterising how brittleness, anxiety, non-linearity, and incomprehensibility manifest in Ukrainian cities and what this implies for planning under uncertainty.
• To synthesise an integrated recovery approach that links governance, infrastructure resilience, digital capacity, and circular-economy principles into a coherent set of policy and implementation pathways.
• To demonstrate applicability through an anchored case by using Kyiv to illustrate how the framework can be operationalized to support prioritisation, sequencing, and accountability in recovery decisions.

Our contribution is fourfold.

• A BANI-to-policy translation artefact. We provide a structured mapping from BANI conditions to concrete policy design requirements (e.g., redundancy and modularity for brittleness, trust and risk communication for anxiety, adaptive sequencing for non-linearity, and transparent metrics for incomprehensibility).
• An integrative recovery framework. We offer a modular structure that connects institutional coordination, critical infrastructure, digital services, and sustainability objectives, making cross-sector dependencies explicit.
• An operationalization toolkit for decision-makers. We outline practical levers and evaluation hooks (principles, criteria/indicators, and governance checks) that enable monitoring, adjustment, and learning during implementation rather than only ex post assessment.
• A case-grounded demonstration for Ukraine’s context. We apply the framework to Kyiv to show how the proposed structure supports actionable choices (priorities, sequencing, and trade-offs) under disruption and resource constraints.

We adopt a pragmatic stance: leverage technologies that clarify complex systems while building capacity to adjust decisions as conditions change. AI-enabled predictive analytics and digital twins are positioned as “sense-making” tools in non-linear, data-rich environments, not as replacements for public judgement [18]. At the same time, we assume adversarial risk: smart grids and sensor networks must be architected and governed for resilience to cyber-physical attack.

Finally, we treat Kyiv’s recovery as an opportunity to institutionalise adaptiveness: continuous monitoring, rapid learning cycles, iterative policy updates, and participatory co-creation with communities, youth, and civic actors. These practices are not auxiliary, they are the operating system of urban management under BANI.

Section 1 characterises BANI in the Ukrainian urban context. Section 2 characterises how brittleness, anxiety, non-linearity, and incomprehensibility manifest in Ukrainian cities under wartime disruption. Section 3 derives design principles for resilient and sustainable recovery in such conditions, including adaptive governance, resilience-by-design, and circular reconstruction. Section 4 examines enabling technologies (AI, IoT, and urban digital twins) and the cybersecurity requirements needed for their safe deployment in a contested environment. Section 5 foregrounds human well-being and equity, including mental health-oriented social infrastructure and participation mechanisms that shape trust, uptake, and access. Section 6 translates these elements into an implementation and monitoring model (collaboration, continuous evaluation, and policy/funding mechanisms) with an indicator set suitable for iterative adjustment. Section 7 then operationalizes the full framework as a portfolio-optimisation problem in which policy levers u_i(t) allocate resources across infrastructure, energy, digital monitoring, and social/well-being investments. At the same time, BANI mechanisms are encoded as stress thresholds, shock-driven risk, non-linear tipping dynamics, and decision delays. Section 8 and Section 9 discuss implications, limitations, and research priorities for Ukraine’s long-term urban transition.

2. Understanding the BANI Environment in the Ukrainian Urban Context

2.1. Literature Review

Urban sustainability and resilience have become central concerns as cities confront complex and unpredictable pressures, particularly in BANI (brittle, anxious, non-linear, and incomprehensible) contexts. Under such conditions, strategic management requires adaptive and flexible approaches to respond to rapid change and uncertainty. The Ukraine case illustrates the need for tools and frameworks that enable urban transformation while reducing risk.

The BANI framework characterises fragile, volatile, and difficult to interpret settings, calling for revised management strategies that can sustain urban resilience and sustainability [4,19]. AI can strengthen risk assessment, support flexible operations, and inform decision-making within this milieu. Empirical applications report measurable gains in project resilience, operational efficiency, and stakeholder confidence when AI is embedded in planning and delivery [20].

A robust resilience posture depends on interlocking capacities. Adaptive capacity draws on education, health, and basic provisioning (food and water); absorptive capacity relies on community support, green spaces, and core infrastructure; and transformative capacity involves technology adoption, multi-stakeholder collaboration, emergency services, and integrated urban planning [21]. Effective governance balances near-term resilience with long-term sustainability, aligning responses to shocks with the pursuit of enduring environmental and social goals [22].

Independent analyses document how Ukraine’s pre-war digital foundations enabled continuity and rapid service adaptation under invasion. OECD and Brookings point to the Diia ecosystem, the Trembita interoperability platform, and e-procurement (Prozorro) as pillars of resilience and recovery governance; these sources also trace policy and institutional reforms behind the technology layer [23,24]. UNDP reports detail how online public services were expanded and kept accessible during displacement and shocks, while sector studies describe continuity and adaptation in procurement and service delivery [25,26]. Together, these works situate our case within a broad evidence base, beyond our own prior publications.

Digital city frameworks and portfolio-level project strategies can underpin sustainable urban management by fostering interdisciplinary collaboration and public-sector innovation beyond simple technology uptake [27]. Closing the gap between research and practice remains essential so that analytic insights translate into concrete pathways for transforming cities toward resilient and sustainable futures [28].

2.2. Defining the BANI Framework

The BANI framework provides a comprehensive vocabulary for describing the chaotic and unpredictable nature of the modern world. Each element highlights a distinct dimension of this contemporary reality:

• Brittleness (B). This term refers to systems that, despite appearing robust, possess an inherent fragility and potential to collapse under stress. This fragility often arises from global interconnectedness and an overemphasis on efficiency at the expense of redundancy and resilience. In urban systems, this translates to critical infrastructure, such as power grids and transportation networks, being vulnerable to cascading failures from minor disruptions. For Kyiv, this is acutely manifested in the direct impact of military actions on its infrastructure, where seemingly stable systems can be instantly shattered, leading to widespread disruptions.
• Anxiety (A). This describes a pervasive emotional and psychological toll stemming from constant uncertainty, rapid change, and information overload. The spread of misinformation can exacerbate this anxiety, leading to risk aversion, feelings of helplessness, and potentially inaction among individuals and organisations. In urban contexts, this manifests as heightened stress among residents due to unpredictable changes and a general sense of uncertainty about the future of their city. Kyiv’s residents experience profound anxiety due to ongoing air raids, displacement, and the constant threat of conflict, which deeply impacts their daily lives and decision-making.
• Non-linearity (N). This characterises situations where cause-and-effect relationships are not predictable. Small inputs can lead to disproportionately large, unexpected, and complex outcomes. This complicates traditional forecasting and planning, as processes can spin out of control with slight disruptions. Climate change, with its far-reaching and often surprising impacts, serves as a prime example of non-linearity in environmental systems. In Kyiv, the non-linear impacts of conflict are evident in unpredictable population displacements, disruptions to supply chains, and the cascading failures of interconnected urban systems due to targeted attacks.
• Incomprehensibility (I). This denotes problems so complex and fast-moving that they defy easy understanding or clear solutions, even when abundant data is available. The sheer volume of information can lead to cognitive overload and a feeling of being overwhelmed, making rational decision-making challenging. This is particularly relevant when dealing with AI technologies whose decision-making processes can be difficult to rationalise or fully explain. For Kyiv, the scale and speed of wartime destruction, coupled with the complexities of humanitarian aid, internal displacement, and future reconstruction, create an environment where traditional planning models struggle to grasp the full scope of challenges and solutions.

2.3. Manifestations and Implications for Ukrainian Urban Systems

The characteristics of the BANI environment manifest in distinct ways within urban systems, presenting unique challenges for sustainable development [29]. Kyiv’s experience provides a stark illustration of these manifestations.

The fragility of infrastructure is a significant concern. Urban infrastructure, often designed for maximum efficiency, lacks the inherent resilience needed in a brittle environment. The interconnectedness of these systems means that a failure in one component can trigger widespread disruptions across the entire urban system. For instance, the COVID-19 pandemic exposed the brittleness of global health infrastructures, demonstrating how interconnected systems can collapse under stress. In Kyiv, this is acutely evident in the deliberate targeting of critical infrastructure, such as energy grids and transportation hubs, which leads to cascading failures across the city’s essential services. This infrastructure damage directly exacerbates crises, hindering recovery efforts and increasing negative outcomes, such as amplified disease transmission during public health emergencies. This pattern reveals that urban systems are not merely facing isolated challenges but are characterised by deeply interconnected vulnerabilities, where a disruption in one domain can trigger or worsen crises across multiple others, including public health and social stability. This necessitates that urban planning in a BANI environment must transcend siloed problem-solving, requiring a holistic, systemic approach to identify and mitigate these cascading risks, focusing on strengthening interdependencies and fostering cross-sectoral resilience rather than addressing individual components in isolation.

Unpredictable resource dynamics further complicate urban sustainability. The non-linear nature of the BANI environment makes urban resource management highly unpredictable. Challenges such as climate change, resource depletion, and large-scale migrations introduce complex, non-linear patterns that defy traditional linear resource allocation and sustainability models. Minor policy adjustments or external shocks can have unforeseen and disproportionate impacts on urban resource flows. For Kyiv, the ongoing conflict has led to unpredictable disruptions in supply chains, energy resources, and labour availability, making long-term resource planning exceptionally challenging.

Another critical implication is cognitive overload in decision-making. The vast and intricate datasets generated by urban systems, combined with the rapid pace of change, can overwhelm urban planners and policymakers. This “incomprehensible complexity” hinders their ability to grasp interdependencies fully and make informed decisions, leading to potential missteps or inaction. In Kyiv, the sheer volume of real-time data from conflict zones, humanitarian needs, and reconstruction demands can create an overwhelming environment for decision-makers, where traditional analytical tools may fall short.

Finally, heightened societal anxiety directly results from the constant flux and uncertainty inherent in a BANI urban environment. This contributes to chronic psychological strain among residents. Such anxiety, amplified by misinformation and a perceived lack of control, can lead to risk aversion, social fragmentation, and a reluctance to engage in adaptive behaviours, ultimately impacting collective urban resilience. Kyiv’s population experiences severe psychological strain due to the constant threat of war, displacement, and the loss of loved ones, necessitating targeted interventions to support mental well-being and foster social cohesion.

2.4. The Human Dimension: How Urban Design and BANI Characteristics Impact Mental Well-Being and Social Equity in Ukraine

Urban design is fundamentally a matter of public health, capable of intensifying psychological strain within populations. Environmental stressors such as relentless noise, overcrowding, insufficient public transport, and a scarcity of accessible green and communal spaces contribute significantly to chronic psychological distress, leading to the emergence of “anxious cities”. The concept of “anxious cities” explicitly links urban design elements, including “spatial fragmentation and a lack of equitable access to green or social infrastructure,” to “chronic psychological strain”. To avoid repetition, we use this subsection to outline the causal pathway from BANI stressors to health and equity outcomes. The detailed evidence-based and actionable design and service interventions (green/blue infrastructure, mobility, social infrastructure, and targeted mental health supports) are developed in Section 5.

For Ukraine, where residents have endured prolonged periods of stress, displacement, and trauma, the need for mental health-promoting urban design is paramount. Equitable access to mental health-promoting urban elements, particularly green and blue spaces, is crucial. Systemic disparities, often rooted in historical patterns of spatial exclusion and uneven investment, mean that lower-income neighbourhoods frequently lack access to nearby nature, exacerbating stress and depression risks. Empirical evidence demonstrates that residents living near “biodiverse, high-quality green areas reported significantly fewer mental health complaints, independent of income or background,” and that children with such access had a “significantly lower risk of developing mental health disorders later in life”. This establishes a direct causal link between physical urban planning decisions and long-term mental health outcomes, particularly highlighting the disproportionate impact on vulnerable populations due to systemic inequalities in access. Integrating natural elements into urban planning as a fundamental right, rather than a mere amenity, is therefore a humane and preventive investment in collective mental health, especially for children and vulnerable populations. This compels urban planners to embed mental health and social equity considerations into the foundational stages of urban development. Proactive design that prioritises equitable access to mental health-promoting environments and services is not just a social good but a strategic imperative for mitigating the “anxious” aspect of BANI, fostering collective resilience, and reducing long-term societal burdens.

Furthermore, long and stressful commutes on fragmented or overcrowded transport systems deplete psychological resilience, eroding opportunities for rest, social connection, and psychological recovery [30]. Chronic exposure to elevated noise levels is directly linked to sleep disruption, anxiety, and depression. In Kyiv, the destruction of transportation infrastructure and the constant threat of air raids exacerbate these stressors, making the provision of safe, accessible, and psychologically restorative urban spaces even more critical for the city’s recovery and the well-being of its citizens [31]. The spontaneous community self-organisation seen in Kyiv during the initial stages of the conflict, where neighbours pooled food, medical students ran first aid training, and teachers arranged online lessons, underscores the inherent human capacity for resilience and the importance of fostering strong community networks through urban design.

Table 1. Characteristics of the BANI environment and its urban manifestations in Kyiv.

BANI Characteristic	Definition	Urban Manifestation/Implication in Ukraine	Key Urban Challenge in Ukraine
Brittle	Systems that appear robust but can collapse under stress.	Direct targeting and destruction of critical infrastructure (energy grids, transportation, housing) leading to cascading failures.	Maintaining essential services and ensuring rapid recovery amidst ongoing attacks and widespread damage.
Anxious	Pervasive emotional and psychological toll from constant uncertainty, rapid change, and information overload.	Heightened stress, trauma, and psychological strain among residents due to air raids, displacement, and misinformation.	Mitigating chronic psychological distress, fostering mental well-being, and rebuilding social trust.
Non-linear	Cause-and-effect relationships are not predictable; small inputs lead to disproportionately large, unexpected outcomes.	Unpredictable population displacement, disruptions to supply chains, and unforeseen consequences of infrastructure damage on urban systems.	Forecasting future needs, adapting to rapid demographic shifts, and managing complex interdependencies in reconstruction.
Incomprehensible	Problems so complex and fast-moving that they defy easy understanding or clear solutions, even with abundant data.	Overwhelming scale of destruction, complexity of humanitarian needs, and challenges in coordinating massive reconstruction efforts amidst uncertainty.	Developing holistic, adaptive planning strategies that can make sense of chaos and guide long-term recovery.

shows BANI dimensions in Kyiv: definition, observed urban manifestation, and the primary challenge for service delivery and equity. Each row links a BANI characteristic to a plain-language definition, the urban manifestation observed in Ukraine, and the primary challenge it creates for continuity of services, recovery, and fairness.

3. Foundational Principles for Sustainable Urban Development in a BANI World: Ukrainian Path to Resilience

3.1. Adaptive Governance and Planning

Adaptive governance is a decision-making approach emphasising collaboration, flexibility, and inclusivity. This approach is essential for urban systems due to their constant evolution and the critical need for responsiveness to new information and changing conditions. Key practices include developing participatory governance structures, such as citizen juries and participatory budgeting, which empower community members in decision-making processes. Furthermore, fostering robust partnerships between government, the private sector, and civil society is crucial, as is encouraging continuous experimentation and learning from failures [32]. This allows for iterative policy development and adjustment based on real-world outcomes [33].

For Kyiv, adaptive governance has been a necessity. The city’s municipal government demonstrated remarkable agility by rapidly reprogramming its public services app (Kyiv Digital) to provide real-time air-raid alerts, shelter directions, and pharmacy supplies updates during wartime.

During the first weeks of the full-scale invasion, the city repurposed Kyiv Digital from transport/parking toward civil-protection functions: push air-raid alerts and “all clear” notifications, in-app navigation to official bomb shelters via the KCSA shelter registry, and live maps of operating essentials (incl. pharmacies) [34]. These features are documented by the Kyiv City State Administration and the app publishers, and contemporaneous reporting details the 24-h pivot and the addition of shelter routing and essential-supplies layers [35].

This exemplifies the flexibility and responsiveness required in a BANI environment. Adaptive management improves resilience to uncertainty and change, enhances stakeholder engagement, and leads to better decision-making through continuous monitoring. Scenario planning is a powerful tool for anticipating and preparing for uncertainty, enabling planners to identify potential risks and opportunities before they fully materialise. Traditional urban planning often relies on fixed, long-term master plans, assuming a predictable future. However, the BANI environment is defined by “non-linear dynamics” and “incomprehensible complexity,” where “outcomes are not always proportional to inputs” and “small changes can have significant effects”. This inherent unpredictability renders rigid plans ineffective. Adaptive governance directly counters this by emphasising “flexibility and responsiveness to changing conditions,” “continuous monitoring and evaluation,” and “experimentation and learning from failure”. This represents a fundamental shift from a static blueprint to an iterative, learning-based system, a necessary response to BANI’s inherent unpredictability and incomprehensibility. This implies a fundamental re-evaluation of how urban planning is conceived and executed, necessitating the institutionalisation of feedback loops, data-driven adjustments, and fostering a culture that views “failures” as crucial learning opportunities rather than terminal setbacks. This proactive, continuous learning approach is vital for navigating BANI’s non-linear and incomprehensible aspects. For Kyiv’s reconstruction, this means moving beyond a single master plan to an iterative, flexible planning process that can adapt to evolving security situations, demographic shifts, and resource availability.

3.2. Building Urban Resilience

Urban resilience is defined as the capacity of a city’s systems, businesses, institutions, communities, and individuals to survive, adapt, and thrive in the face of chronic stresses and acute shocks. It requires a holistic assessment of capacities and risks, with meaningful engagement from all community members, especially the most vulnerable [36].

Building urban resilience includes incorporating green infrastructure, such as parks, gardens, and green roofs, to mitigate the urban heat island effect, reduce air pollution, manage stormwater runoff, and enhance biodiversity and public well-being. Designing mixed-use developments promotes walkability, reduces transportation emissions, and improves residents’ overall quality of life.

For Ukraine, building resilience means reconstructing damaged infrastructure and building it back stronger and smarter, with redundancy and decentralised systems to withstand future shocks. Fostering social structures and community networks, including community gathering places and resilience hubs, strengthens the social fabric of a city, providing crucial support during emergencies and reducing anxiety among residents. This also involves ensuring local food security through regional food systems and low-energy storage and building community facilities that can serve as emergency hubs with essential services. BANI’s “anxious” characteristic manifests as pervasive uncertainty, misinformation, and feelings of helplessness [37]. While technological and infrastructural resilience are important, the definition of urban resilience extends to the capacity of “individuals, communities, institutions, businesses and systems” to adapt and grow. Strategies for building community resilience explicitly emphasise “social structures that strengthen the fabric of community” and “community gathering places”. This indicates that fostering strong social bonds and mutual support is not merely a desirable social outcome but a fundamental mechanism for collective psychological and practical resilience, directly counteracting anxiety’s isolating and overwhelming effects in a BANI world. Therefore, urban planners must intentionally design for social interaction and community building, recognising these as critical infrastructures for mental well-being and collective adaptive capacity. Investments in public spaces, community programmes, and participatory initiatives are not just “nice-to-haves” but essential components of a robust urban resilience strategy, particularly in addressing the human dimension of the “anxious” city. Kyiv’s experience with community self-organisation, where neighbours pooled food and medical students ran first aid training, demonstrates the vital role of social resilience in a crisis. Post-war reconstruction must prioritise creating and restoring such community hubs and green spaces to support psychological recovery and social cohesion.

3.3. Circular Economy Principles

Cities are uniquely positioned to spearhead the transition to a circular economy, an approach fundamentally driven by three design principles: eliminating waste and pollution, circulating products and materials at their highest value, and regenerating natural systems [38,39]. This systemic approach minimises waste and maximises resource value through continuous cycles of use and regeneration.

Notable examples include Amsterdam’s ambitious goal of becoming a fully circular city by 2050, and initiatives like BioMakeries that treat and recycle wastewater, demonstrating practical applications of circularity in urban environments. The broader transition towards circular business models directly responds to escalating environmental concerns, promoting eco-friendly products and practices across various sectors. BANI’s “brittle” characteristic refers to systems prone to collapse under stress, often due to global interconnectedness and an overemphasis on efficiency, leading to vulnerabilities in resource supply chains. The circular economy, by design, aims to “circulate products and materials (at their highest value)” and “regenerate nature”. This approach inherently reduces dependence on distant, linear supply chains and finite virgin resources, thereby directly mitigating the “brittleness” of resource availability and the “anxiety” associated with supply chain disruptions and resource scarcity. This implies that promoting circular economy principles in urban development is not merely an environmental objective but a strategic imperative for enhancing urban resilience in a BANI world [40]. It transforms cities from vulnerable consumption hubs into more self-sustaining ecosystems, reducing external dependencies and building intrinsic robustness against resource shocks and anxieties. For Kyiv, circular reconstruction can convert demolition and repair waste into regulated secondary materials, reduce dependence on imports for basic construction inputs, and create local repair and remanufacturing capacity. The strategic benefit is resilience as much as sustainability: shorter supply chains and structured reuse pathways can keep critical rebuilding on track during volatility.

4. Leveraging Technology for Urban Sustainability and Resilience in Ukraine

4.1. The Transformative Role of AI, IoT, and Digital Twins

Advanced technologies are pivotal in navigating the complexities of a BANI environment, offering capabilities that traditional methods cannot.

• AI: AI can support urban decision-making in BANI conditions by enhancing triage, forecasting, and logistics under uncertain conditions. For Kyiv, realistic near-term applications include rapid damage assessment, prioritisation of repair queues, forecasting localised humanitarian needs, and anomaly detection across energy, health, and environmental signals. These benefits are conditional on data governance, model transparency, and integration with human-led crisis protocols. Without such safeguards, AI may increase institutional complexity and perceived opacity rather than reduce the incomprehensibility that BANI highlights.
• IoT: IoT enables near-real-time visibility across transport, energy, water, and public safety systems, creating the data foundation for faster repairs and more targeted resource allocation. For Kyiv, this could include monitoring the status of critical infrastructure, neighbourhood-level air quality, and energy demand variability during crisis periods. However, IoT expansion must be designed with cyber and physical redundancy in mind, especially where sensors and control systems interface with power and water networks. In wartime-degraded conditions, we assume partial observability and design for graceful degradation: satellite and drone assessment, utility continuity logs, and field reporting can substitute when local sensors or networks fail, and any digital monitoring should be deployable on low-power, low-bandwidth links with backup power at critical nodes.
• Digital Twins: Urban digital twins can serve as controlled test environments for reconstruction choices, helping planners compare design options under multiple threat and budget assumptions. In Kyiv, a phased approach could begin with district-level energy and mobility models that integrate outage histories, repair backlogs, and new distributed generation scenarios. The primary value is not perfect prediction, but rather improved coordination and earlier detection of trade-offs across infrastructure, climate goals, and service equity.

4.2. Closed-Loop Integration Model: Digital Twins, Mental Health Service Distribution, and Urban Resilience Indicators

Closed-loop integration model: digital twins, mental health service distribution, and urban resilience indicators.

To strengthen the cross-disciplinary integration, we specify a closed-loop mechanism that links urban digital twin capabilities, the spatial distribution of mental health services (including digital and hybrid delivery), and resilience performance indicators used in this paper. Urban digital twins are treated not only as scenario sandboxes for infrastructure choices, but as spatial decision engines for locating and operating mental health support in ways that improve equity and system robustness under BANI constraints.

Loop step 1. Spatial inputs and demand surfaces. The digital twin ingests population distribution and mobility constraints; existing service inventory (clinics, school counsellors, community hubs, mobile teams, hotlines); disruption signals (outage histories, access constraints, transport fragmentation); and anonymised demand proxies from digital mental health tools and referral pathways (e.g., hotline load, app/web utilisation, waiting times) consistent with the DMHI monitoring logic already discussed.

Loop step 2. Spatial allocation and tiered service design. Using these inputs, the twin evaluates alternative service distribution configurations (e.g., placement of community hubs, school-based programmes, mobile team routes, and low-bandwidth digital access points). Each configuration is assessed using simple, interpretable spatial-access measures, such as coverage within a defined travel-time threshold, capacity-to-demand ratios by district, and an equity gap metric (the difference in coverage between high- and low-access neighbourhoods).

Loop step 3. Policy implementation and operational controls. Selected configurations translate into actionable urban planning and service operations decisions (resource allocation, staffing, routing, facility siting, and digital/hybrid service modality). This enables AI-enabled analytics and digital twins to be operational within the governance cycle, rather than merely descriptive.

Loop step 4. Feedback into resilience indicators and model monitoring. The observed service performance and spatial access outcomes are mapped onto the paper’s composite performance logic, particularly service uptime, recovery speed, and equity of access, which are explicit components of the portfolio index F(t). Improvements in coverage and continuity of support increase the equity-of-access component directly. In contrast, reduced service bottlenecks and faster restoration of support after shocks contribute to recovery speed and uptime. These observed values then feed back into the monitoring and iterative adjustment cycle, which is already in place as necessary in BANI environments (continuous monitoring, learning cycles, course correction).

Loop step 5. Iteration and recalibration. The twin is updated with new observations (utilisation, waiting times, dropout/engagement, district-level access changes) and used to re-test placements and modalities as constraints evolve. This closes the loop from sensing—allocation—implementation—indicator change—reallocation, making the interdisciplinary integration explicit and analytically testable.

4.3. Smart Energy Management

Modern cities are increasingly adopting smart energy systems powered by advanced sensors, IoT devices, and AI-driven analytics. These systems enable dynamic monitoring and optimisation of energy consumption, significantly reducing carbon footprints and reliance on fossil fuels by integrating renewable energy sources like solar and wind [41,42].

A key strategy is decentralising energy distribution, allowing entire districts to be supplied independently. This approach reduces dependence on fragile central networks, enhancing urban energy resilience in a brittle environment. BANI’s “brittle” aspect refers to systems that, while seemingly stable, are prone to breaking under stress, often due to global interconnectedness and an overemphasis on efficiency. Centralised energy grids represent a classic example of such a brittle system, vulnerable to single points of failure. The research highlights the strategic shift towards “decentralising energy distribution” to enable “entire districts to be supplied with electricity independently,” which “reduces dependence on too-fragile central networks”. Smart grids, by leveraging AI and sensors to balance supply and demand, also contribute to this distributed resilience. This indicates that transitioning towards decentralised, smart energy systems is a direct strategic response to the “brittle” characteristic of urban infrastructure in a BANI environment, significantly enhancing energy security, reducing vulnerability to large-scale disruptions, and contributing fundamentally to urban sustainability and resilience. For Kyiv, which has experienced repeated attacks on its centralised energy infrastructure, decentralising energy distribution and implementing smart grids are critical for enhancing energy security and ensuring a continuous supply to residents and essential services.

Intelligent building systems and smart home applications further optimise energy consumption, contributing to a reduced ecological footprint across urban areas.

4.4. Data-Driven Urban Planning

Advanced technology, including smart city platforms and Geographic Information Systems (GIS), combined with IoT sensors, drones, and AI-powered analytics, gathers comprehensive insights on urban dynamics such as traffic patterns, energy usage, population density, and environmental conditions.

Predictive analytics tools are crucial for forecasting future urban needs, such as housing demand and transportation requirements, enabling efficient resource allocation and early risk detection. AI solutions can predict flood risks, detect leaks, optimise energy use, and manage waste more efficiently. The non-linearity in the BANI framework implies unpredictable cause-and-effect relationships where “minor issues can lead to unexpected and complex outcomes”. This unpredictability, coupled with general uncertainty, fuels “anxiety”. Powered by AI, predictive analytics directly addresses these challenges by “analysing historical and real-time data to identify potential risks and forecast future trends”. Specific examples include AI modelling of climate scenarios to predict floods and optimise building materials for heat waves. This capability allows urban planners to anticipate and prepare for unforeseen consequences, reducing the “anxiety” associated with unpredictability and enabling more effective navigation of non-linear dynamics. Implementing robust predictive analytics capabilities is thus crucial for transforming urban planning from a reactive to a proactive discipline. This strategic use of data helps mitigate the psychological and systemic impacts of non-linearity and anxiety by providing foresight and enabling preemptive interventions, making urban development more adaptable and resilient. In Kyiv, data-driven urban planning is essential for managing the influx and outflow of displaced populations, prioritising housing reconstruction, and optimising resource allocation for recovery efforts, all while adapting to the unpredictable nature of the conflict.

Smart traffic systems, informed by real-time data, reduce congestion and pollution, further contributing to urban sustainability.

4.5. Cybersecurity as a Critical Enabler

The increasing reliance on digital technologies in urban systems introduces significant cybersecurity risks. Cyber threats substantially threaten urban energy systems, potentially leading to widespread blackouts, transportation disruptions, and compromised public safety.

To mitigate these risks, cities must adopt a multi-layered approach to energy security, embedding “secure-by-design” principles from the outset of urban planning and infrastructure development. This includes robust physical safeguards, encryption, network segmentation, and continuous monitoring of digital systems. Protecting critical infrastructure involves ensuring data integrity, system availability, and safeguarding against unauthorised access, as manipulated data in smart grids could lead to incorrect energy distribution and public safety risks. BANI’s “brittle” characteristic means systems are prone to collapse under stress, and cyberattacks represent a direct form of such stress on increasingly digital urban infrastructure. Furthermore, BANI’s “anxious” aspect is exacerbated by misinformation and feelings of powerlessness. A major cyberattack on critical urban services could cause physical disruption, severely erode public trust, and amplify societal anxiety. The emphasis on “secure-by-design principles” and protecting “data integrity, system availability, and protection against unauthorised access” suggests that cybersecurity is not merely a technical add-on but a foundational element for maintaining the integrity and trustworthiness of the digital systems that underpin urban sustainability. Therefore, cybersecurity must be a core strategic pillar in all smart city initiatives. Without robust cybersecurity, the transformative benefits of digital transformation for urban sustainability in a BANI world are severely undermined by the risk of catastrophic failure and the erosion of public confidence. It is about ensuring the foundational stability and trustworthiness of the digital layer that supports urban resilience. For Kyiv, which has been a target of extensive cyber campaigns, robust cybersecurity is not just a best practice but a national security imperative for its urban systems, protecting critical infrastructure and combating misinformation that fuels anxiety.

Table 2. Key technologies for sustainable urban development in a BANI context: relevance for Kyiv.

Technology/Tool	Specific Application in Urban Development	BANI Characteristic Addressed	Key Benefit for Sustainability/Resilience in Kyiv
AI	Predictive analytics for damage assessment, reconstruction logistics, humanitarian needs; dynamic simulations for urban planning; misinformation countering; mental health support.	Incomprehensibility, Non-linearity, Anxiety.	Enhanced decision-making for rapid recovery, proactive risk mitigation in conflict zones, optimised resource allocation for reconstruction, and psychological support for residents.
IoT	Real-time monitoring of damaged infrastructure, energy distribution, public safety, resource consumption.	Brittleness, Anxiety, Incomprehensibility.	Immediate situational awareness during attacks, efficient resource management for emergency response, and data-driven insights for rebuilding efforts.
Digital Twins	Immersive virtual environments for reconstruction planning, simulating resilience of new designs against threats, optimising material/energy flows in rebuild.	Incomprehensibility, Non-linearity.	Risk-free testing of urban regeneration plans, optimised infrastructure design for resilience, and identification of bottlenecks in post-war recovery.
Smart Grids	Decentralised energy distribution, integration of renewables, dynamic optimisation of energy consumption.	Brittleness, Non-linearity.	Enhanced energy security against attacks, improved energy efficiency during reconstruction, and increased urban resilience to power disruptions.
Predictive Analytics	Forecasting housing demand for displaced populations, anticipating resource needs for reconstruction, early risk detection (e.g., disease spread, panic).	Non-linearity, Anxiety.	Proactive urban planning for demographic shifts, reduced uncertainty in resource allocation, and timely interventions for public health and social stability.
Cybersecurity Solutions	Secure-by-design principles for urban infrastructure, encryption, network segmentation, and continuous monitoring against cyberattacks.	Brittleness, Anxiety.	Safeguarding critical urban services from cyber warfare, maintaining public trust in digital systems, and ensuring operational continuity during conflict.

shows the key technologies for sustainable urban development in a BANI context. The table maps each technology/tool to its specific application in urban management, the BANI characteristic addressed, and the key benefit for sustainability/resilience in Kyiv.

5. Cultivating Well-Being in the “Anxious” City: Kyiv’s Human-Centred Recovery

5.1. The Socio-Economic Burden of Mental Health in Urban Populations

Mental health challenges impose a substantial economic burden on societies, costing Europe an estimated €600 billion annually. This figure is primarily attributed to lost productivity, increased healthcare costs, and social welfare expenditures, representing over 4% of the EU’s Gross Domestic Product (GDP). The prevalence of mental health issues has worsened, particularly among young people, with almost one in two young Europeans (aged 15–24) reporting unmet mental healthcare needs.

Youth mental health problems significantly impact educational attainment, leading to lower school attendance and poor learning outcomes, especially concerning exams. These issues also increase the likelihood of young people becoming “Neither in Employment nor in Education or Training” (NEET), with high school dropout playing a notable, though small, mediating role in this relationship. In the United States, the ripple effects of the adolescent behavioural health crisis are estimated to result in up to $185 billion in lifetime medical costs and a staggering $3 trillion in lifetime lost productivity and wages. The research explicitly details the immense economic burden of mental health challenges, with figures in the trillions for lost productivity and billions for healthcare costs. Crucially, the data connects childhood psychological problems to “diminished educational accomplishments,” “reduced adult family incomes by 20%,” and “seven fewer weeks worked per year”. It further states that neglecting adolescent mental health can result in severe long-term consequences, including limiting opportunities for adolescents to lead meaningful lives. This demonstrates a compounding, long-term economic and social cost that extends across individuals’ lifespans and significantly impacts overall societal productivity and welfare, far beyond immediate healthcare expenditures. This elevates investment in youth mental health from a mere social expenditure to a critical economic imperative with substantial long-term returns. Policymakers and urban planners should recognise that the “anxious city” is not just a psychological concept but a significant economic drain. Therefore, interventions that foster youth mental well-being directly contribute to future urban prosperity, economic stability, and overall sustainability.

For Kyiv, the socio-economic burden of mental health is exacerbated by the direct trauma of war, displacement, and the disruption of education and employment. Addressing these challenges is a humanitarian imperative and crucial for the city’s long-term economic recovery and social stability.

5.2. Impact of Urban Design on Psychological Strain

Building on the “human dimension” link established in Section 2.4, this subsection focuses on operational levers. How specific design choices and service configurations reduce psychological strain and inequity under Kyiv’s BANI conditions.

Urban design is fundamentally a matter of public health, as it can significantly intensify psychological strain within populations. Key environmental stressors include relentless noise, overcrowding, insufficient public transportation, and a scarcity of accessible green and communal spaces, all contributing to chronic psychological strain and the emergence of “anxious cities”.

Equitable access to green spaces is paramount for mental well-being. Systemic disparities, often rooted in historical patterns of spatial exclusion and uneven investment, mean that lower-income neighbourhoods frequently lack access to nearby nature, exacerbating stress and depression risks. The concept of “anxious cities” directly links specific urban design elements, such as “spatial fragmentation and a lack of equitable access to green or social infrastructure,” to “chronic psychological strain”. The evidence provides strong empirical support that residents living near “biodiverse, high-quality green areas reported significantly fewer mental health complaints, independent of income or background,” and that children with such access had a “significantly lower risk of developing mental health disorders later in life”. This compelling evidence transforms the perception of urban green and blue spaces from mere aesthetic amenities to critical, long-term public health infrastructure, particularly for mental well-being and social equity. Integrating natural elements into urban planning as a fundamental right, particularly for children and vulnerable populations, represents a humane and preventive investment in collective mental health. This implies that urban planning must prioritise creating and equitably distributing high-quality green and blue infrastructure as a core, rather than peripheral, strategy for public mental health. This is a direct, tangible way to mitigate the “anxious” aspect of BANI environments at the human scale, fostering resilience, reducing long-term healthcare burdens, and promoting social equity.

Furthermore, long and stressful commutes on fragmented or overcrowded transport systems deplete psychological resilience, eroding opportunities for rest, social connection, and psychological recovery. Chronic exposure to elevated noise levels is directly linked to sleep disruption, anxiety, and depression. In Kyiv, the reconstruction efforts must prioritise the creation of accessible green spaces, quiet zones, and efficient, safe public transport to mitigate the psychological strain on its residents, many of whom have experienced profound trauma.

5.3. Digital Mental Health Interventions

Digital Mental Health Interventions (DMHIs) offer scalable solutions to address the urgent need for mental health treatments, particularly for disadvantaged populations with limited access to traditional care. They can overcome barriers such as geographical isolation, inconvenience, privacy concerns, and anonymity, providing flexible and accessible support. DMHIs encompass various digital tools, including smartphone apps, web-based resources, chatbots, and virtual reality (VR) systems. Common therapeutic modalities in these interventions include Cognitive Behavioural Therapy (CBT), mindfulness, and positive psychology.

AI-powered tools are increasingly pivotal, enabling early detection and diagnosis through behavioural analysis and predictive analytics. They also provide therapeutic solutions such as chatbots, virtual therapists, and personalised therapy programmes for youth. AI models can proactively identify adolescents at high risk for future mental health issues by analysing underlying causes like sleep disturbances and family conflict. VR interventions create lifelike environments for training and treatment, allowing for personalised therapeutic experiences and utilising virtual characters for social interaction and engagement. VR therapy has demonstrated effectiveness for anxiety disorders, PTSD, and phobias, and can improve social skills in a safe, controlled environment.

Blended care models, which combine digital tools with human support (e.g., therapists, coaches, digital navigators), are particularly effective in improving user adherence and therapeutic outcomes. Human guidance is consistently found to augment the effects of digital interventions and is crucial for engagement and adherence, with users often preferring interventions that include professional support. The paradox of digital accessibility is evident here: DMHIs are widely lauded for their “enhanced accessibility, flexibility, and scalability,” and their potential to “overcome perceived barriers such as privacy and anonymity”. This suggests a democratisation of mental healthcare. However, the evidence also explicitly identifies “digital disparities” as a significant barrier, where “children and adolescents in low-income communities often lack access to devices or stable internet, limiting their participation”. This creates a critical paradox: while digital tools can increase access, existing socio-economic inequalities in digital infrastructure and literacy can inadvertently exacerbate mental health disparities, undermining the very goal of equitable access. This implies that simply developing and deploying digital mental health tools is insufficient. Urban sustainability strategies must actively invest in bridging the digital divide by providing equitable access to digital infrastructure and implementing comprehensive digital literacy programmes, particularly in underserved communities.

In Ukraine’s post-conflict setting, digital access is not uniform and should be treated as an explicit precondition for AI-enabled mental health support. We therefore propose a rapid Digital Accessibility Assessment (DAA) at the neighbourhood/district level that distinguishes core urban zones (typically characterised by higher connectivity and service density) from peripheral or heavily disrupted areas (often characterised by intermittent networks, unstable electricity, and constrained device availability). The DAA evaluates five practical dimensions: network availability and bandwidth (including outage frequency), electricity reliability (charging feasibility), device access (smartphone ownership/shared devices), digital literacy and language accessibility, and safe access points (schools/community hubs where privacy and support can be ensured).

Based on the DAA, DMHIs should be implemented in tiers rather than uniformly: Tier A (high access—core areas) enables full-featured tools (apps, secure chat and telehealth, AI triage, analytics) with blended human support; Tier B (intermittent access—transition zones) prioritises low-bandwidth and asynchronous modalities (lightweight web, downloadable content, SMS/USSD-style check-ins, hotline + navigator support) and reduces reliance on continuous connectivity; Tier C (low access—disrupted areas) shifts to offline-first delivery anchored in community hubs and mobile teams (paper-based screening, in-person group support, radio/printed psychoeducation), with digital tools used only when connectivity is available. This tiering ensures that AI-enabled support functions as an equity-preserving resilience measure: the modality is adapted to infrastructure conditions so that the digital layer does not widen gaps during disruption.

This ensures that DMHIs genuinely serve as tools for health equity rather than inadvertently widening existing gaps, which is crucial for addressing BANI’s “anxious” aspect through inclusive solutions.

From a Social-Ecological Systems (SES) perspective, urban resilience depends not only on physical infrastructure but also on human and social capital (mental health, trust, and social cohesion) and on the information and governance mechanisms that enable rapid adaptation during shocks. In this framing, DMHIs function as a system-layer intervention: at the micro-level, they support individual coping and continuity of care during disruptions; at the meso-level, they strengthen linkages between households, schools, community hubs, and clinicians (e.g., triage, navigation, blended care); and at the macro-level, they provide an interoperable channel for municipal and health-system coordination (including privacy-preserving, aggregated demand signals) that improves adaptive response and equitable resource allocation. This is consistent with Ecosystem Theory’s multi-level logic (individual-community-institutional-policy), and it clarifies why closing the digital divide is not a standalone equity issue, but a resilience constraint: without inclusive digital infrastructure and literacy, interventions cannot reliably operate across system levels under BANI conditions.

For Kyiv, digital mental health interventions are crucial for providing accessible support to a population experiencing widespread trauma and displacement, complementing overstretched traditional services. This includes AI-enabled mental health support as part of public health projects in Ukraine.

Despite their potential, DMHIs face significant challenges, notably low user engagement due to poor usability, privacy and data security concerns, scepticism about benefits, limited digital literacy, and a lack of personalisation. Furthermore, evidence from recent large trials and reviews suggests that some universal, school-delivered mental health programmes can be associated with null effects or even adverse outcomes for certain measures or subgroups at follow-up, underscoring the need for careful programme selection, fidelity monitoring, and equity-aware implementation [43,44,45].

To enable the empirical evaluation of mental health effects in future implementations, outcomes can be tracked using a small set of standardised indicators across symptom burden, service access and utilisation, and functional and community outcomes. For digital interventions specifically, engagement metrics should be interpreted in conjunction with equity-of-access indicators to ensure that digital delivery does not exacerbate existing disparities.

5.4. Integrating Mental Health into Urban Policy

Quality mental health education in schools is a foundational strategy, increasing students’ mental health literacy, reducing stigma, and promoting help-seeking behaviours. Such curricula should be medically accurate, developmentally appropriate, and address key concepts like causes and symptoms of mental illness, treatment opportunities, stigma, and communication with trusted adults.

Comprehensive School Mental Health Systems (CSMHS) provide a tiered approach to support (universal, targeted, individualised interventions) and necessitate robust collaboration among educators, mental health professionals, and families. Evidence-based decision-making in public health is paramount, involving the systematic use of the best available research data, practitioner expertise, and community preferences to inform programme and policy development. Resources like SAMHSA’s Evidence-Based Practices Resource Centre provide essential tools and frameworks for policymakers and practitioners.

Youth-centred practices (YCP) are critical for effective policy integration, emphasising young people’s active engagement and participation in the design, delivery, and evaluation of their healthcare needs. This approach fosters autonomy and ensures interventions are relevant and impactful. Some universal school-based mental health interventions have shown unintended negative effects (including higher emotional difficulties or worse scores on selected outcomes at follow-up) in a subset of high-quality evaluations, although mechanisms and susceptibility factors remain insufficiently understood. In line with global guidance that recognises the low certainty of evidence for universally delivered psychosocial prevention in adolescents, implementation should incorporate explicit monitoring of adverse outcomes alongside intended benefits [46,47]. Conversely, “youth-centred practices” explicitly advocate for “active engagement and participation of YEA in all aspects of their healthcare needs, including the design, delivery, and evaluation of their services”. This direct involvement of the target population has led to higher usability, satisfaction, and adherence. To mitigate the risk of unintended negative consequences and enhance the effectiveness of urban mental health policies and interventions, particularly for youth, genuine co-creation and youth-centric design principles must be adopted. This means moving beyond tokenistic consultation to empower young people as active partners in shaping the solutions that affect their well-being, directly addressing BANI’s “anxious” and “incomprehensible” aspects by building trust, relevance, and ownership.

To translate YCP from principle into practice, we propose a structured participation process with three components.

• Stakeholder identification matrix. Kyiv should establish a matrix that identifies youth subgroups affected by war and reconstruction (e.g., displaced/returning youth, youth with disabilities, youth in heavily affected districts, and school-attending vs. out-of-school youth), and maps them to institutional stakeholders (schools, municipal health and social services, NGOs, community hubs, and digital platform operators). The matrix specifies each group’s expected role (user, co-designer, evaluator), accessibility needs (language, disability supports), and safe engagement channels.
• Participation levels and decision rights. Participation should be explicitly classified by level of influence to prevent tokenism:
  Information sharing (transparent communication of options and constraints); consultation (structured feedback on needs and usability); co-design (joint development of programme components such as referral pathways, platform features, or school-based delivery); and co-decision-making (youth representatives hold defined voting/approval rights on selected elements such as prioritisation of interventions, service standards, and evaluation indicators). Each programme component should declare its participation level and what decisions youth input can change.
• Conflict resolution, safeguarding, and accountability. Because youth participation occurs in a high-stress context, Kyiv should implement a simple dispute-resolution and safeguarding mechanism: facilitated sessions with documented agendas and outputs; an escalation pathway to a small cross-sector steering group (municipality-school-service providers-youth reps); and a grievance channel (including anonymous reporting) for concerns related to privacy, stigma, or harmful content. Participation outputs (decisions made, items deferred, and rationale) should be published in a concise “you said—we did” log to maintain trust and ensure the process informs iterative planning updates.

For Kyiv, integrating mental health support into urban policy means prioritising trauma-informed approaches in schools and community programmes, ensuring that the city’s recovery plans address the psychological well-being of its population, particularly children and youth.

Table 3. Strategies for promoting mental well-being in urban environments: Kyiv-specific applications.

Strategy Category	Specific Action/Intervention	Key Benefit for Mental Well-Being	BANI Characteristic Addressed (Primary)	Kyiv-Specific Application
Urban Design & Green Infrastructure	Equitable access to green and blue spaces.	Reduced stress, improved emotional regulation, lower risk of mental disorders.	Anxiety	Prioritising the rapid restoration and creation of parks, public gardens, and accessible green corridors in war-affected districts for psychological recovery.
	Noise reduction zoning and acoustic sensitivity in urban design.	Reduced sleep disruption, anxiety, and depression.	Anxiety	Implementing noise mitigation strategies during reconstruction and designing quiet zones for respite from conflict-related stressors.
	Design of mixed-use developments and walkable urban areas.	Reduced psychological strain from commutes, increased physical activity, and enhanced social connection.	Anxiety	Rebuilding neighbourhoods with integrated services and pedestrian-friendly layouts to reduce reliance on fragmented transport and foster local community bonds.
Digital Mental Health Solutions	AI-powered chatbots and virtual therapists.	Immediate, on-demand support; reduced stigma; increased accessibility.	Anxiety	Deploying AI-driven mental health chatbots accessible via Kyiv Digital or other platforms to provide immediate, low-stigma support for trauma and stress.
	VR therapy for exposure and social skills training.	Safe practice environments, personalised therapeutic experiences, and improved social interaction.	Anxiety	Developing VR modules to help residents process trauma in a safe, controlled environment and practice social reintegration skills.
	Personalised therapy programmes (AI-driven).	Tailored interventions, enhanced engagement, better outcomes.	Incomprehensibility	Utilising AI to tailor mental health interventions based on individual needs and trauma responses, adapting to the unique experiences of Kyiv’s population.
	Blended care models (digital + human support).	Improved adherence, increased effectiveness, enhanced trust.	Anxiety	Integrating digital mental health tools with human support networks (psychologists, social workers, community volunteers) for comprehensive care.
Policy & Community Engagement	Comprehensive school mental health systems (MTSS, CSMHS).	Increased mental health literacy, reduced stigma, improved help-seeking, and better academic outcomes.	Anxiety	Implementing trauma-informed mental health programmes in schools to support children and youth affected by conflict.
	Youth-centred policy design and co-creation.	Increased relevance and impact of interventions, enhanced autonomy, reduced unintended negative effects.	Anxiety, Incomprehensibility	Actively involving Kyiv’s youth in designing mental health and urban reconstruction initiatives to ensure relevance and foster ownership.
	Investment in digital literacy programmes.	Bridged the digital divide, ensured equitable access to DMHIs, and increased engagement.	Anxiety	Providing digital literacy training to ensure all residents, especially vulnerable groups, can access digital mental health and urban services.
	Community-led urban sustainability initiatives.	Enhanced social cohesion, community pride, local resilience.	Brittleness, Anxiety	Supporting and empowering local community groups in Kyiv to lead reconstruction and greening projects, building on wartime self-organisation efforts.

shows strategies for promoting mental well-being in urban environments. The table lists strategy categories with specific actions/interventions, the key mental-health benefit, the primary BANI aspect addressed, and a Kyiv-specific application relevant to post-war recovery.

6. Implementation Strategies and Best Practices: Ukrainian Adaptive Recovery Model

6.1. Multi-Stakeholder Collaboration and Co-Creation

Successful implementation of complex initiatives in a BANI environment necessitates robust multi-sector collaboration and co-creation. This involves forging strong partnerships across academia, healthcare providers, policymakers, patient organisations, and citizens. Close collaboration and formal engagement with end-users and stakeholders are critical success factors, influencing content, design decisions, and dissemination strategies from the project’s outset. This ensures the relevance and usability of interventions, making them more likely to be adopted and effective. For scalable digital mental health interventions, collaboration between researchers, organisations capable of scaling, and lived experience experts is explicitly required to ensure an implementation focus from the beginning of a project.

The BANI environment is defined by “incomprehensible complexity” and “non-linear dynamics,” where traditional, siloed approaches are ineffective. The evidence consistently identifies “multi-sector collaboration and co-creation” and “close collaboration with stakeholders (end-user engagement)” as “key factors for success” in implementing complex interventions. The emphasis on engaging “diverse stakeholders” and “breaking down silos” in adaptive governance further reinforces this. This suggests that BANI challenges’ sheer complexity and unpredictable nature necessitate a collective intelligence approach, where diverse perspectives and expertise are integrated to achieve a more comprehensive understanding and adaptive response. Therefore, effective urban sustainability management in a BANI environment demands a fundamental shift from hierarchical, top-down decision-making to a networked, collaborative ecosystem. This means intentionally building cross-sectoral partnerships, fostering shared understanding, and embedding co-creation throughout the project lifecycle to harness collective intelligence and navigate the inherent complexities and non-linearities. Kyiv’s wartime experience already demonstrated powerful multi-stakeholder collaboration, with national institutions, local government, communities, and international partners working together. This collaborative spirit forms a strong foundation for future sustainable urban development and reconstruction.

6.2. Continuous Monitoring, Evaluation, and Learning

Adaptive governance frameworks for urban sustainability emphasise continuous monitoring and evaluation, utilising data and analytics to inform urban planning decisions and track key performance indicators (KPIs).

To make evidence-based decision-making operational, we propose a multi-level evaluation framework that links routine measurement to adaptive planning cycles. Metrics are organised into three layers: technical (digital system performance), social (service and equity outcomes), and resilience (shock absorption and recovery dynamics). Technical metrics ensure that the digital operating layer is reliable enough to support decision-making (e.g., monitoring platforms and registries). Social metrics track whether reconstruction translates into equitable access and trusted services. Resilience metrics directly reflect the model structure: service uptime and recovery speed inform observed resilience trajectories. At the same time, redundancy and response time serve as practical proxies for adaptive capacity consistent with Equation (5).

Measurement and feedback schedule. Technical indicators are monitored daily and weekly (automated logs), social/service indicators monthly/quarterly (administrative data + short surveys), and resilience indicators event-based and quarterly (restoration curves after shocks plus rolling summaries). Results are reviewed through a recurring Plan-Do-Study-Act cycle: if thresholds are breached (e.g., rising response times, worsening equity coverage, slower restoration), the city updates operational priorities, adjusts implementation tactics, and, at minimum quarterly, revisits the indicator targets and weights used for portfolio comparison.

Proposed indicator set:

Technical metrics: system response time for the monitoring stack (median time from event detection to dashboard update), data update frequency by source (in hours/days), platform availability and uptime, interoperability success rate across registries, and incident response time for cyber and operational disruptions.

Social metrics: service coverage at minimum standards by district (water, heat, electricity, transport), distribution-sensitive equity measure (coverage gaps for vulnerable groups), user satisfaction and trust score for municipal communications and digital services, utilisation and engagement metrics for digital supports (e.g., active users, retention), and grievance resolution time.

Resilience metrics: mean/median time-to-restoration for critical services after shocks (recovery speed), cumulative outage minutes (service uptime loss), redundancy coverage for critical facilities (share with backup power/islanding readiness), and an adaptive-capacity proxy combining redundancy and operational response time, aligned with Equation (5).

This includes regular reviews and assessments of policies and programmes to ensure ongoing relevance and effectiveness. Continuous quality improvement (CQI) is essential for digital mental health interventions, employing iterative, data-driven methods like the Plan-Do-Study-Act (PDSA) cycle. This involves closely monitoring patient DMHT use, tracking outcomes, and refining implementation processes based on feedback. Measuring the use and reach of digital resources through web metrics and analytics provides crucial feedback for iterative improvement and adaptation.

In a non-linear and incomprehensible BANI environment, “outcomes are not simply hard to foresee, they are completely unpredictable”. This means initial solutions may not work as expected, or may even have “unintended adverse effects”. Continuous monitoring and evaluation, particularly through “data-driven decision-making” and “continuous quality improvement”, provide the necessary feedback loops to detect deviations, learn from real-time performance, and “continuously adjust and improve”. This iterative process is a direct counter to unpredictability, allowing for “course correction when needed”. This implies that robust data collection, real-time analytics, and embedded evaluation frameworks are good practices and a strategic imperative for managing urban sustainability in a BANI context. This enables agile adaptation, minimises the impact of unforeseen negative outcomes, and fosters continuous learning, which is crucial when dealing with non-linear and incomprehensible systems. For Kyiv, continuous monitoring of reconstruction progress, population movements, and the effectiveness of new urban solutions will be vital for adaptive planning and resource optimisation.

6.3. Policy Frameworks and Funding Mechanisms

Effective policy frameworks for urban resilience must incorporate comprehensive risk assessment, strategic infrastructure investment, adaptive design principles, and robust community engagement. These frameworks should prioritise the needs of vulnerable populations and be flexible enough to integrate emerging trends and innovations. Funding mechanisms, such as Horizon Europe, support health research, including AI and mental health initiatives, by emphasising clear objectives, defined impact pathways, and detailed work plans. Proposals must align with broader EU strategies and embed ethical considerations and inclusivity.

For Kyiv, adaptive governance therefore includes not only data-informed feedback, but also clarification of land and property rights, permitting and procurement processes, funding negotiations with international donors, and dispute resolution across affected stakeholders. The proposed monitoring platforms are intended to increase transparency and coordination within these processes, not to replace them.

To make this more actionable for Kyiv, examples of Kyiv-specific projects that could realistically align with Horizon Europe-type calls include:

• District-level energy resilience packages that combine solar-plus-storage for critical facilities (hospitals, shelters, water pumping) with grid-hardening and islanding protocols, evaluated through outage reduction and recovery-time metrics.
• An AI-enabled reconstruction monitoring and prioritisation platform that fuses satellite or drone damage assessment, municipal asset registries, and transparent progress dashboards to support adaptive sequencing and anti-corruption accountability.
• Cyber-secure digital public services for recovery delivery, including interoperable registries (housing, benefits, permits) and incident-response-ready architectures that protect critical municipal and utility operations.
• Community-scaled digital mental health and psychosocial support for displaced and war-affected populations, integrated with primary care referral pathways and privacy-preserving analytics to guide resource allocation.

In all cases, Kyiv’s realistic role would be as a living laboratory implementation site within a consortium, with measurable impact pathways and explicit data governance and ethics safeguards.

Economic evaluations and high-quality real-world data are essential for informing mental health resource allocation decisions and shaping effective policies at both national and global levels. Cost-effectiveness and Return on Investment (ROI) analyses are crucial for demonstrating the value and justifying investment in interventions. Digital mental health policy frameworks face significant challenges, including inconsistent oversight, rapid evolution of AI-integrated tools, and critical data privacy and security concerns. Opportunities exist to improve access and adoption through targeted evidence requirements, expanded reimbursement models, fostering consensus on modern regulatory frameworks, and enhancing data privacy and security awareness.

The BANI environment is characterised by rapid change and “incomprehensible complexity,” particularly with the swift emergence and integration of technologies like AI. Digital mental health technologies rapidly evolve, with AI integration enhancing usability and performance. However, existing regulatory mechanisms are often “outdated” and “do not account for how tools may change as companies use AI to improve usability and performance”. This regulatory lag can hinder innovation and safe, widespread adoption. The call for “new regulatory authority to augment existing rules” and “a modern and efficient regulatory framework” addresses this critical gap. Policy and regulatory frameworks must transition from static, reactive models to agile, adaptive ones for urban sustainability, especially in domains leveraging advanced digital solutions. This means designing regulations that can evolve in tandem with technological advancements and unforeseen challenges, fostering innovation while rigorously safeguarding public interest (e.g., data privacy, ethical AI use) in a “non-linear” and “incomprehensible” technological landscape. For Kyiv, this means developing agile policy frameworks for urban planning and reconstruction that can adapt to the evolving security situation, integrate new technologies, and secure international funding for long-term sustainable development.

The framework’s core claim is that Kyiv’s recovery strategy should be evaluated as a portfolio of actions under BANI conditions, balancing sustainability and resilience objectives while accounting for shock risk, decision delays, and binding resource constraints. In Section 7, this is implemented by representing policy actions as time-varying control variables u_i(t) (e.g., renewable energy investment, infrastructure hardening, refugee integration spending, AI-enabled monitoring capacity), translating desired outcomes into a composite performance index F(t) built from normalised, policy-defined targets (service uptime, recovery speed, emissions intensity, equity of access), and encoding BANI mechanisms as explicit stress thresholds, shock-driven risk/variance, non-linear tipping dynamics, and entropy-linked decision delays (

Table 4. BANI dynamics.

BANI Factors	Mathematical Representation
Brittle	Infrastructure failure is Stress > Thresholdmaterial
Anxious	Risk perception ∝σ2 (Shock Frequency)
non-linear	$\mathrm{Tipping}\mathrm{points}{\displaystyle \frac{dS}{dt}}=aS-bS3+ϵ(t)$
Incomprehensible	Information entropy H(t) ≥ H0 ⇒ decision delays

).

Conceptually, the digital-capacity and governance sections motivate how measurement and adaptation occur (data streams, monitoring cadence, and participatory target-setting), the circular-economy section motivates which material and energy transitions affect the emissions and continuity components of F(t), and the mental-health and equity section motivates how distribution-sensitive access is represented within the equity component and associated constraints. The result is that each narrative pillar corresponds to a model object (variable, constraint, parameter, or metric), enabling traceable interpretation of the optimisation outputs for urban policy.

7. Case Studies and Examples of Adaptive Urban Development in BANI-like Contexts

7.1. Core Mathematical Model Framework

Section 7 formalises the preceding framework as a portfolio-optimisation problem for post-conflict urban recovery in a BANI environment. The qualitative sections define the policy objectives (continuity, recovery, low-carbon transition, equity, and well-being), the constraints (budget, minimum service requirements, and institutional feasibility), and the mechanisms that make Kyiv’s context BANI (brittleness thresholds, anxiety and risk under shock frequency, non-linear tipping behaviour, and incomprehensibility as decision delay). These elements are operationalized here through a composite performance index, F(t), aggregated from normalised indicators with explicit baselines and policy targets, a set of controllable investment/programme levers, u_i(t), that represent implementable municipal actions, and BANI-specific dynamics encoded as model terms and constraints (summarised in Table 4). The purpose is to ensure that the optimisation outputs can be interpreted as transparent, auditable prioritisation guidance, i.e., a quantitative extension of the framework rather than a separate methodological strand.

The objective is to maximise long-term urban functionality F(t) under BANI constraints.

$$F\left(t\right)=\max {\int }_0^T{e}^{-\rho t}\left[\alpha ·S(t\right)+\beta ·R(t)]dt,$$

(1)

where S(t) is Sustainability index (0–1), R(t) is Resilience index (0–1), α, β are policy weights (α + β = 1), ρ is discount rate, T is time horizon.

We evaluate portfolio performance using a composite index F(t) ∈ [0, 1] that aggregates four normalised components: service uptime, recovery speed, emissions intensity, and equity of access. Each component is normalised against a transparent baseline and a policy-defined target. The weights (w_j) is set through a participatory process involving municipal agencies, sector operators, and civil society, and then tested through sensitivity analysis to confirm that a narrow parameter choice does not drive recommended allocations. We use F(5) to compare five-year portfolio outcomes.

The four components of the composite index, service uptime, recovery speed, emissions intensity, and equity of access, are computed from a defined “data dictionary” that specifies the unit of measurement, temporal resolution, spatial unit (citywide or district-level), and provenance (administrative records vs. monitoring streams vs. survey and statistical sources). In implementation, service uptime can be derived from sector operator outage logs and continuity reports (e.g., electricity, district heating, water, transport service availability); recovery speed from time-to-restoration distributions following discrete shock events; emissions intensity from energy mix and consumption estimates mapped to standard emissions factors; and equity of access from distribution-sensitive coverage measures (e.g., the share of residents meeting a minimum service threshold by district, weighted by population and vulnerability).

Weights w_j are elicited through a structured multi-criteria decision process with municipal agencies, sector operators, and civil society (as noted above). Specifically, stakeholders first agree on the policy targets and acceptable trade-offs (e.g., minimum equity threshold versus emissions reductions versus continuity). We then elicit weights using a short, documented protocol (workshop + survey) with consistency checks (e.g., pairwise comparisons aggregated across stakeholder groups, followed by reconciliation in a joint session).

We validate the framework at three levels. First, face validity: domain experts review whether the indicators and targets reflect operational reality and statutory obligations. Second, historical plausibility: where data permit, we back-check whether the model reproduces observed directions of change in continuity and restoration under past shock episodes when applying the same baselines/targets. Third, robustness: We conduct a sensitivity analysis over weights w_j (e.g., ±20% reweighting with renormalization), alternative baselines and targets, and shock-severity parameters, and report whether the recommended portfolio rankings remain stable. These steps ensure that recommendations are not artefacts of a narrow parameter choice but persist across credible policy preferences and uncertainty ranges.

The index is dimensionless. Higher values indicate better overall performance relative to the stated targets. We report F_optimized(5) at the end of the five year horizon.

Key components:

• Sustainability sub model

$$S\left(t\right)=wE·SE\left(t\right)+wS·SS\left(t\right)+wG·SG\left(t\right),$$

(2)

We structure the sustainability submodel around three dimensions aligned with post-conflict priorities: environmental continuity, social protection, and governance integrity. Environmental indicators can include the green energy ratio, air quality, and water security. Social indicators can include housing accessibility, healthcare coverage, and a distribution-sensitive equity measure. Governance indicators can include anti-corruption performance and the effectiveness of decentralisation. The indicator set is intentionally modular, allowing Kyiv to adjust it as reconstruction phases shift from emergency restoration to long-term modernization.

2.

Resilience sub model

$$R(t)=exp(-\lambda \int Shocks(t)dt)Radaptive(t),$$

(3)

Shock events are

$$Shocks(t)=\sum \left|\gamma k\cdot \delta (t-tk\right|),$$

(4)

where γ_k is severity of shock k.

Adaptive capacity is

$$R_{adaptive}(t)={\displaystyle \frac{1}{1+e^{-({\theta }_0+{\theta }_1\cdot Redundancy+{\theta }_2\cdot ResponseTime)}}},$$

(5)

In this framework, the “anxiety effect” is not a psychological-scale-to-engineering conversion. It is an operational decision and implementation delay penalty (e.g., procurement and coordination latency) calibrated from city governance and service-delivery delay proxies. The 15% value used below is an illustrative placeholder for a moderate delay regime.

3.

BANI dynamics is presented in Table 4. For each BANI factor, the table gives the mathematical representation used to encode its effect in the modelling framework (e.g., thresholds, variance-linked risk, tipping dynamics, entropy-driven delays).

• Strategic decision variables

Let control vector include

u(t) = [u₁(t), u₂ (t),…,u_m(t)],

(6)

where u₁ is investment in renewable energy, u₂ is budget for critical infrastructure hardening, u₃ is refugee integration spending, u₄ is AI-enabled monitoring networks.

Constrains include:

-

Budget

∑u_i(t) ≤ GDP(t)⋅τ(t),

(7)

-

War-driven

u_i(t) ≥ u_i min(t),

(8)

Stochastic optimisation

$$\underset{u}{\mathit{max}}E\left|F\left(T\right)\right|-\kappa \cdot Var\left(F\right(T\left)\right),$$

(9)

where k is risk aversion parameter.

Robustness check is

$$F_{worst-case}={\displaystyle \frac{min}{ϵ\in \Xi }}F(T,u,ϵ)$$

(10)

To align the model with the temporal structure used in recovery and reconstruction needs assessments, we operationalize reconstruction as a three-phase process over the planning horizon, rather than treating it as a single undifferentiated period. Phase 1: Emergency Repair (0–6 months) prioritises the restoration of critical lifelines (energy, water, communications, and emergency mobility) and the continuity of essential services. Phase 2: Functional Recovery (6–24 months) focuses on network stabilisation, rehousing, service reopening, and restoring baseline municipal capacity. Phase 3: Resilience Enhancement (2–5 years) prioritises modernization and adaptation (e.g., decentralised energy, low-carbon upgrades, equity-oriented service access, and institutional learning).

In modelling terms, the phase structure is implemented by allowing phase-specific policy targets and parameter settings: the normalisation baselines and targets used to compute the components of F(t) are updated by phase; the policy weights (α, β) can be shifted (e.g., Phase 1 emphasises resilience and service continuity, Phase 3 increases the weight on sustainability and equity); and minimum-allocation constraints u_imin(t) and budget shares τ(t) can be specified differently across phases to reflect binding emergency requirements versus longer-term investment capacity.

The same temporal logic also clarifies technology sequencing: Phase 1 emphasises rapid damage and outage sensing (e.g., satellite-based damage assessment, lightweight IoT monitoring, low-bandwidth public information services) and triage-level psychosocial support delivery; Phase 2 expands to coordinated reconstruction planning and service mapping (e.g., “good-enough” digital-twin layers for prioritisation and scheduling, smart-grid pilots, and scaled referral pathways); Phase 3 supports full scenario-based optimisation and stress testing (e.g., mature digital twins for investment pathway comparison, early-warning thresholds for shocks, and integrated digital mental-health service ecosystems).

7.2. Implementation Protocol

• Data inputs
  • Satellite imagery (damage assessment)
  • UNDP resilience indicators
  • Energy grid vulnerability maps
• Calibration
  • Machine learning for λ, θ_i
  • Participatory modelling with local communities
• Outputs
  • Adaptive investment pathways
  • Stress-test reports for cities
  • Early-warning thresholds for BANI shocks

Figure 1 summarises the methods workflow from inputs to outputs, showing calibration, the stochastic objective and constraints, and the scenario analysis layer.

Ukrainian contextualization:

• War Effects. Explicit shock terms γ_k in the resilience submodel.
• Decentralisation. S_G(t) depends on local governance capacity.
• EU Integration. S(t) increases with alignment to EU Green Deal.
• Energy. S_E(t) penalises fossil fuel dependence (post-strike recovery).

This model offers a dynamic decision support tool for Ukrainian policymakers, striking a balance between immediate survival needs (resilience) and long-term sustainability in the face of extreme uncertainty.

Let us look at the numerical example, which applies the model to a hypothetical Ukrainian city (e.g., Lviv or Kharkiv) under post-war reconstruction, calibrated with realistic (simplified) data. We focus on a 5-year horizon (2025–2030) with BANI shocks (

Table 5. Parameter calibration (Baseline values).

Parameter	Value	Source/Assumption
Initial Sustainability S(0)	0.45	Post-war assessment (low due to infrastructure damage)
Initial Resilience R(0)	0.30	High vulnerability to energy/water disruptions
Shock Severity γk	0.15 (annual attack), 0.3 (major flood)	Based on 2020–2023 event frequency in Northern Ukraine
Discount rate ρ	0.10	High urgency of recovery
Policy weights α, β	0.6 (sustainability), 0.4 (resilience)	Prioritising green reconstruction (EU alignment)
Budget τ(t)⋅GDP	$100 M/year	Estimated post-war GDP for Chernihiv Oblast + international aid

). Baseline inputs used in the illustrative run, including initial conditions, shock parameters, discounting, policy weights, and budget scale, with brief source or assumption.

Step 0. Empirical data assembly for Kyiv (spatial, energy, and social inputs).

Although the numerical example below is illustrative, the operating model is designed to be parameterised with empirical inputs at citywide and district scales. For Kyiv, the data package can be constructed as follows:

(a)

Spatial/service-access layer: district-level population and vulnerability profiles; geolocated critical facilities (hospitals, shelters, water pumping, schools); service catchments and travel-time-to-access metrics (e.g., share of residents within X minutes of shelter/primary care); green/blue space accessibility indicators and exposure proxies (noise/heat where available).

(b)

Energy and critical-infrastructure layer: utility continuity logs and restoration records (e.g., outage minutes and restoration times for electricity/heating/water), redundancy coverage for critical facilities (backup power, islanding readiness), distributed generation and storage capacity additions over time, and energy/emissions accounting for reconstructed assets.

(c)

Social and governance layer: displacement/return dynamics, housing availability and repair throughput, service utilisation metrics (healthcare, social support), and transparency/accountability indicators (procurement cycle time, completion rates, grievance reporting).

Step 1. Parameter calibration (baseline values).

Step 2. Simulated scenarios.

• Scenario A. “Baseline Recovery” (No Major Shocks)

Investments (u(t)).

• u1. 20% to renewables (solar microgrids)
• u2. 30% to infrastructure (water pipes, roads)
• u3. 25% to social housing
• u4. 15% to AI monitoring (e.g., air quality sensors)

Yearly Dynamics.

F(t) = 0.45 + 0.1⋅t (linear improvement from repairs) = 0.3 + 0.05⋅t (redundancy from microgrids) = 0 (no new events) = 0.6⋅S(t) + 0.4⋅R(t)

Result by 2030 (t = 5).

F(5) = 0.6⋅0.95 + 0.4⋅0.55 = 0.79 (Strong recovery)

• Scenario B. “BANI Crisis” (Shocks + Anxiety)

• Year 2 (2027). Major flood (γ2 = 0.3) damages 30% of the infrastructure.
• Year 4 (2029). Energy grid cyberattack (γ4 = 0.2).
• Anxiety Effect (illustrative). Decision and implementation delays reduce R_adaptive by 15% (placeholder for a moderate delay regime; calibrated in practice from procurement/coordination latency proxies as described in Section 7.1).

Resilience Submodel.

R(t) = exp(−0.3 − 0.2)⋅(0.55⋅0.85) = 0.30 (Collapse)

Sustainability Submodel.

S(t) = 0.45 + 0.1t − 0.3⋅δ(t − 2) − 0.1⋅δ(t − 4) = 0.55 (Partial recovery)

Result by 2030.

F(5) = 0.6⋅0.55 + 0.4⋅0.30 = 0.45 (Stagnation)

Step 3. Policy Optimisation

Objective. Adjust u(t) to maximise E[F(5)] under shock risk.

Optimal Allocation (from stochastic programming) (

Table 6. Optimal allocation.

Investment	Baseline (%)	Adjusted for BANI (%)	Rationale
Renewables (u1)	20	30	Reduce energy brittleness
Infrastructure (u2)	30	25	Shift funds to redundancy
Social (u3)	25	20	Temporary reduction
AI Monitoring (u4)	15	25	Early warning for shocks

).

Table 6 summarises an illustrative reallocation that increases investment in distributed renewables and AI-enabled monitoring, while moderately reducing the shares of conventional infrastructure and social programmes. The logic is to lower energy brittleness and improve early warning and coordination under compound risk. In the example, the adjusted portfolio yields F_optimized(5) = 0.65, indicating a mid-range improvement toward defined targets across service continuity, recovery speed, emissions, and equity. This result should be interpreted as a demonstration of the framework’s use rather than a definitive prescription without local calibration.

Key Takeaways for Ukraine:

• Renewables + Redundancy. High u₁ and u₄ mitigate energy shocks (critical after Russian strikes).
• Adaptive Budgeting. Dynamic reallocation (e.g., reduce u₃ during crises) improves outcomes.
• Anxiety Tax. BANI effects reduce resilience by ~15%—Invest in transparent governance (SG).

To make the proposed operating model genuinely workable in Kyiv, several enabling institutional and capacity conditions are required. First, the city needs a clearly mandated lead recovery delivery function that can coordinate municipal departments, utilities, and external partners, run short feedback loops, and translate scenario outputs into prioritised work plans and budget reallocations. Second, Kyiv requires agile procurement and contracting capacity so that iterative upgrades in energy, mobility, and municipal services can proceed without excessive decision delays. Third, the digital layer relies on a minimum data governance backbone, including interoperable registries, shared data standards, and APIs, as well as a transparent indicator or weight-setting process that involves municipal agencies, sector operators, and civil society. Fourth, because the model relies on connected platforms and critical infrastructure, Kyiv needs operational cybersecurity capacity embedded as a baseline requirement rather than an add-on.

8. Discussion

This paper operationalizes the BANI lens for wartime urban sustainability through the case of Kyiv. The analysis shows that resilience under sustained attack is socio-technical by necessity: digital tools can accelerate situational awareness and service coordination, but their impact depends on adaptive municipal decision-making, community capacity, and public trust.

The modelling study provides a compact way to connect policy levers to expected five-year performance under uncertainty. In the “no major shocks” pathway, the system reaches F(5) = 0.79, labelled “strong recovery”, reflecting high sustainability performance combined with moderate resilience (F(5) = 0.6·0.95 + 0.4·0.55). In the “BANI crisis” pathway, compound shocks and decision delays reduce performance to F(5) = 0.45, primarily driven by a collapse in the resilience term (R(t) = 0.30), even when sustainability remains partial (S(t) = 0.55). This highlights a policy-relevant asymmetry: under compound risk, marginal improvements in adaptive capacity (redundancy and response time) have a disproportionately large effect on the overall outcome, whereas sustainability gains alone do not prevent severe degradation when resilience fails. The model also formalises an anxiety tax: an illustrative decision and implementation-delay penalty (η = 0.15 in the worked example) materially worsens outcomes, indicating that investments in transparent governance, crisis communication, and rapid procurement are not auxiliary but performance-critical.

Under explicit uncertainty, the illustrative optimisation yields F_optimized(5) = 0.65 by shifting the portfolio toward distributed renewables and AI-enabled monitoring (Table 6). For policy, this should be read as a quantitative justification for prioritising energy brittleness reduction (distributed generation, storage, and islanding readiness) and early-warning and coordination capacity, while maintaining baseline social protection to avoid equity backsliding. The score remains below the “strong recovery” pathway, reinforcing that the optimisation improves robustness but does not eliminate structural exposure without local calibration and sustained institutional capacity.

Kyiv’s rapid retooling of citizen-facing services for air-raid alerts, shelter navigation, and outage communication demonstrates how governance and technology must evolve together in response to shock. Building on these observations, we distil four cross-cutting insights and outline implications for policy design and future empirical work.

Across brittle, anxious, and opaque conditions, Kyiv’s most effective responses emerged from a hybrid model that combined civic improvisation, municipal agility, and selective use of data and AI. The rapid iteration of the municipal app for air-raid alerts, shelter navigation, and outage updates demonstrates how digital services can expand institutional reach during periods of acute uncertainty. Similar patterns in national service continuity efforts suggest that wartime governance in Ukraine has relied on technology to enhance the speed and scale of human decision-making, rather than replace it. This reading is consistent with evidence of adaptive, networked arrangements between digital service teams and frontline institutions that sustained delivery under duress [48].

Evidence from Kyiv Digital and Diia suggests that citizen-facing platforms have become key nodes for information, transactions, and risk communication, helping to reduce anxiety and restore interpretability in rapidly changing conditions [49]. At the same time, grassroots initiatives in the early months of the full-scale invasion functioned as social infrastructure, absorbing shocks and stabilising routines. This pattern aligns with agency-centric resilience research, which highlights the role of civil society, local firms, and municipal actors in sustaining communities during crises [50]. In practical terms, resilience planning should treat social infrastructure as a design requirement alongside physical reconstruction, funding spaces and programmes that cultivate participation, mutual aid, and daily cohesion [51]. The mental health evidence similarly underscores that access to quality green space and walkable public realms is associated with lower anxiety and depression burdens, particularly among vulnerable groups [52]. This priority is especially consequential in an “anxious city” [53].

AI, IoT, and urban digital twins are most valuable in Kyiv, where system behaviour is non-linear and operational data is dense. They can surface patterns, enable scenario testing, and support allocation decisions under scarcity. The most immediate applications include rapid damage assessment, reconstruction logistics, energy balancing, and selected public health operations. The evidence that urban digital twins are moving beyond pilots toward city-scale and citizen-centric uses reinforces the feasibility of this direction [54]. However, the expansion of connected platforms also enlarges the cyber-physical attack surface [55]. For Kyiv and similar contexts, smart-grid, OT/IoT, and microgrid designs require secure-by-design architectures, segmented control, and continuous monitoring to prevent cascading failures. This is aligned with recent smart-grid/OT security syntheses that emphasise embedding cyber-resilience as a baseline requirement for energy and municipal platforms [56].

Urban brittleness in Ukraine is also material: over-optimised and import-dependent supply chains magnify exposure to disruption. Kyiv’s reconstruction can mitigate this risk by closing material loops through rubble recycling, reuse-oriented design, and localised supply chains. This framing positions circular economy measures as resilience instruments, not only environmental upgrades. Recent Ukrainian campaigns and technical studies, which suggest feasible pathways for scaled debris valorisation, support this direction, highlighting potential gains in cost, carbon reduction, and rebuilding speed [57].

First, adaptive governance should be institutionalised rather than treated as a temporary crisis mode. Short feedback loops, iterative policy updates, and cross-functional teams can be formalised through agile procurement rules, transparent service-level dashboards, and regulated experimentation in critical domains such as energy, mobility, and health. Second, social resilience should be funded and managed as core infrastructure. This implies earmarked investment for community hubs, green/blue networks, and co-created spaces that improve daily well-being and provide surge capacity during shocks. It also supports joint deployment of digital mental health services through trusted public platforms when in-person care is constrained. Third, cyber-resilience must be embedded in the baseline design of every connected project. For energy systems and city platforms, this includes OT security requirements, continuous monitoring, and routine cyber-physical exercises that test cascading-failure scenarios. Fourth, circular reconstruction should be scaled through aligned national guidance and financing that enable rubble processing, secondary material markets, and reuse-oriented design standards. At the municipal level, pilots of “materials passports” for major projects can improve traceability and support asset redeployment over time. Public digital platforms should also be positioned as trusted infrastructure for risk communication. City-service apps can integrate verified alerts, outage maps, and recovery workflows to increase reach, maintain trust, and counter misinformation, consistent with WHO/Europe recommendations on responsible AI-enabled risk communication.

These findings suggest a more grounded understanding of resilience under BANI: robust recovery depends on the deliberate integration of human agency, institutional adaptability, and secure, data-driven systems. Kyiv’s experience indicates that this integration is feasible and scalable when social infrastructure, cybersecurity, and circularity are treated as foundational design constraints.

9. Conclusions

This paper examines urban sustainability in Kyiv under BANI conditions and proposes an operational framework linking adaptive governance, secure data-driven systems (AI/IoT/urban digital twins), and circular reconstruction. The Kyiv case indicates that risk reduction and service continuity depend on coordinated socio-technical solutions that combine community capacity with cyber-resilient infrastructure.

The study’s contribution is to shift the BANI lens from diagnosis to design through city-scale governance, technology, and circularity measures. Social infrastructure and equity are positioned as primary constraints, while cybersecurity is viewed as a baseline requirement for digital and energy initiatives, rather than an auxiliary safeguard.

The study provides actionable steps for Kyiv and comparable cities, including institutionalising agile governance and procurement, utilising citizen-facing platforms for risk communication and service delivery, adopting secure-by-design standards for energy and municipal systems, and investing in community hubs and green/blue networks, as well as digital mental health support. In parallel, debris-to-materials programmes can be scaled through the use of instruments such as materials passports.

Key research priorities include causal evaluations of hybrid (community + digital) interventions, cyber-physical stress testing of interconnected platforms and microgrids, and cost–benefit analyses of circular reconstruction pathways. Longitudinal studies should track equity, mental health, and trust outcomes and examine data governance models that can sustain public adoption of digital services in high-uncertainty environments.

Our illustrative scenarios help quantitatively anchor the argument. In the absence of major shocks, the system reaches F(5) = 0.79 by the end of the five-year horizon. Under a BANI crisis pathway with shocks and decision delays, performance declines to F(5) = 0.45. When uncertainty is explicitly modelled, the optimised portfolio (30% renewables, 25% core infrastructure, 20% social programmes, 25% AI-enabled monitoring) achieves F(5) = 0.65, suggesting a feasible balance between recovery and robustness. The limitations are typical of wartime research: the model is applied to a hypothetical city calibrated with realistic (simplified) inputs, relies on structured data sources (satellite damage imagery, UNDP indicators, energy-grid vulnerability maps), and is validated through stochastic scenarios and a worst-case check rather than ex-post measurement.

Unlike conventional resilience frameworks that primarily catalogue hazards or prescribe static robustness measures, our BANI-oriented approach couples governance, digital capacity, and circular recovery levers into an adaptive operating model that supports continuous re-prioritisation in response to non-linear shocks and contested information environments.

The proposed framework is intended as an operating model for structuring decisions under BANI conditions. However, its numerical outputs should be interpreted with caution. First, the present application is illustrative: it is calibrated with simplified, realistic inputs and scenario parameters rather than full administrative datasets. It therefore does not constitute an ex-post validated forecast for Kyiv. Second, several elements require local governance and data calibration, including the specification of baselines/targets for each indicator, stakeholder-derived weights, and shock-severity assumptions. Therefore, the direct transfer of parameter values to other cities (or non-war contexts) is not appropriate without re-elicitation and recalibration. Third, validation in the current manuscript relies on stochastic scenarios and a worst-case check, not measured outcome tracking. As data access improves, external validation should compare model outputs against observed restoration times, service continuity logs, and distributional equity metrics.

Future research should therefore prioritise comparative replication across Ukrainian cities to test robustness under different infrastructure configurations and governance capacities; indicator-level empirical estimation and publication of a transparent “data dictionary” so readers can reproduce the composite scores; causal evaluation of hybrid interventions and cyber-resilient microgrid and digital-twin deployments; and extension to longer horizons and iterative re-optimisation, where early-warning thresholds and stress-test reporting become part of continuous monitoring within adaptive governance.