Advancements Toward a Standard System for Intelligent Operation and Maintenance of Buildings and Municipal Facilities

Abstract

The building and municipal facility sectors in many countries are shifting from rapid construction to a balanced focus on construction and operation & maintenance (O&M). However, O&M practices remain largely manual, with poor digital integration, fragmented data management, and inconsistent performance standards. The absence of a unified theoretical and standardization framework for intelligent O&M represents a critical research and practice gap. To address this, this paper proposes a comprehensive framework for intelligent O&M standards, grounded in operations management theory and supported by extensive research. The framework is structured across three dimensions: (a) functional services, including perception, data fusion, decision-making, and disaster prevention; (b) system hierarchy, ranging from perception layer and algorithm layer to human–computer interaction layer; and (c) intelligence characteristics, spanning presentation and monitoring to autonomous maintenance. In addition, existing standards and representative applications are reviewed to provide valuable references for the future development of intelligent O&M standard systems.

1. Introduction

Buildings and municipal facilities (BMFs) generally encompass power grids, communications, transportation networks, and various municipal services. Their availability is vital in improving the quality of life and public well-being, driving productivity, economic growth, and poverty reduction. In China, rapid urbanization has resulted in the addition of over 1 billion square meters of new housing annually. The country’s underground networks span over 1 million km of water, 700,000 km of gas, and 200,000 km of heating pipelines. By 2024, highways reached 5.44 million km, with 1.08 million bridges, including 17.4% that were large or extra-large [1,2]. However, constraints such as limited population growth and resource availability have slowed the pace of new construction. Therefore, the infrastructure model is transitioning from a “construction-focused” approach to integrating construction and long-term management worldwide. Public demand for better BMF service quality has increased, but their performance has declined due to aging and environmental pressures [3,4,5]. Traditional operation and maintenance (O&M) modes lack long-term planning, underutilize data, and suffer from inefficiency, making it challenging to meet the demands of the extensive and numerous BMF for O&M. More seriously, the BMF O&M industry currently faces severe aging of practitioners and a labor shortage. To address these problems, a new O&M management model is urgently needed to allocate limited resources reasonably and improve operational efficiency [6].

Similar to other fields, intelligent O&M refers to a management approach that leverages next-generation information and innovative technologies, such as the Internet of Things (IoT), robotics, big data, and Artificial Intelligence (AI), to comprehensively monitor, analyze, and manage the service performance of BMFs. As illustrated in Figure 1, its core lies in real-time monitoring and integrated analysis of diverse O&M data, combined with periodic inspections and routine patrol information. This enables timely detection of abnormal conditions, fault risks, and performance degradation, supports early safety warnings, and provides a scientific basis for maintenance management and decision-making. A standard system is an integrated and systematic whole formed by a set of interrelated standards within a defined scope. It represents the framework for formulating and implementing standards, guiding standardization activities, and ensuring the practical application of these standards. The standard system is essential for orderly industry development, as it is both the top-level design and the foundational structure of standardization. Although focused on O&M, the framework also addresses design, construction, and especially end-of-life, enabling intelligent support for decommissioning, resource recovery, and safety.

Recent research highlights several challenges in intelligent O&M for BMFs: (a) existing standards primarily address design and construction, with limited coverage of intelligent O&M processes such as data collection, transmission, database management, and integration of visualization and data mining tools; (b) incompatibility among vendor products hampers system interoperability, leading to fragmented information, inefficiency, and increased costs; (c) many existing facilities lack the infrastructure to support intelligent O&M, as they were not designed with such systems in mind; and (d) retrofitting older utilities is hindered by the absence of unified standards, complicating planning, budgeting, and technical coordination. These issues reflect the lack of a comprehensive intelligent O&M standard system and a maturity evaluation model. The specific challenges associated with current intelligent O&M practices across different sectors of buildings and municipal facilities are further illustrated and summarized below.

Intelligent O&M modes have been explored across various sectors [7,8,9,10]. In mining, Wang proposed the concept of an intelligent mine and developed a technology architecture based on the deep integration of multiple systems, laying the foundation for intelligent O&M in mining. However, standardization in coal mines remains in its infancy, lacking top-level planning, leading to overlapping regulations, unclear standards, and repeated construction [11]. In the field of healthcare, Cai examined O&M services in large hospitals in China and built a hospital standard O&M service system based on ITSS (Information Technology Service Standard), demonstrating that ITSS frameworks can enhance hospital O&M management [12]. Xu reviewed intelligent hospital implementations, proposed construction paths, and highlighted the importance of standard research, information system restructuring, and quality control mechanisms [13]. In the financial sector, Fu et al. analyzed data center O&M from the perspective of data-driven development and data thinking [14]. At the same time, Yang classified the digital-to-intelligent transformation into three stages—digital O&M, intelligent O&M, and brilliant operation—based on AIOps practices [15]. In agriculture, Hu studied intelligent agricultural systems [16], and Yang et al. examined the development of intelligent agriculture in China, identifying technical gaps, insufficient standards, and a lack of standardized talent. They proposed top-level planning and a national intelligent agriculture standards system [17]. Other sectors, including wind power and hydropower, have also extensively researched intelligent O&M standardization [18,19].

For infrastructure systems, Zhu highlighted that underground space O&M suffers from the absence of a complete technical system for maintenance and renewal of underground BMFs, inadequate management of O&M information, and incomplete emergency rescue systems, emphasizing the urgent need for a digital and intelligent safety O&M system [20]. Wang et al. introduced Japan’s railway tunnel maintenance management system, automated inspection, and intelligent technologies, identifying challenges that hinder the development of high-quality O&M and increase costs. They proposed a predictive O&M system for Chinese tunnels to extend service life, based on structural performance and lifecycle cost [21]. Li et al. analyzed digital highways and their standard requirements, proposing a model and architecture for a digital highway standards system [22]. Liu reviewed China’s intelligent highway standards, exploring paths to align standards with practical applications and technological trends [23]. Jing et al. addressed the complexity and high costs of cross-sea cluster infrastructures, reviewing key intelligent O&M technologies for the Hong Kong–Zhuhai–Macao Bridge and summarizing nine critical engineering elements [24]. Jiao defined “intelligent utility tunnels” and proposed construction standards to guide tunnel design, construction, and management [25]. Yang et al. discussed key scientific issues in full-lifecycle intelligent infrastructure [26]. Wang et al. proposed a comprehensive safety monitoring platform for road, bridge, and tunnel infrastructure, demonstrating its effectiveness in reducing operational risks and optimizing maintenance resources [27]. More practical or pilot implementations of bridge O&M systems can be found in several well-known projects, including the Tsing Ma Bridge [28], Kap Shui Mun Bridge [29], Great Belt East Bridge [30], Akashi Kaikyo Bridge [31], and Commodore Barry Bridge [32], among others.

In the field of smart cities, Ren analyzed domestic and international intelligent city development, identified gaps and standardization needs, and proposed a hierarchical standards framework detailing various standards [33]. Liu provided a comprehensive overview of smart city standardization, proposing a scientific architecture aligned with development requirements and summarizing the research status of each standard [34]. Dai, using Fuzhou City as a case study, highlighted the urgent need for a new intelligent city standards system, proposed a top-level design spanning information, object, and application layers, and systematically outlined eight key areas, including information support, innovative infrastructure, and security standards [35]. To enhance the international relevance of the proposed framework, this study further considers cross-national experiences with O&M systems in smart city contexts across Asia and Europe. In Asian cities such as Singapore, Seoul, and Tokyo, O&M systems are characterized by centralized governance, high-frequency data collection, and strong governmental support for digital transformation, which enable the rapid deployment of integrated monitoring platforms [36,37,38]. In contrast, European cities such as Amsterdam, Copenhagen, and Barcelona emphasize open-data ecosystems, interoperability standards, and citizen-oriented service models, reflecting a more decentralized and participatory approach to infrastructure management. Such cross-national insights not only validate the general applicability of the proposed framework but also guide tailoring its implementation to different socio-technical contexts worldwide [39,40,41,42].

From the literature review above, current O&M systems for BMFs tend to focus only on specific functions, such as perception or performance prediction, without adequately emphasizing next-level intelligent decision-making. Others concentrate on data collection and transmission while neglecting algorithm development and human–computer interaction, resulting in a relatively low intelligence level of the O&M system. Therefore, it is evident that establishing a scientific and comprehensive standards system is essential for the development of intelligent O&M for BMFs [43]. Recently issued standards primarily focus on monitoring, testing, and maintenance, with limited attention to intelligent aspects. The definition and understanding of intelligent O&M remain unclear. Key challenges in developing intelligent O&M standards include the following: (a) inconsistent and ambiguous concepts and terminology; (b) the lack of data standards for collection, transmission, and integration, which limits lifecycle data utilization; (c) outdated technical standards, particularly for data analysis, intelligent control, and decision-making; (d) the absence of evaluation standards for system quality and efficiency; (e) overemphasis on traditional models without technical indicators for intelligent development; (f) fragmented and poorly coordinated standard-setting efforts, leading to overlap and conflict; and (g) many intelligent O&M standards remain immature, untested, or misaligned with practical needs.

Intelligent O&M of BMFs is a systematic endeavor requiring analysis grounded in established theoretical frameworks. The comprehensive micro-analysis method provides a fundamental approach for constructing such systems [44]. This method involves analyzing the component subsystems of the intelligent O&M system, performing conceptual analyses at both system and component levels, and characterizing microstates, macrostates, and component state distributions to build an integrated composite system. Based on operations management theory and the comprehensive micro-analysis method, this paper examines the intelligent O&M standards system through a literature review and industry investigation. The research approach is as follows: First, by reviewing various intelligent O&M systems, standard abstract features are extracted to construct a system architecture comprising functional services, system hierarchies, and intelligence characteristics. The objects and boundaries of intelligent O&M are defined, existing and missing standards are identified, and overlaps among standards are recognized. Next, standardization needs are analyzed, and logical relationships across architectural dimensions are integrated. Functional services and intelligence features are then mapped onto six system hierarchy levels: perception, transmission, data, algorithm, business, and human–machine interaction layers. Together with general and industry application standards, these form the intelligent O&M standards system architecture. Finally, the architecture is decomposed to establish a framework guiding standards development and implementation. Overall, this study aims to develop a comprehensive standardization framework for intelligent O&M of BMFs based on a multidimensional system analysis. The workflow is illustrated in Figure 2.

2. Architecture of Intelligent O&M Standard Systems

2.1. Objectives

This study focuses on developing a comprehensive intelligent O&M standard system that synthesizes relevant GB/T standards in China for the O&M of buildings and municipal facilities. This study analyzes the complete lifecycle technical requirements of intelligent O&M systems for BMF. Vertical correlations are examined through industry management and business process analysis, while horizontal correlations are assessed via technical classifications and cross-industry interactions. Specifically, the research combined a systematic literature review of 21 relevant studies and standards (published between 2015 and 2025) [13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,45] with an industry investigation that included expert interviews and case analyses of 10 representative smart city O&M systems in China, Japan, and Europe [33,34,35,36,37,38,39,40,41,42]. These sources were examined to identify core service functions, data structures, and system hierarchies. The analysis and decomposition followed four main steps: (1) identification of standard components from existing practices; (2) categorization by service function and system level; (3) cross-analysis to detect overlaps and gaps; and (4) integration into a coherent multi-level architecture. Based on this analysis, a multi-level technical standards system is established, comprising general intelligent O&M standards (A), key technology standards (B), and industry application standards (F). This framework is designed to improve industry practitioners’ understanding of intelligent O&M standardization, facilitate communication on infrastructure O&M standards across regions and sectors, and serve as a reference for implementing intelligent O&M in BMFs.

2.2. Basic Principles

Intelligent O&M of buildings and municipal facilities can be conceptualized as a complex mega-system composed of multiple interacting subsystems. For example, perception subsystems include sensors, imaging devices, and structural health monitoring instruments that collect real-time data; communication subsystems such as IoT networks, cloud platforms, and edge computing systems transmit and integrate data; decision-making subsystems leverage AI algorithms and digital twins to analyze information and support predictive maintenance; and service subsystems provide interfaces for operators, facility managers, and the public. Constructing an intelligent O&M standards system requires reflecting the internal relationships within each subsystem and the interactions among different subsystems, while accounting for both static and dynamic factors. The standards framework should follow a systematic, hierarchical principle, ensuring coherence, practical relevance across lifecycle stages, and adaptability for various infrastructure sectors. By grounding the framework in concrete examples, the standards system can serve as a practical theoretical guide, supporting consistent, interoperable, and effective implementation of intelligent O&M systems.

2.3. O&M Standard System Deconstruction

The intelligent O&M standard system architecture characterizes the activities, equipment, and features of BMFs from three dimensions: functional services, system hierarchy, and intelligence characteristics. This framework defines the standardization requirements, objectives, and scope of intelligent O&M, guiding the development of a comprehensive intelligent O&M standards system. In addition to technical and managerial dimensions, intelligent O&M also requires robust mechanisms for information security and risk management to ensure the safe handling of sensitive operational data, as shown by Dimension 1 of Functional services in Figure 3. These aspects form an essential foundation of the proposed standardization framework. The architecture of this system is illustrated in Figure 3.

2.3.1. Functional Services (Dimension 1)

Functional services are the core services the intelligent O&M system provides, aligned with its objectives. They include eight categories: intelligent perception, data fusion, status assessment, early warning and forecasting, thoughtful decision-making, intelligent control, disaster prevention, and emergency management.

1.

Intelligent perception: Employing sensing methods such as sound, light, electricity, magnetism, and heat combined with intelligent sensors, an Internet of Everything is formed. Data on functional performance, durability, safety, and energy consumption is collected through online monitoring, inspections, and routine patrols.

2.

Data fusion: Collected data are modeled and analyzed to understand changes in facility status and service functions caused by loads, environmental factors, or disasters.

3.

Status assessment: Real-time data, physical knowledge, and AI techniques (e.g., machine learning, large models like ChatGPT-4o and DeepSeek-R1) are used to build models for status recognition, evaluation, and risk identification, forming an “O&M brain” that is updated regularly or post disaster.

4.

Early warning and forecasting: Big data and machine learning algorithms process information to detect risks and issue timely warnings and forecasts.

5.

Intelligent decision-making: Status assessments, combined with knowledge bases and AI techniques, generate decision models to recommend maintenance and repair actions.

6.

Intelligent control: Robotics, VR/AR, drones, and other intelligent technologies detect and respond to maintenance needs and emergencies.

7.

Disaster prevention: “Intelligent monitoring + intelligent equipment” technologies mitigate significant hazards such as floods, freezing, earthquakes, mudslides, snow accumulation, fires, and rockfalls.

8.

Emergency management: A self-circulating closed-loop system coordinates pre-event warnings, event response through rapid resource scheduling, and post-event optimization for resilience and improved emergency management.

2.3.2. System Hierarchy (Dimension 2)

System hierarchy defines the organizational levels of intelligent O&M activities, comprising six layers: perception, transmission, data, algorithm, business, and human–computer interaction.

• Perception layer: Senses environmental and infrastructure changes, collecting O&M information.
• Transmission layer: Enables real-time interaction between physical and information spaces through wired/wireless media, cloud–edge–terminal networks, and high-capacity, low-latency transmission technologies.
• Data layer: Aggregates collected, generated, and fused data, forming the foundation for platform operations.
• Algorithm layer: The core driver of intelligent O&M is analyzing high-quality data to make decisions, generate insights, and control automated O&M operations.
• Business layer: Provides essential services—including limit alarms, status assessment, and emergency control—based on algorithm outputs, representing the system’s functional core.
• Human–computer interaction layer: Supports data visualization, digital twin models, AI interfaces, intelligent alerts, and automated work orders for various infrastructures and O&M scenarios.

2.3.3. Intelligence Characteristics (Dimension 3)

Intelligence characteristics classify intelligent O&M systems by maturity levels, divided into five stages: presentation display, monitoring and alarm, comprehensive fusion, future projection, and autonomous maintenance, as shown in Figure 4.

• Presentation and display (Level 1): Displays infrastructure service and maintenance information but cannot assess facility status, relying mainly on expert intervention. For example, bridge technicians may monitor vehicle operations using video and weighing systems, ensuring some structural safety; however, the system cannot assess structural integrity.
• Monitoring and alarm (Level 2): Presents real-time facility status and issues alerts when measurements exceed thresholds, prompting personnel intervention. For instance, sensors on a bridge collect data on deflection and stress, triggering alarms if abnormal readings are detected.
• Comprehensive fusion (Level 3): Integrates multisource data to reflect the facility’s real-time condition, laying the foundation for predictive analysis. In bridges, sensors and algorithms monitor vehicle, wind, temperature, and structural responses, while digital twins enable the real-time fusion of data and evaluation of service status.
• Future projection (Level 4): Uses models and AI algorithms to predict infrastructure performance over time, enabling proactive maintenance. Time-variant material models combined with dynamic fusion data forecast future bridge service states, transforming sudden issues into predictable events.
• Autonomous maintenance (Level 5): The system continuously adapts through long-term interaction with infrastructure, reconstructing digital twins, diagnosing, maintaining, and optimizing operations autonomously. Bridges benefit from dynamic perception, prediction, decision-making, and the automated deployment of maintenance robots, enabling sustainable, low-cost, high-quality intelligent O&M.

2.4. Framework of Standard System

The intelligent O&M standard system architecture for BMFs is mapped downward to form the overall standard system framework and its fundamental components. The architecture is illustrated in Figure 5. The framework primarily comprises six parts: General Standards (A), Key Technology Standards (B), Intelligent O&M Platform Standards (C), Intelligent O&M Evaluation Standards (D), Intelligent O&M Safety Standards (E), and Industry Application Standards (F). This structure clearly reflects the composition and interrelationships among the different components of the standard system.

Specifically, General Standards (A) include Terminology Standards (A1), Reference Architecture Standards (A2), Applicability Guidelines (A3), and O&M Indicator Systems (A4). These define intelligent O&M concepts, reference frameworks, applicable conditions, and multidimensional status indicators for various structures, providing foundational support for the standard system. Key Technology Standards (B) cover six layers: Perception (B1), Transmission (B2), Data (B3), Algorithm (B4), Business (B5), and Human–Computer Interaction (B6). They guide the research and implementation of critical digital intelligent O&M technologies, ensuring effectiveness and enabling collaborative development and modular interchangeability.

Intelligent O&M Platform Standards (C) comprise Technical Standards (C1) and Basic Standards (C2), which define the design, construction, acceptance, and technical requirements for developing intelligent O&M software platforms. Intelligent O&M Evaluation Standards (D) focus on maturity assessment and guide intelligent upgrades of O&M systems. Intelligent O&M Safety Standards (E) include Physical System Safety (E1), Functional Safety (E2), and Information Security (E3), standardizing personnel safety operations and secure management of information within the system. Industry Application Standards (F) account for technological differences across various domains, regulating intelligent O&M implementation for Roads (F1), Bridges (F2), Buildings (F3), Municipal Pipelines (F4), and other sectors. The detailed relationships among these components are illustrated in Figure 6.

3. Content of Intelligent O&M Standard System

3.1. General Standards

The contents of the four sections of general standards are illustrated in Figure 7.

3.1.1. Terminology

This section defines concepts and abbreviations related to intelligent O&M, facilitating user understanding and supporting the development of other standards. Key terms include core definitions, critical technologies, and concepts such as intelligent O&M for BMF, multi-source heterogeneous data, data fusion, intelligent early warning, intelligent identification, intelligent assessment, intelligent decision-making, O&M intelligent entities, connectivity, and intelligent O&M platforms.

3.1.2. Reference Architecture

Intelligent O&M is divided into two levels based on business processes: system-level and unit-level. System-level reference architecture standards govern the layering, overall architecture, and interrelationships between components, enabling users to understand the system design. Unit-level standards define reference architectures for subsystems, including the perception, transmission, data, algorithm, business, and human–computer interaction layers.

3.1.3. Application Guidelines

These guidelines standardize the applicability requirements of intelligent O&M, enabling users to determine whether and how to implement intelligent O&M. The guidelines encompass functional, performance, safety, and maintenance requirements.

3.1.4. O&M Indicator System

This section outlines multidimensional status indicators for various structures, encompassing safety, functionality, service life, energy consumption, and social factors. The indicator system defines key elements across the O&M business chain, including monitoring data, status assessment, performance evaluation, maintenance decision-making, and post-maintenance evaluation, thereby guiding perception, assessment, decision-making, and optimization activities.

3.2. Key Technology Standards

Key technology standards define and regulate the critical technical requirements for implementing intelligent O&M systems in BMFs. They are organized into six main categories: perception layer standards, transmission layer standards, data layer standards, algorithm layer standards, business layer standards, and human–computer interaction layer standards, as illustrated in Figure 8.

3.2.1. Perception Layer

The perception layer is responsible for sensing environmental conditions and changes in BMF, as well as collecting various O&M information, serving as a critical support for the intelligent O&M system. Relevant standards and key contents are illustrated in Figure 9 and include the following three aspects [46]:

• O&M perception devices and technology: These standards establish technical requirements for intelligent sensor performance, data output, installation testing, and protection during O&M information collection. They regulate the specifications and usage rules for sensors and detection equipment integrated into intelligent O&M systems, thereby enhancing the reliability, compatibility, and operability of perception devices.
• Perception device optimization and deployment: Sensors represent the primary cost component in constructing intelligent O&M systems and serve as the primary data source. Optimizing the spatial placement of sensors and perception networks is essential for a scientifically and efficiently built system. Standards should provide guidelines for monitoring point deployment, optimization calculation methods, typical structural sensor placement atlases, and strategies for cooperative perception deployment, ensuring practical and rational sensor layouts.
• Collaborative perception: These standards define requirements and interaction rules for the collaborative operation of multi-source heterogeneous devices (e.g., structural response sensors, environmental detectors, space–sky–ground–sea–vehicle platforms). They cover 3D perception networks for structural response, ecological and sudden-change detection, and spatial information acquisition systems, with clear identification and interconnection criteria to enable integrated collaboration and system modeling.

3.2.2. Transmission Layer

The transmission layer ensures real-time interaction between the physical and information spaces, serving as the backbone of intelligent O&M systems. It typically involves two primary types of transmission media (i.e., wired and wireless) and cloud–edge–end architectures that support efficient data exchange. Relevant standards and key contents are outlined in Figure 10 and address the following aspects [47,48]:

• Private network technology: These standards define information exchange and communication technologies across heterogeneous network platforms. By interconnecting sensing devices through different media, data transmission and information exchange are enabled. Key technologies include the Internet of Things (IoT), Wireless Sensor Networks (WSNs), and next-generation Internet technologies. For example, IoT technologies are regulated by international ISO 18000 security standards [49] and LAN IEEE 802 transmission protocol standards [50]. WSNs follow standards such as Zigbee [51] and ISA100 [52]. Next-generation Internet technologies, including IPv6 and 5G, offer enhanced connectivity, scalability, and transmission capacity.
• Transmission control technology: These standards regulate the formats, conversion, and control mechanisms that data must follow during transmission. They cover real-time access, multi-source aggregation, and trusted, controllable, and manageable network technologies. Widely adopted protocols include TCP/IP and the Real-Time Control Protocol (RTCP). Additionally, control technology standards promote interoperability by facilitating the conversion of diverse device control protocols, including network transmission protocols, control protocols, and device addressing schemes.
• Data transmission system architecture: Large-scale and complex infrastructure O&M systems must process massive and rapidly growing data streams. Edge computing is a critical technology for ensuring efficient data transmission and interaction. A well-structured cloud–edge–end collaborative architecture is required to achieve low-latency, high-reliability transmission. This includes standards for local data collection at end devices, transmission to the edge or cloud, and edge computing capabilities for real-time preprocessing and regional data management. Big data processing, intelligent analysis, and long-term storage architectures are standardized at the cloud side to support decision-making and system optimization.

3.2.3. Data Layer

The data layer serves as the foundation of the intelligent O&M system, aggregating all data obtained through collection, reading, generation, and fusion to support platform operation. Relevant standards and key contents are shown in Figure 11 and include the following four aspects [53,54]:

• Data standards regulate the formats and interface requirements for diverse data types within the intelligent O&M system. They cover structured data, device data from intelligent O&M systems, online monitoring data, offline inspection data, intelligent inspection data, feature data, and data interaction protocols, ensuring consistency and interoperability across all data types.
• Data processing technology: These standards specify the technical processes for analyzing and processing data, including massive data storage, urban data mining, large-scale data cleaning, dynamic modeling of correlated data, and trend prediction. Since different fields often adopt symbols, charts, and notations, unified standards are critical for effective transmission and utilization. International Organization for Standardization (ISO) standards provide a basis for the systematic analysis, organization, calculation, and editing of raw data.
• Data management: These standards address integrating isolated data and restoring correlations to overcome data utilization barriers. They cover data description and cognition, maintenance and management, association and growth, as well as active security and privacy protection. These standards promote dynamic, integrated, and secure data sharing and exchange by establishing clear rules for data identification, association, and governance.
• Database design: Given the wide variety of data involved in the O&M of BMF, rational classification and structuring of databases are essential. Standards in this area regulate database types, partitioning rules, functional requirements, and interaction mechanisms. Examples include structural information databases, static and dynamic response databases, data feature libraries, and O&M information libraries, all of which provide critical support for the intelligent O&M system.

3.2.4. Algorithm Layer

The algorithm layer represents the core competency of intelligent O&M systems, where data-driven analysis and decision-making enable automated control and optimized operations. This layer uses high-quality data to support reliable analysis and decision-making. Although artificial intelligence (AI) has achieved significant success in domains such as facial recognition and autonomous driving, its application in intelligent O&M still faces challenges, including computational governance and efficiency [55,56]. As illustrated in Figure 12, the algorithm layer comprises two main components: algorithm governance and algorithm performance.

• Algorithm governance ensures that intelligent O&M algorithms are secure, ethical, and systematically managed. It consists of three dimensions:
  • Security: Security is the prerequisite for algorithm adoption in O&M. Standards for security evaluation cover aspects such as the safety of objective functions, resistance to algorithmic attacks, reliability of dependent libraries, traceability, and internal control. These provide requirements, evaluation methods, and criteria to ensure secure implementation.
  • Compliance: It defines the ethical and legal boundaries of algorithm use. Standards are needed to promote the industrial growth of intelligent O&M algorithms and prevent malicious or unlawful applications by ensuring accountability and legal enforcement.
  • Classification and Grading: Just as data governance requires data classification and grading, algorithms must also be categorized and graded to build a systematic governance framework.
• Algorithm performance: It is essential to support intelligent O&M at scale, which includes two critical aspects:
  • Explainability: It determines the transparency of an algorithm’s internal logic, technical pathways, decision-making processes, and intended objectives. Algorithms with higher explainability are easier to understand, validate, and manage, facilitating broader adoption. Evaluation criteria focus on the modeling preparation, process, and application stages.
  • Accuracy and efficiency: Accuracy and efficiency are the primary indicators of an algorithm’s effectiveness. High accuracy ensures reliable outputs, while efficiency determines the applicability of solutions in real-time O&M scenarios. Performance evaluation emphasizes requirements, methods, and criteria across the modeling and application stages to ensure dependable results.

3.2.5. Business Layer

The business layer constitutes the core functional level of intelligent O&M systems, translating the data analysis outputs of the algorithm layer into actionable services such as alarms, status assessments, and decision support for BMF. As shown in Figure 13, related standards and key contents cover the following aspects:

1.

Data fusion

This standard regulates data standardization, conversion, and fusion technology. Data standardization refers to the use of uniform methods for describing and expressing structured, monitoring, inspection, and resource data. Data conversion technology outlines processes for transforming unstructured heterogeneous data into usable formats. Data fusion technology defines mechanisms and methods for integrating multi-source heterogeneous data to enable comprehensive analysis.

2.

Early warning and forecasting

These standards establish requirements for selecting early warning indicators, setting thresholds, and implementing mechanisms to reduce false alarms. Indicator selection must consider environmental stimuli, structural responses, and overall structural performance under abnormal events. Thresholds are categorized into structural safety, serviceability, durability, and other dimensions. False-alarm reduction is addressed through structural alarm mechanisms and supporting technologies.

3.

Performance assessment

Standards specify both exceptional and comprehensive performance evaluations. Special assessments target safety (e.g., steel structure fatigue, instability, and bridge overturning) and serviceability (e.g., abnormal vibrations and vortex-induced vibrations). Comprehensive assessments encompass overall condition evaluation, safety evaluation, and lifespan prediction, including reliability and fatigue life assessments for bridges.

4.

Intelligent decision-making

Current maintenance decisions largely depend on expert judgment, which is subjective, limited in its ability to handle complexity, and lacks real-time adaptability. These standards provide technical specifications for applying intelligent technologies and systems in maintenance decision-making, aiming to reduce subjectivity and improve efficiency, reliability, and dynamic capability. Incorporating facilities management can enhance interoperability, planning, and execution of intelligent decision-making.

5.

Resilience improvement

Standards define requirements for preventive maintenance systems and intelligent defect remediation technologies. Preventive systems include typical defect identification, adaptive, and periodic maintenance methods. Intelligent remediation technologies utilize environmentally friendly, high-performance repair materials, targeted repair processes, and innovative remediation equipment to correct defects.

6.

Disaster prevention

This section establishes a disaster protection framework addressing risks throughout the service life of BMF. Key hazards include floods, freezing, mudslides, earthquakes, snow accumulation, fires, and rockfalls. The standards align with national and industry disaster prevention and mitigation guidelines where applicable.

7.

Emergency management

The emergency management process is structured into three stages: before, during, and after an event. The “before” stage includes monitoring, early warning, and emergency support systems. The “during” stage addresses event alarms, reporting, and emergency dispatch. The “after” stage focuses on statistical analysis and emergency evaluation, closing the loop for continuous improvement.

3.2.6. Human–Computer Interaction Layer

The human–computer interaction layer provides the data-driven foundation for large BMF models and diverse operation and maintenance (O&M) scenarios, primarily encompassing digital twin models and AI-based interaction technologies [57], as illustrated in Figure 14.

• O&M digital twin model

The digital twin system is a key tool for enabling intelligent O&M, with standardized requirements for constructing operation and maintenance models, including Building Information Modeling (BIM), digital twin models, 3D surface models, and integrated combined models.

2.

AI-driven human–computer interaction technology

The primary objective is to facilitate seamless information exchange between users and systems, encompassing traditional hardware interfaces and advanced modalities such as voice recognition, gesture recognition, eye tracking, and multi-channel interaction.

3.3. Platform Standards

Platform standards address the design, construction, and acceptance of intelligent O&M systems for BMFs. System software configuration and sensor selection should be demand-driven, prioritizing performance-to-cost ratio rather than scale or novelty. Consensus is required throughout research, development, and application to ensure feasibility, practicality, and ease of operation, while encouraging the development of software products with independent intellectual property rights. Service platforms should be flexible, reusable, upgradable, and maintainable, with clear standards to promote interoperability and resource sharing. These standards cover two key areas [58], as indicated in Figure 15.

3.3.1. Construction Technology

It specifies the design, construction, and acceptance requirements for intelligent O&M systems in BMF, ensuring standardized implementation and quality control.

3.3.2. Capability

It defines the platform-related technical requirements for implementing intelligent O&M systems, including overall architecture, functional design, and key performance attributes such as extensibility, scalability, interoperability, cooperative computing, and dynamic feedback.

3.4. Evaluation Standards

Evaluation standards provide the indicator system for measuring the effectiveness of intelligent O&M in BMF. They should adhere to scientific rigor, comprehensiveness, reliability, comparability, and foresight, ensuring a systematic and quantitative assessment of the entire O&M process. Emphasizing the concept of “promoting construction through evaluation,” these standards aim to guide the effective implementation and sustainable development of intelligent O&M [59]. The main contents are shown in Figure 16 and include the following aspects:

3.4.1. Capability Framework

It defines the evaluation capabilities of intelligent O&M systems, including the perception layer, transmission layer, data layer, warning and assessment layer, decision-making layer, and system resilience.

3.4.2. Indicator System

It specifies detailed evaluation factors under each capability framework, such as the applicability and intelligence of the perception layer, as well as the mean time between failures and maximum single failure time for system resilience.

3.4.3. Algorithms

These standardized evaluation algorithms, including the Analytic Hierarchy Process (AHP) and the Delphi method, correspond to these factors.

3.4.4. Maturity Levels

The development stages of intelligent O&M are assessed, reinforcing the principle of “promoting construction through evaluation” to ensure healthy and sustainable growth.

3.5. Security Standards

The security standards for intelligent O&M systems regulate the protection requirements of the intelligent O&M framework. The related standards and main contents are shown in Figure 17 and include the following aspects [60,61]:

3.5.1. Physical System Security Requirements

These standards specify the safety requirements of physical systems within the intelligent O&M framework. They include physical system risk analysis, electrical system safety, mechanical system safety, and intrinsic safety.

3.5.2. Functional Safety Requirements

These standards regulate the safety-related technical requirements in the design, construction, and operation processes of intelligent O&M systems. They cover safety risk analysis, safety function design, and safety integrity level assessment of the O&M system.

3.5.3. Information Security Technology

These standards define the security requirements for network and information technologies used in intelligent O&M. They include security reference architectures and technical requirements, data encryption and decryption, secure transmission, identity authentication, network identity management, and information labeling. Establishing clear data governance policies, defining access rights, and ensuring responsible use of sensitive information are essential to maintaining public confidence and supporting the sustainable implementation of intelligent O&M frameworks.

3.5.4. Information Security Management

This is crucial to intelligent O&M standardization for BMFs. It establishes standards for security management itself, including security management systems, organizational management, personnel management, system construction management, and system operation and maintenance management.

3.6. Industry Applications

Based on the general, key technical, platform, evaluation, and security standards of intelligent O&M, and considering the specific needs and characteristics of different industries and fields, industry-specific application standards should be developed. These include intelligent O&M technical standards for roads, bridges, and tunnels; intelligent O&M technical standards for buildings, urban complexes, and integrated station–city three-dimensional spaces; and intelligent O&M technical standards for municipal water supply networks, drainage networks, and natural gas pipelines, as illustrated in Figure 18.

As indicated in the literature review in the Introduction Section, several components of the standard systems proposed in this study have already been applied or implemented across various civil and infrastructure domains. Structural sensing technologies have become increasingly advanced, with significant improvements in hardware, including data acquisition systems. Currently, comprehensive monitoring of structural responses is possible, encompassing strain fields, linear shapes, displacements, accelerations, and related parameters. In parallel, the surrounding environmental conditions, such as traffic loads, temperature, and wind speed, can also be effectively monitored. On-site inspection, testing, and monitoring techniques are becoming increasingly diverse. These include non-destructive testing methods such as electromagnetic and acoustic emission techniques, load testing, and approaches based on environmental or impact-induced vibrations, as well as rapid testing using moving vehicles. Collectively, these methods provide essential data foundations for multidimensional structural characterization and performance evaluation, ranging from microscopic to macroscopic levels. Meanwhile, the development of data analysis, structural feature identification, and information inversion technologies, such as wavelet analysis, neural networks, pattern recognition, and data mining, has dramatically advanced the estimation of structural parameters and information inversion based on monitoring data. Finally, structural performance assessment theories have also matured, and various methods for evaluating performance and load-bearing capacity based on inspection and monitoring data have been proposed and successfully applied in practice [49,62].

However, numerous challenges and unresolved issues remain. Broadly, the main difficulties lie in two aspects. On the one hand, the diversity of sensor types, the complexity of system integration, limited durability, and the absence of effective data processing and management strategies hinder the efficient utilization of massive monitoring datasets. On the other hand, extracting reliable structural modal parameters from such high-dimensional data for accurate damage detection and performance evaluation remains a significant obstacle. Taking the three-dimensional characteristics of intelligent O&M system architectures in Figure 3 as an example, considerable progress has been achieved primarily in the field of intelligent perception. In contrast, other dimensions remain in their early stages of development. Similarly, the algorithm layer, business layer, and human–computer interaction layer within hierarchical system frameworks continue to evolve. Consequently, the current maturity level of most intelligent O&M systems can generally be classified only as Level 1 (L1) or Level 2 (L2).

4. Discussion

The new generation of information technology and the resulting Industry 4.0 are reshaping contemporary society, bringing new opportunities for developing and managing O&M in BMFs. Standardization is a crucial technical support for intelligent O&M management in these facilities, functioning as the foundation and the driving force. Existing O&M standards provide a wealth of effective technical accumulation, laying the groundwork for intelligent O&M. However, these standards mainly focus on traditional manual O&M modes, exhibiting intense subjectivity and failing to address the challenges posed by labor shortages and complex O&M scenarios in the modern BMFs. Based on operations management science theory and a systematic, integrated micro-analysis approach, this paper analyzes the connotations of intelligent O&M for BMFs and its corresponding standard system, drawing on a literature review and industry surveys. The main conclusions are as follows:

1.

Intelligent O&M for BMFs refers to a management approach that leverages new-generation intelligent technologies and equipment, such as IoT, robotics, big data, and artificial intelligence, to achieve a comprehensive perception of service conditions, deep fusion analysis, intelligent interaction, and scientific decision-making. Its core lies in real-time monitoring and analyzing BMF’s various O&M data. It integrates periodic inspections and routine patrol information to detect abnormal conditions, fault risks, and performance degradation, providing timely safety warnings and supporting maintenance decision-making.

2.

Intelligent O&M systems deliver seven key services: perception, data fusion, status assessment, early warning and forecasting, decision-making, intelligent control, disaster prevention, and emergency management. These functions are structured across hierarchical layers, from perception to human–computer interaction, and their intelligence maturity progresses through five levels, from basic presentation and monitoring to dynamic fusion, performance projection, and autonomous maintenance. This framework highlights the comprehensive and multi-layered nature of intelligent O&M, providing a clear path for advancing system capabilities.

3.

The intelligent O&M standard system framework for BMFs mainly comprises six parts: general standards (A), key technical standards (B), intelligent O&M platform standards (C), intelligent O&M evaluation standards (D), intelligent O&M security standards (E), and industry application standards (F). According to their differing attributes, their subordinate standards and main technical contents also vary accordingly.

4.

Beyond the technical and managerial dimensions, the successful implementation of the proposed framework is inherently influenced by the surrounding socio-economic and policy environment. The economic feasibility, including factors such as initial investment and maintenance expenditure, is achieved through data-driven decision-making and plays a pivotal role in determining the adoption and scalability of the framework. At the policy level, supportive regulations, institutional coordination, and incentive mechanisms are essential for promoting the standardization, interoperability, and large-scale integration of O&M systems across different agencies and regions. Furthermore, stakeholder alignment, which involves government authorities, private operators, and end-users, has a significant impact on implementation efficiency and sustainability. Incorporating these socio-economic and policy perspectives enables a more comprehensive and realistic understanding of how the proposed framework can be effectively deployed, institutionalized, and maintained in real-world practice.

5. Conclusions

In the future, to further improve the adaptability and general applicability of the proposed framework, interdisciplinary integration beyond the scope of operations management should be emphasized. From the perspective of information systems, ensuring the interoperability and seamless integration of multi-source monitoring and inspection data is fundamental to building a unified digital management ecosystem. Within the domain of urban informatics, the framework can be extended to link infrastructure-level data with city-scale information networks, thereby facilitating cross-domain and multi-scale decision-making in smart city environments. Moreover, insights from organizational behavior offer a valuable understanding of the human and institutional dimensions that influence the adoption and long-term sustainability of digital O&M systems, such as user training, interdepartmental collaboration, and organizational resistance to technological transformation. By incorporating these interdisciplinary perspectives, the proposed framework can evolve into a more resilient, adaptive, and human-centered lifecycle management paradigm for infrastructure systems.