Digital Engineering for Future Smart Cities

1. Introduction

Cities worldwide are confronting major challenges associated with rapid urbanization, climate change, resource constraints, aging infrastructure, and the transition toward decarbonized, sustainable economies. At the same time, advances in digital technologies are creating unprecedented opportunities to transform how urban systems are planned, operated, maintained, and increasingly automated. Digital engineering, encompassing aspects such as the internet of things (IoT), artificial intelligence (AI), machine learning, digital twins, advanced sensing, cyber–physical systems, edge computing, intelligent energy management, robotics, and data-driven decision-making, is increasingly a key enabler of achieving sustainable and resilient smart cities [1,2,3,4].

The smart city paradigm has considerably evolved since the mid-2010s. Early smart city initiatives primarily focused on information and communication technologies and infrastructure; contemporary approaches increasingly emphasize sustainability, energy efficiency, resilience, interoperability, autonomy, and citizen-centered outcomes. Digital engineering provides the mechanisms through which these objectives can be achieved, enabling diverse infrastructure—including energy systems, buildings, transportation networks, communications platforms, and urban services—to become increasingly interconnected and intelligently coordinated [1,2]. More recent developments in urban digital twins illustrate this progression, providing dynamic representations through which physical infrastructure, sensing, simulation, AI, and operational decision-making can be integrated [5].

This Special Issue, entitled Digital Engineering for Future Smart Cities, was launched to showcase emerging research and innovation at the intersection of digital technologies, engineering systems, energy and utilities, and urban sustainability. The original scope encompassed technologies including the IoT, AI and machine learning, robotics, and domotics, alongside smart buildings and infrastructure, smart energy, transport decarbonization, eMobility, and smart logistics. The resulting contributions reflect the inherently interdisciplinary nature of smart city development and span urban energy systems, smart buildings, nonintrusive load monitoring, demand-side flexibility, reinforcement learning, mobile edge computing, advanced semiconductor technologies, privacy-preserving digital infrastructure, and urban-scale energy modeling.

A total of eighteen papers were received for inclusion in this Special Issue, and nine research articles were selected and published following peer review and appropriate revisions (alongside this Editorial). These contributions (listed in List of Contributions) comprise seven original research articles and two systematic state-of-the-art review articles. Collectively, the studies provide a snapshot the breadth of contemporary digital engineering research and illustrate how advanced computational methods, intelligent control systems, low-power computing technologies, and digital infrastructure can support the development of cleaner, smarter, and more sustainable cities.

2. Summary of Contributions

This Special Issue includes two review papers that provide broad perspectives on the future development of smart cities and urban energy systems. Drăgan et al. [Contribution 1] present The Smart City from the Energy Perspective, a comprehensive review examining smart cities through a system-of-systems lens. The authors emphasize the growing interdependencies among digitalization, energy infrastructure, transportation systems, data management, and advanced communications technologies. Attention focuses on renewable energy integration, intelligent energy management systems, the transition from 5G toward 6G communications, cybersecurity, standardization, data governance, and participatory governance. The study reinforces the need to regard the smart city not as a collection of independent technologies but as an interconnected sociotechnical system whose constituent infrastructure must increasingly operate cooperatively.

Complementing this contribution, Anselmo and Boccardo [Contribution 2] reviewed the state of the art in urban building energy modeling (UBEM) using geographic information system (GIS) and remote sensing technologies. Their analysis demonstrates how geospatial technologies can address one of the principal barriers to scalable urban energy modeling: the acquisition and integration of sufficiently detailed input data. The authors identify neighborhood-scale modeling as particularly prominent, with building geometry constituting the dominant model input, and remote sensing is increasingly being employed to determine geometric, nongeometric, and climatic parameters. The authors further highlight opportunities to integrate increasingly automated UBEM pipelines within urban digital twins, strengthening the connection between spatial intelligence, energy modeling, and city-scale decarbonization.

Two contributions particularly focus on smart buildings and residential energy intelligence. Alrashidi et al. [Contribution 3] developed a smart building energy and water management framework based on spiking neural networks (SNNs). The event-driven approach integrates adaptive control, occupancy-related information, anomaly detection, and edge intelligence within a computationally efficient framework. By exploiting the low-power characteristics of neuromorphic computation, the study demonstrates the potential of SNN-based approaches for real-time building automation in which sensing, anomaly identification, and control increasingly migrate toward distributed edge devices.

Abu Sbeitan et al. [Contribution 4] addressed appliance identification through a novel nonintrusive load monitoring (NILM) methodology. Their approach combines phase-space reconstruction with two-dimensional Fourier descriptors to characterize the nonlinear geometric properties of appliance current demand waveforms. The results of evaluation using the COOLL dataset demonstrated the effectiveness of the resulting low-dimensional feature representation for supervised appliance classification. Such techniques can provide increasingly detailed visibility of energy consumption without requiring dedicated sensing for individual appliances, supporting improved predictions and analytics for future applications in building energy management, demand response, and intelligent load coordination.

In a related study at the larger energy system level, Lee et al. [Contribution 5] investigated the application of deep reinforcement learning to multiunit combined heat and power (CHP) operation under demand uncertainty and progressive equipment degradation. Their proximal policy optimization (PPO)-based framework jointly addresses unit commitment, coupled heat and power dispatch, and preventive maintenance. By transferring a large part of the computational burden to offline training, the approach enables near-real-time operational decisions and demonstrates how reinforcement learning can support increasingly complex and uncertain energy infrastructures.

Lewicki et al. [Contribution 6] explored an unconventional but potentially main source of demand-side flexibility: domestic freezers providing power-grid frequency regulation. A dynamic thermal–electrical model was used to investigate the aggregated freezer response to frequency deviations. For a simulated population of 100,000 80 W deep freezers, approximately 3.2 MW of instantaneous downregulation was available under a 40% initial duty cycle. Extrapolation to the estimated eligible freezer population in Türkiye indicated potential headroom of approximately 150–225 MW. The authors also identified practical challenges, including compressor synchronization and postdisturbance demand peaks, illustrating the considerable potential and control challenges associated with coordinating large populations of small distributed loads. In a world increasingly requiring large amounts of cooling for, e.g., buildings and data centers, identifying flexibility potential and practical challenges, this is an important area to investigate. Along with [A3–A5], these contributions imply an ongoing, critical area of research is urban energy management at multiple levels using aspects of control and AI.

Digital infrastructure and computational sustainability are represented by the contribution of Chaudhry et al. [Contribution 7], who developed an incentive-based energy-efficient workload scheduling framework for mobile edge computing. The approach combines blockchain-based incentives with an enhanced whale optimization algorithm to address the competing requirements of energy consumption, task deadlines, device availability, and economic reward. The results indicate reductions in deadline violations alongside increases in net income and device availability. The study is particularly relevant as IoT and AI applications increasingly transfer computation from centralized cloud infrastructure toward distributed edge platforms located throughout the urban environment. This again provides useful results given the increasing energy and cooling requirements for large-scale AI adoption.

At the underlying hardware level, Devi et al. [Contribution 8] investigated low-power hetero-dielectric gate-all-around MOSFET architectures as potential enablers of sustainable embedded and smart city technologies. Device and inverter architectures were developed using TCAD simulation, demonstrating reductions in short-channel effects and increases in inverter noise margins relative to values reported in the literature. Although operating at a fundamentally different abstraction level from the other contributions, the study highlights an important consideration: increasingly pervasive urban intelligence ultimately depends on physical computing hardware, and increases in transistor-level efficiency can contribute to reducing the cumulative energy requirements of large populations of embedded devices. In considering the study along with [A6, A7], wider ideas start to emerge around hardware–software codesign and operations for energy-aware pervasive computing and AI infrastructures in smart cities.

Finally, Tereñes et al. [Contribution 9] addressed the increasingly important issue of trust and privacy within digital energy ecosystems and smart contracts through Privacy-Preserving Verification of Energy Performance Contracts: A Zero-Knowledge Proof Approach Within the FORTESIE Framework. Their framework applies zero-knowledge proof techniques to enable the verification of energy performance contract compliance without requiring the disclosure of the underlying sensitive operational data. This addresses fundamental tension in future smart cities: the effective optimization of energy, utilities, and related resources increasingly depends on data sharing among multiple stakeholders, whereas commercial confidentiality, privacy, cybersecurity, and regulatory compliance simultaneously require stronger control over those data. Privacy-preserving computation therefore represents an important layer enabling trustworthy digital urban infrastructure.

3. Synthesis and Research Directions in Digital Engineering for Sustainable Smart Cities

Despite spanning markedly different scales and research domains from semiconductor devices and household appliances to buildings and city-scale infrastructure, the contributions to this Special Issue reveal several common themes that collectively characterize the evolution of digital engineering for sustainable smart cities.

The first is the growing role of AI. AI-driven approaches appear throughout this Special Issue in the form of reinforcement learning, spiking neural networks, machine learning classification, signal interpretation, and intelligent optimization. These technologies are progressively shifting engineered urban systems from conventional monitoring and reactive control toward predictive and adaptive operation. The next stage of this transition will likely involve systems capable of not only interpreting their environment but also coordinating decisions across multiple physical and digital domains.

The second theme is the increasing importance of distributed intelligence. Several contributions move computation and decision-making away from exclusively centralized architectures toward edge devices, individual buildings, household appliances, and distributed energy resources. Such decentralization can increase scalability, responsiveness, resilience, and privacy while reducing communication burdens and computational bottlenecks. The smart building, NILM, mobile edge computing, and domestic flexibility contributions particularly exemplify this trend. The decentralization of computing resources—if securely managed and appropriately implemented—also has the potential to ameliorate some of issues associated with the large resource footprints (cooling and electrical power) of larger AI datacenters.

Third, this Special Issue demonstrates the critical role of flexibility and adaptive control in future urban energy systems. Traditional energy infrastructure has been designed around comparatively predictable patterns of centralized generation and demand. Future cities must instead accommodate renewable generation, electrified transport, distributed storage, flexible demand, prosumers, and increasingly dynamic patterns of consumption. The CHP scheduling, smart building, and freezer aggregation studies illustrate three different scales at which digital engineering can support and unlock resource flexibility. Future systems will increasingly need to coordinate such resources across buildings, neighborhoods, microgrids, transport systems, and wider electricity networks.

A fourth theme concerns the importance of data integration and interoperability. Effective digital engineering depends on accurate, timely, contextualized, and interpretable information from heterogeneous sources. The GIS and remote sensing review demonstrates how urban-scale spatial datasets can support energy planning and modeling, whereas the NILM contribution illustrates how advanced signal processing and machine learning methods can transform aggregate measurements into appliance-level information. Urban digital twins provide a natural architectural progression for gathering such heterogeneous information, combining IoT sensing, AI, simulation, and cyber–physical infrastructure within dynamic representations of urban systems [5,6].

The contributions also indicate the need to consider computational sustainability alongside urban sustainability. Digital technologies are becoming ubiquitous in cities, from sensors and embedded controllers to communications networks, edge platforms, AI models, datacenters, and cloud infrastructures. Although these systems can enable substantial resource and energy savings, their own energy and material requirements cannot be ignored. The SNN, edge computing, and semiconductor contributions demonstrate that sustainability needs to be considered across the complete digital stack—from transistor and device architectures through embedded intelligence and communications to large-scale AI and urban computing infrastructure.

A key emerging research direction extending beyond the contributions to this Special Issue is the increasing role of robotics and autonomous physical systems for directly implementing actions derived from sensor and AI-based insights. The original scope of this Special Issue explicitly identified robotics and domotics alongside AI and the IoT as technologies enabling future smart cities. Although the published papers principally address the sensing, computation, optimization, energy, and digital infrastructure layers of the smart city ecosystem, a logical next step is more directly connecting this intelligence to physical action through building management systems (BMSs); traffic management systems (TMSs) and telematics, industrial DCS and SCADA platforms; energy trading systems (ETSs; and energy management systems (EMSs), alongside traditional, mobile, and humanoid robotics [7,8,9].

AI-enabled robotic systems are already progressing from deterministic automation toward adaptive systems incorporating multimodal perception, learning-enabled control using digital twins, and featuring increasingly sophisticated human–robot interactions. Within future cities, these capabilities have potential applications in automated inspection and the maintenance of energy and transport infrastructure; robotic building and facilities services; waste collection, sorting, and circular economy processes; autonomous logistics; environmental monitoring; emergency response; and mobility. Recent advances in adaptive control similarly demonstrate the continuing need for the robust, computationally efficient real-time control for next-generation robotic systems, including those operating under uncertainty in these related application areas [9]. The deployment of such precision control techniques is vital for future, sustainable smart city operations.

This progression suggests a critical conceptual transition. Earlier smart city systems were primarily designed to sense and communicate; the current systems increasingly analyze, predict, and optimize; future cyber–physical cities may be able to decide, coordinate, and physically act. Robotics, autonomous vehicles, drones, and other embodied intelligent systems provide this physical-action layer. Their integration with IoT infrastructure, edge AI, urban digital twins, smart buildings, and energy systems therefore represents a particularly important research frontier that is likely to have large impacts on energy management [8].

Such autonomy also introduces new challenges. Safety assurance, explainability, cybersecurity, human–robot interaction, energy efficiency, privacy, interoperability, and responsibility for autonomous decisions will become more important as AI migrates from advisory applications into systems capable of directly modifying the physical urban environment. Human-centered design must consequently remain fundamental. The objective should not simply be increasing technological autonomy but deploying autonomous systems that demonstrably increase sustainability, resilience, safety, accessibility, productivity, or quality of life.

This Special Issue additionally highlights the importance of trustworthy digitalization. As sensing and autonomous decision-making become more pervasive, urban systems will depend on information exchanged among residents, infrastructure operators, utilities, commercial organizations, and public authorities. The zero-knowledge proof contribution illustrates one approach through which useful verification can be achieved while preserving confidentiality. Future research should further investigate privacy-preserving AI, federated and distributed learning, cyber-secure digital twins, explainable autonomous decision-making, and architectures that establish trust without requiring unrestricted centralization of sensitive urban data.

Finally, smart cities should continue to be viewed as complex sociotechnical systems rather than purely technological constructs. Their energy, transportation, communications, environmental, industrial, building, and human systems are deeply interconnected. Future research should therefore consider cross-domain interactions, for example, buildings simultaneously operating as energy resources, workplaces, data-generating cyber–physical environments, and operating spaces for autonomous systems; or electric and autonomous mobility simultaneously interacting with transportation networks, electricity grids, communications infrastructure, and urban logistics.

Taking these studies together, several priority research directions emerge:

• Scalable and interoperable urban digital twins capable of progressing from monitoring to prediction/forecasting, and closed-loop operation;
• Integrating AI with multienergy systems, buildings, transport, and distributed flexibility;
• Energy-efficient AI extending from semiconductor hardware to edge and cloud infrastructures;
• Trustworthy and privacy-preserving data exchange across multistakeholder urban ecosystems;
• AI-enabled robotics, autonomous vehicles, and multirobot systems for sustainable urban services;
• Tighter integration of robotics and autonomous systems with digital twins and edge intelligence;
• Enhanced geospatial and remote sensing intelligence for dynamic urban modeling;
• Cyber-secure and explainable autonomous decision-making architectures;
• Human-centered approaches to AI, automation, and robotics;
• Real-world demonstrators capable of validating these technologies beyond simulation and isolated pilot studies.

Collectively, these directions indicate a transition toward highly connected, intelligent, adaptive, and increasingly autonomous urban environments in which digital engineering provides the link among information, computation, energy, utilities, transportation, infrastructure, and physical action.

4. Conclusions

The papers published in this Special Issue showcase the diversity and maturity of contemporary research in digital engineering for sustainable and smart cities. From smart buildings and appliance-level energy intelligence to urban energy systems, distributed flexibility, edge computing, advanced optimization, privacy-preserving computation, geospatial modeling, and semiconductor technologies, the contributions demonstrate the transformative potential of digital engineering across multiple levels of urban infrastructure.

A common message emerging from this Special Issue is that future smart cities will depend on not any single technology but the integration of intelligent software, energy-efficient hardware, cyber–physical systems, energy infrastructure, communication networks, sensing technologies, and trustworthy data-driven operational frameworks. Increasingly, this integration will extend beyond sensing and optimization toward autonomous physical action through robotics and other embodied intelligent systems.

Increasing autonomy must simultaneously remain aligned with the fundamental objectives of sustainable urban development. Digitalization is valuable not as an end in itself but where it contributes to lower resource consumption and emissions, increased resilience, safer infrastructure, better public services, increased accessibility, and improved quality of life. Achieving these objectives will require continued collaboration among engineering, computer science, energy, the built environment, transportation, robotics, social science, governance, and policy.

The breadth of the contributions received and the research opportunities identified through this Special Issue demonstrate that digital engineering for smart cities remains a rapidly developing field. To continue this discussion and provide a venue for emerging work, a related edition of this Special Issue, Digital Engineering for Smart and Sustainable Cities, has subsequently been launched in Energies (https://www.mdpi.com/journal/energies/special_issues/E75LAI01W8 [Accessed 11 September 2026]). This related edition provides an opportunity to extend the themes established here while encouraging research on emerging areas including AI-enabled and autonomous urban systems, robotics, urban digital twins, intelligent infrastructure, smart energy, sustainable mobility and logistics, trustworthy digitalization, and human-centered approaches to the design and operation of future cities.

The editors hope that the research presented in this Special Issue will stimulate further interdisciplinary collaboration and encourage the continued development, demonstration, and deployment of digital engineering solutions capable of supporting the next generation of sustainable, resilient, intelligent, and human-centered cities.