Smart or Intelligent Assets or Infrastructure rely of technology to deliver its services and functionality. Technology is further divided into four categories: sensor, network, server and workstation.
5.1. Sensor
Field devices embedded into the asset provide specific functionality such as video surveillance, access control, intercoms, environmental, location, and occupancy. These field devices can be powered via a dedicated power supply, the network Power over Ethernet (PoE) or batteries. IoT devices are normally battery-powered based on simple hardware with limited computing power designed to perform simple computing operations and transmit data quickly at low bandwidth while consuming as little energy as possible.
Sensors transmit information via open standard protocols to the server layer located at the edge or in the cloud. HyperText Transfer Protocol Secure (HTTPS) is designed for the transmission of documents in client-server applications rather than IoT data communication from mobile devices or sensors. Message Queuing Telemetry Transport (MQTT) and Constrained Application Protocol (CoAP) are the main OSI Application Layer IoT protocols for constrained networks defined as low bitrate, high packet loss and high asymmetric links. Both are designed as client/server models via the transmission of TCP/UPD IP packets with mechanisms for asynchronous communications with binary data messages for small payloads therefore low RAM memory and power consumption.
MQTT protocol transmits mainly binary live data via TCP connection data stream to a detached broker as a client-sensor/server-broker model. Every message in MQTT is a discrete piece of information published to an address, denominated as a topic where the broker transmits every message published to the topic to the various subscribed clients and clients can subscribe to various topics. The main issue with MQTT is the topic string can be composed of large strings, therefore, inhibiting its low power, low computing, and low bandwidth features. COAP transmit information via UDP connectionless datagrams with an inbuilt functionality for content negotiation and discovery that allows devices to validate each other and to discover different methods for exchanging data. CoAP is primarily a client-to-server protocol for transferring state information rather than events that integrates with the HTTP and RESTful protocols due to its packet fragmentation mechanism. CoAP performs better than MQTT in terms of bandwidth usage and round-trip time, therefore, reducing network utilization, device memory and power, on the other hand, MQTT is more reliable with congestion control mechanisms and Quality of Service (QoS).
Lightweight M2M (LwM2M) aims to simplify and standardize device management and data transfer between sensor and server in the IoT based on TCP, CoAP and REST architecture.
5.2. Network
The fast and large increment of mobile devices and content coupled with server virtualization and cloud services where users access applications anywhere and anytime are the main reasons traditional network architectures are adapting to changing traffic patterns. The network transmits information between the sensors and the server for machine and human communications.
5.2.1. Software Defined Networks
A software-defined wide area network (SD-WAN) over the public Internet creates virtual private networks and simplifies the management and operation of a WAN by decoupling the networking data plane hardware from its control plane software. Bandwidth is shared without an established Quality of Service (QoS). Secure Access Service Edge integrates networking and security functions into a cloud solution. Software-defined networking enables controlling and optimizing the routing of data packets through a centralised server.
5.2.2. Internet Connectivity
Fixed wired Internet Service Providers (ISP) provide external connectivity between the different users of the assets and infrastructure, including datacentres. Technologies include Asymmetric Digital Subscriber Line (ADSL) and Integrated Services Digital Network (ISDN) based on the 2-wire copper telephone line, Fibre to the Cabinet (FTTC), or Fibre Broadband where the telephone line connects the cabinet to the property. Hybrid Fibre-Coaxial is the same as a Fibre to the Cabinet (FTTC), however, Coaxial Cable is used instead of phone lines. Fibre to the Premises (FTTP) the connection is via a fibre optic cable from the exchange to the asset at speeds up to 2 Gbps per connection.
Wireless Internet Service Providers (WISPs) deploy solutions including Wireless Fidelity (Wi-Fi-IEEE 802.11), Worldwide Interoperability for Microwave Access (WiMAX-IEEE 802.16) or Satellite. Low-cost Wi-Fi connectivity is designed with coverage of around 100 m and 150 m. WiMAX normally delivers the last mile/kilometre wireless broadband access as a substitute to fixed DSL and cable at speeds (approx. up to 1 Gbit/s with 2.3–3.6 GHz). Satellite Internet access provides data rates ranging from 2 kbit/s to 1 Gbit/s downstream and from 2 kbit/s to 10 Mbit/s upstream at 1–40 GHz. Satellited can be in Geostationary Earth Orbit (GEO) with a total latency between 0.75 to 1.25 s that affect real-time applications. Satellites in low Earth orbit (LEO) and medium Earth orbit (MEO) require more satellites at variable positions above the Earth although they provide lower latencies (125 ms, and 7 ms), respectively, with associated higher speeds.
5.2.3. Mobile
The 5G mobile infrastructure provides low latency services for large data streams within very short, unobstructed transmission links to enable applications based on mobile broadband, IoT, and mission-critical applications; 5G is designed to provide up to 10 Gbit/s therefore with competing services Internet Service Providers (ISPs). Furthermore, 5G provides higher transmission speeds and bandwidth; therefore, it is able to connect more devices, improving the quality of service (QoS) in crowded areas. Additionally, 5G can be implemented in low-band, (600–900 MHz, 30–250 Mbit/s), mid-band (1.7–4.7 GHz 100–900 Mbit/s) and high-band (24–47 GHz, 1–10 Gbit/s. The 5G Radio Access Network (RAN) is supported by macro-cells based on Multiple Input, Multiple Output (MIMO) antennas and millimetre waves (mmWave). In addition, small-cell base stations with edge computing capabilities complement the macro network are distributed in dense clusters supported by indoor distributed antenna systems. Normally, the lower frequency of the radiofrequency spectrum provides the widest coverage and presents lower penetration losses. On the other side, the higher frequency the higher the bandwidth although requires a line of sight.
5.2.4. Wi-Fi
Wi-Fi 6 (802.11ax) provides higher bandwidth wireless connectivity with greater throughput at lower latency enabling connectivity to more IoT and mobile devices in the network in the frequency range between 1 and 7.125 GHz. The main two technologies that support Wi-Fi 6 are Multi-User, Multiple Input, Multiple Output (MU-MIMO) permit communication from multiple devices to the access point at the same time. Orthogonal Frequency Division Multiple Access (OFDMA) enables a higher spectral efficiency via the transmission of information to multiple devices at the same time.
5.2.5. Radio
Private Radio Systems (PRS) mostly support the communications of emergency services, public safety organisations and asset operators via VHF or UHF bands that transmit power typically limited to 4 watts to provide a reliable coverage between 5 to 30 km depending on terrain. Private Radio technologies include Private Mobile Radio (PMR), Trunked Radio System (TRS) such as Terrestrial Trunked Radio (TETRA) (Bandwidth-25 kHz Frequency: 380–430 MHz) or P25 (Frequency 136–859 MHz, Bandwidth: 12.5 kHz), Digital Mobile Radio (DMR) (12.5 kHz, 66–860 MHz). Due to PRS being privately owned with its associated additional OPEX and CAPEX and the low bandwidth provision that does not support Big Data applications such as Virtual Reality, BIM or video calls, the trend is the outsourcing of its services to 5G or replacement by Wi-Fi.
5.2.6. Low Power Wide Area
A Low Power Wide Area (LPWA) is a wireless Wide Area Network with long-range coverage up to 10 km at a low bit rate (0.3 kbit/s to 50 kbit/s per channel) designed for the IoT based on low-cost lightweight protocols, reduced hardware complexity and low power normally via battery or directly harvested. The network itself is also low cost based on a simplified design: star topology device to a gateway, reduced expensive infrastructure requirements, and the usage of license-free bands. Low power low bandwidth devices transmit long-range data to a gateway mainly via LoraWAN, and EnOcean wireless communication protocols, the gateway provides edge computing and forwards relevant information to the cloud via a fixed or wireless network. Due to its extensive coverage, LPWA includes smart city applications such as air-quality data, waste-management data, parking-availability data, or smart meter readings.
Long Range WAN (LoRaWAN) is considered the equivalent to the OSI Layer 2 datalink and OSI Layer 3 network layers (863–928 MHz and 2.4 GHz). Typically, LoRaWAN operates on top of LoRa which is the equivalent of Layer 1 physical OSI Layer. LoRaWAN is mostly designed for outdoor applications with battery powered sensors. On the other side, the EnOcean wireless protocol (868.3 MHz, 125 kbit/s) was designed for self-powered devices that transmit very low data and use very reduced amounts of power. The required power can be locally harvested via thermal differences or equipment vibration to recharge batteries and expand their autonomous working life for many years with no maintenance. EnOcean supports low latency suited for mission critical applications and short range (30 m) for indoor applications such as building or transport automation.
5.2.7. Local Area Networks
A local area network (LAN) interconnects devices within one limited physical location or area (1 Gbps–10 Gpbs) via twisted copper pairs or fibre optics. LANs enable access to centralized applications such as servers, sharing resources including printers and access to the Internet; therefore, it can enhance communications and flexibility while holistically protecting the network from external attacks. Local Area Networks in access, distribution and core configuration for the asset converged network provides connectivity and power to fixed sensors based on the OSI Layer 2 Ethernet (IEEE 802.11) based on 48-bit MAC addresses.
5.2.8. Wide Area Networks
In contrast, a wide area network (WAN) or metropolitan area network (MAN) covers larger areas by interconnecting several LANs together via the OSI Layer 3 IP protocols. Several technologies support a WAN. Private circuit mechanisms typically are delivered over Synchronous optical networking (SONET)/synchronous digital hierarchy (SDH) and have been widely utilized for the past several years. The reliability inherent in SONET/SDH is due to the automatic protection switching element, which provides recovery within 50 milliseconds. However, the lack of bandwidth flexibility makes private circuit services less adaptable to current network and traffics demands. Asynchronous Transfer Mode (ATM) circuit and packet switching protocols via asynchronous time-division multiplexing are used in the SONET/SDH backbone of the public switched telephone network (PSTN) and the Integrated Services Digital Network (ISDN) based on the transmission of a fixed size 53-byte cells for real-time, low-latency content such as voice and video at approx. 135 Mbit/s. Frame Relay (FR) is a standardized packet switching methodology designed for transport across ISDN infrastructure. Multi-Protocol Label Switching (MPLS) is a key protocol for delivering voice, video, and data services on IP networks, it operates at OSI layer 2.5. Metro Ethernet involves circuits with Ethernet interfaces where assets can subscribe to high bandwidth at rates of 1 Gbit/s or higher.
5.2.9. Personal Area Networks
Personal Area Networks (PANs) provide wireless data transmission among devices replacing the serial cables or “RS” protocols based on low power (mW), and low range (10 m) communications. ZigBee protocol targets self-powered IoT sensor solutions in the 2.4 GHz IEEE 802.15.4 standard (868 MHZ–20kbits 2.4 GHZ 50 kbits/s). Bluetooth (IEEE 802.15.1) uses UHF radio waves in the ISM bands, from 2.402 GHz to 2.48 GH with a transmission power limited to 2.5 milliwatts for a range of up to 10 m.
5.3. Server
The server element collects data from the sensors and performs specific actions according to the required functionality or service. Server or computing can be distributed at the device level (fog computing), at the gateway level (edge computing) or at the datacentre level (cloud computing) to balance the triangle between power consumption, latency and bandwidth.
5.3.1. Systems
The specific functionality of the server function for assets or infrastructure includes security systems (video surveillance, access control, intrusion detection), Building management systems (HVAC, energy, lighting, fire, public address voice alarms) Building monitoring systems (environmental conditions, occupancy, energy), user experience systems (visitor management systems, resource and space booking, lockers, wayfinding), asset management (devices, facilities, property, space, leases, maintenance, project management).
Specific asset management systems also reside in the server layer such as Integrated Workplace Management System (IWMS), Computer Aided Facility Management (CAFM), Computerized Maintenance Management System (CMMS), and Asset information for asset register and maintenance. These systems enable functionality including maintenance, project, lease, rent, cost, auditing, investment, and space management supporting cost efficiencies and expenditure effectiveness while reducing human interventions that lead to time consuming and prone-to-error inefficient operations. One of the main challenges of asset management solutions is the alignment between daily activity and short-term plans between individual departments and the longer-term objectives of the organization while keeping accountability or placing undue constraints on individual departments. This alignment includes other management aspects such as maintenance. Costs are normally reduced with routine maintenance; however, unnecessary planned maintenance also intensifies expenses. Timing these decisions support extending an asset’s life cycle at a reduced cost.
Asset accounts management systems are used by a variety of stakeholders to assess the financial and business performance of the organization, supported by consistent and robust evidence with the alignment of technical and financial data. Accounts are subject to auditing and policies such as expenditure, revenue recognition, capitalisation, valuation, depreciation policy, asset impairment, capital rationing policy or tax treatment policies. Finally, asset financial models include financial requirements, capital investment requirements, capital and revenue investment streams, projects, maintenance, energy, utilities, and operational and facilities management.
5.3.2. Open Standard Protocol
The asset middleware combines the individual functionality of the different system servers for advanced functionality via system integrations based on open protocols and standards. This overarching middleware removes organizational or team silos as different users’ access and benefit from a single interface.
Digital Addressable Lighting Interface (DALI) is designed for the interoperability of lighting control in the IoT (IEC 62386). Each device is assigned a unique logic address between 0 and 63 where devices can also be programmed to operate in groups devices, report a failure, or answer a query about its status or other information. A single pair of wires generates the bus used for signal and power. Data are transferred between devices via asynchronous, half-duplex, serial protocol over a two-wire bus with a fixed data transfer rate of 1200 bit/s.
KNX is developed for commercial and domestic building automation mostly in Europe (EN 50090, ISO/IEC 14543, ANSI/ASHRAE 135). KNX supports devices to form distributed applications or decentralized topology via a twisted pair bus. This is implemented via interworking models with standardised datapoint types and objects that model process and control variables in the system. KNX can connect up to 57,375 devices using 16-bit addresses distributed in areas lines and segments.
BACnet is mostly used in building automation and control (BAC) networks (ANSI/ASHRAE 135-2016, and ISO 16484-5). BACnet was designed to allow communication of devices regardless of the particular building service or systems. It is normally applied in commercial HVAC control and Building Energy Management Systems (BEMS).
Modbus is a network communications protocol suited for industrial automation systems. It connects electronic equipment over serial lines in a master (requesting information)/slave (transmitting information) arrangement in building automation, transport, and energy such as metering.
5.3.3. System Integrations
Two separate applications communicate information via an intermediary bridge called Application Programming Interfaces (APIs) that enable one system to access the information or functionality of another via standardised protocols. The most widely used APIs are SOAP (Simple Object Access Protocol), REST (Representational State Transfer), GraphQL, and Remote Procedure Call (RPC).
Remote Procedure Call RPC defines a remote execution of a function into a separate environment. The process consists of the client invocation of a remote method, the serialisation of the parameters and additional information into a message, and finally the transmission of the message to the server. After the server receives the message, the process consists of the deserialization of its content, execution of the requested operation, and the transmission of a result back to the client. RPC uses GET to collect information and POST for everything else based on a high message rate and very low overheads. With its tight coupling, RPC is normally applied in internal microservices.
SOAP makes data available as Web-based services based on an XML-formatted, highly standardised Web communication protocol language and a platform-agnostic environment. A SOAP message consists of an envelope tag that starts and finalises every message, a body that contains the request or response, a header to determine any specifics or extra requirements, and a fault that reports any errors from the process of the request. Due to SOAP presenting a static structure coupled with security and authorization features it enforces formal software contracts and complies with legal contracts between the API provider and consumer. SOAP applications include billing, booking, and payments for private or enterprise distributed applications between heterogeneous platforms.
Representational State Transfer (REST) is architected for developing high-performance, scalable services based on Web services. It is based on the concept of resource-oriented architecture, where resources are identified by their URI (Uniform Resource Identifier). REST makes server-side data available and represents them in simple formats, often JSON and XML, that decouple client and server via HTTP Communication between the client and server.
GraphQL is based on building a schema aiming to make precise JSON data requests to retrieve data from multiple sources. GraphQL messages are self-describing for distributed environments and therefore suitable for mobile devices to load data from multiple APIs.
5.3.4. Data Structure
Data or information have their own life cycle and management, similar to the asset that supports monitoring, audit, assurance, and performance benchmarking. Information requires maintenance to comply with legal and other statutory requirements, this includes change management, storage, retention and final. Information Lifecycle, processes physical and functional requirements. Raw asset data automatically retrieved from the different sensors are stored in a data lake following specific naming conventions such as Brick Schema, Haystack, OSCRE and Google Digital Buildings. Data ontologies are based on the Semantic Web principles and provide a uniform schema and toolset for representing structured information and model systems and components relationships. This standardisation enables its portability and consistency between multiple assets independently of individual inputs from the different users through the asset’s WLC. Data ontologies support the management and operations of very large, heterogeneous asset portfolios in a scalable way. Data models support advanced functionality based on energy audits, automated fault detection and diagnostics, asset automation, the complex search of information and analytics and optimisations.
BIM Standards based on ISO 19650 [
45] structured information management applied to the entire WLC of a built asset to support collaboration and procurement competition. Specifically, ISO 19650 is based on Information Requirements (Employer’s Information Requirements, Organisational Information Requirements, Asset Information Requirements, BIM Execution Plan), Information Models (Project Information Model, Asset Information Model), Collaborative Practices(Common Data Environment, Interoperability, Industry Foundation Classes and COBie).
The Digital Building Logbook [
46] is a proposal aiming at establishing a common European approach that aggregates relevant building data and ensures that authorised people can access accurate information about the buildings. Users are intended to be the market players such as property owners, tenants, investors, financial institutions and public administrations.
5.3.5. Data Formats
Different data formats support the digital creation of information and document in terms of the content, as the displayed information, structure as how the information within the document is structured and finally the formatting as the visual appearance. The first data standard was Hyper Text Markup Language (HTML) which structures the content of the document in tags and provides formatting rules to display data. HTML is a rendering protocol for Web browsers that lacks structure within tags. As a consequence, Extensible Markup Language (XML) was developed to encode the information structure of the document, or the representation of arbitrary data structures and transmit data readable for humans and machines. XML sends structured data within a web-based system and does not include a programming syntax. As Web pages become more dynamic or interactive via Web applications developed via software such as Javascript, rather than just displaying content via markup languages JavaScript Object Notation (JSON), structures information as key-value pairs to support data storage and interchange between software applications, although still hierarchical and human-readable. Geography Markup Language (GML) models geographic features via the Web Feature Service (WFS) Interface Standard.
5.3.6. Artificial Intelligence
Artificial Intelligence (AI) provides different applications to asset management based on classification, and regression algorithms for the different asset management functions. By analysing different sources of data, AI reduces asset risk under several predicted scenarios such as political, economic, social, technological, legislative and environmental. AI analyses and refines the Big Data stored from the different sensors to find patterns and detect anomalies supporting applications that optimize variables such as energy consumption or space utilization or predict the growth of demand, trends in customer behaviours, analyses of market conditions, and long-term resources. AI has the potential to expand the asset life cycle via the learning of the right intervention at the right time for its rehabilitation, reparation and replacement based on preventive and proactive maintenance.
5.3.7. Tokenisation
Web3, based on Distributed Ledger Technologies (DLTs) enables the movement of asset information between parties transparently where every asset transaction is auditable and verified without the need for a third-party central authority. DLT has the potential to disrupt ingrained legacy operating models based on the manual reconciliation of “Book of Records” or ledgers that are the barrier to change with an almost real-time solution. The time between trade date and settlement date is still from two to three days in most markets where the risk of counterparty default is often mitigated with collateral, which adds unnecessary transaction costs. DLTs enable a single version of truth between investors, managers, and sellers. Different users can share asset information and transaction history without compromising trust or confidentiality or human error. The tokenization of the asset supports a faster resolution of issues related to dispute resolution and improves the time to find information and solve discrepancies in data while enabling new digital parties to deliver new or traditional services. Smart contracts or tokenisation enhance the ease of asset management and the respective supply chain by prescribing the terms and conditions of the service which must occur for transactions to take place and making the communication of such conditions being reached irrefutable. These applications have the potential of reducing costs and processes time, increasing operational efficiency, improving transparency and facilitating a series of innovative investments. As services become digitally managed, Artificial Intelligence can draft automatically smart contracts based on Natural Language Processing and make autonomous decisions via Big Data analytics and decision trees. Tokenisation supports also regulatory compliance, information security rules, privacy laws and investor reporting. New services include asset tokenization, or tokenized ownership, as the securitization of high-value goods that broadens the investor base, increases liquidity, and decreases resell risks where parties trade and settle directly within minutes at low cost.
5.4. Workstation
Workstation identifies the different interfaces or channels users of the smart asset or infrastructure have for its management, operations, or usage. As the access to the servers and information is performed via a Web Brower, there is no need to have a dedicated workstation, laptops, mobile phones and tablets also can be used where functionality can be bundled into mobile apps that aggregate different services and functionality into a common user channel. Although users have the information via their mobile phones, digital signage in fixed locations is still relevant to easily show relevant information, such as train, flight timetables or asset conditions including occupancy and environmental conditions. Digital signage can the updated in real-time and used to visually reinforce messages from the Public Address/Voice Alarm in emergencies.
Dashboards support asset managers and operators to visually group, analyse and filter information to make data-driven decisions for asset and portfolio levels. These decisions seek the enhancement of the asset performance in terms of energy consumption, environmental conditions, maintenance, or space utilisation. Dashboards support longer strategic decisions on longer-term strategic decisions by specific criteria such as Key Performance Indicators (KPI) based on value generation whereas balancing the asset’s performance, service, capacity, expenditure, risks, operational constraints, compliance to regulatory safety or another statutory requirement and environmental. Decisions in asset management include the time dimension from day to day for operational delivery to yearly tactical planning and provision for longer-term life cycle considerations.
The digital model representation of the physical asset as the single source of truth via Geographic Information Systems (GIS), Building Information Models (BIM) and Digital Twins support asset manager, operations and maintenance to find relevant asset information within a 3D model where different dimensions or layers can be added such as time, financial, sustainability, facilities management, utilisation. Real-time device information can be embedded within the Digital model supported via IoT networks.
BIM models have the potential to include the asset information through its lifecycle, from design and construction to its operations and final decommissioning. In addition to the traditional 3D BIM, more dimensions have been added: 4D includes time and construction sequences, BIM 5D adds CAPEX and OPEX cost; 6D provides facilities management information; 7D is for sustainability information; 8D brings health and safety considerations. Geographic information places the asset to the earth coordinates within a spatial and cartographic landscape. Augmented Reality provides an interactive experience of real-world environments for training in terms of operations and maintenance.