Inter national J our nal of Inf ormatics and Communication T echnology (IJ-ICT) V ol. 15, No. 2, June 2026, pp. 909 924 ISSN: 2252-8776, DOI: 10.11591/ijict.v15i2.pp909-924 909 Semantic inter operability in IoT f or Industry 4.0: Re view , taxonomy , challenges, and futur e r esear ch De v amekalai Nagasundaram 1 , Erum Ashraf 2 , Selv akumar Manickam 1 , Shams Ul Arfeen Laghari 3 , Shankar Karuppayah 1 1 Cybersecurity Research Centre, Uni v ersiti Sains Malaysia (USM), Gelugor , Malaysia 2 Department of Computer Science, Bahria Uni v ersity Islamabad, Islamabad, P akistan 3 F aculty of Engineering Design Information and Communication (EDICT), Bahrain Polytechnic, Isa T o wn, Bahrain Article Inf o Article history: Recei v ed Aug 28, 2025 Re vised Apr 1, 2026 Accepted Apr 15, 2026 K eyw ords: Industry 4.0 Internet of things Ontology Re vie w Semantic interoperability ABSTRA CT Semantic interoperability is a critical enabl er for achie ving the Industry 4.0 vi- sion, ensuring that heterogeneous IoT de vices, systems, and applications can e x- change and interpret data consistently . Despite its importance, achie ving seman- tic interoperability continues to pose signicant challenges due to the di v ersity of data formats, standards, and ontol ogies used across industrial IoT en viron- ments. This paper presents a comprehensi v e re vie w and taxonomy of semantic interoperability within Industry 4.0, analyzing e xisting frame w orks, protocols, and ontological models. W e classify current approaches based on their architec- tural layers, semantic technologies, and application domains. Additionally , this study identies the limitations of pre v ailing solutions, highlights open research challenges, and proposes future directions for enhancing semantic interoperabil- ity in industrial IoT systems. The insights pro vided aim to support researchers and practitioners in de v eloping scalable, secure, and semantically aligned IoT ecosystems for Industry 4.0. This is an open access article under the CC BY -SA license . Corresponding A uthor: Selv akumar Manickam Cybersecurity Research Centre, Uni v ersiti Sains Malaysia (USM) Gelugor , Pulau Pinang, Malaysia Email: selv a@usm.my 1. INTR ODUCTION The IoT refers to a rapidly e xpanding ecosystem of interconnected ph ysical objects ranging from v ehicles and home appliances to industrial machinery which are embedded with electronics, s o f tw are, sensors, and netw ork connecti vity , enabling autonomous data collection, e xchange, and processing [1]. This inte gration of the ph ysical and digital w orlds has dri v en transformati v e changes across sectors such as manuf acturing, healthcare, transportation, and smart cities, impro ving operational ef cienc y , decision-making accurac y , and economic producti vity . The emer gence of Industry 4.0 has further accelerated the deplo yment of IoT , by mer ging c yber - ph ysical systems (CPS) with intelligent industrial i nfrastructures to enable autonomous, real-time, and adapti v e production en vironments [2]. Central to the success of Industry 4.0 is the seamless interchange, comprehen- sion, and ut ilization of data generated by heterogeneous IoT de vices, platforms, and services. A major barrier to achie ving this vision is the lack of semantic interoperability , which ensures that de vices and systems from di v erse manuf acturers interpret and process e xchanged data with a consistent, shared understanding [3]. W ith- out this capability , Industry 4.0 infrastructures struggle to inte grate ne w de vices and services ef ciently and J ournal homepage: http://ijict.iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
910 ISSN: 2252-8776 manage data-dri v en decision-making w orko ws reliably . The global IoT landscape continues to e xpand at a remarkable pace, with recent projections esti mating that o v er 75 billion IoT de vices will be operational by 2026 [4]. This e xponential gro wth is fueled by the gro wing demand for smart homes, industrial automation, connected v ehicles, and wearable technology [5], [6]. Consequently , the v olume of dat a generated by IoT systems is anticipated to surpass 175 zettabytes annually by 2025, presenting unprecedented challenges in terms of data storage, real-ti me processing, inte gration, and analysis [7], [8]. A primary obst acle lies in the signicant heterogeneity of IoT de vices, which v ary widely in terms of their hardw are capabilities, communication protocols, data formats, ontological models, and security archi- tectures [9], [10]. This heterogeneity contrib utes to fragmented IoT ecosystems, dat a silos, and interoperabil- ity bottlenecks, undermining the scalability , adaptability , and reliability of Industry 4.0 infras tructures. The problem is further compounded by the accelerated pace of digital transformation catalyzed by the CO VID-19 pandemic, which highlighted the ur gent need for interoperable, resilient, and scalable IoT architectures capable of supporting autonomous and distrib uted industrial operations [11], [12]. Recent studies ha v e e xplored v arious semantic interoperability frame w orks, ontologi cal models, and middle w are solutions that aim to harmonize data semantics across heterogeneous IoT en vironments [10]. No- tably , Multidisciplinary Digital Publishing Institute (MDPI) research has proposed a metamodeling-based inter - operability and inte gration testing platform that formalizes IoT system interactions and enables cross-platform v alidation across di v erse de vices and data o ws [13]. This w ork demonstrates the feasibility of systematic in- teroperability management approaches b ut highlights ongoing limitations in dynamic semantic alignment and real-time inte gration for lar ge-scale industrial deplo yments. Moreo v er , research in applied domains has emphasized the practical adv antages of semanti c interop- erability . F or instance, a spatio-temporal semantic data management frame w ork has been deplo yed in precision agriculture to enhance interoperability in IoT -dri v en f arming en vironments [14]. Similarly , an ontology-based semantic middle w are for smart campus infrastructures has demonstrated the ability to automate de vice and data inte gration w orko ws in heterogeneous IoT systems [15]. These implementations reinforce the importance of semantically a w are architectures, though uni v ersal, scalable, and domain-independent solutions remain elusi v e. Despite these adv ancements, achie ving seamless, scalable, and dynamic semantic interoperability in IoT systems for Industry 4.0 re mains an open research challenge. The absence of uni v ersally adopted seman- tic frame w orks and the limited maturity of real-ti me semantic alignment mechanisms continue to hinder the inte gration of heterogeneous de vices, platforms, and services [15]. T o address these challenges, this paper presents a comprehensi v e re vie w and taxonomy of semantic in- teroperability frame w orks, ontol og i es, and enabling technologies for IoT in Industry 4.0. It cate gorizes e xisting approaches based on their architectural layers, semantic models, and application domains; identies persistent limitations and open research challenges; and proposes future research directions to guide the de v elopment of scalable, dynamic, and domain-independent semantic interoperability solutions for industrial IoT ecosystems. Semantic interoperability has become a critical research topic within the industrial IoT domain, g ain- ing increasing attention due to its role in enabling seamless data e xchange and system inte gration in Industry 4.0 en vironments. Numerous studies ha v e in v estig ated v arious approaches to achie ving semantic interoperability ho we v er , se v eral open challenges remain. T o understa n d the current state of the art, identify e xisting g aps, and propose future directions, this paper conducts a comprehensi v e re vie w of recent research on semantic interoperability in IoT for Industry 4.0. In particular , the follo wing research questions (RQs) guide the scope and objecti v es of this study: - RQ1: Ho w is semantic interoperability dened in the conte xt of Industry 4.0 and industrial IoT? - RQ2: What approaches and strate gies ha v e been proposed in pre vious studies to address semantic interop- erability in IoT , and ho w ef fecti v e are the y? - RQ3: What are the primary challenges and limitations f aced by IoT systems and Industry 4.0 infrastructures due to the lack of semantic interoperability? - RQ4: What future research directions and open challenges need to be addressed to enhance semantic inter - operability in IoT systems for Industry 4.0? T o address these research questions, the remainder of this paper is or g anized as follo ws: Section 2 pro- vides an o v ervie w of interoperability components within IoT systems. Section 3 and sect ion 4 discuss semantic interoperability technologies and the k e y obstacles to achie ving semantic interoperability , respecti v ely . Sec- tion 5 outlines the operat ional and inte gration challenges caused by the lack of semantic interoperability in IoT Int J Inf & Commun T echnol, V ol. 15, No. 2, June 2026: 909–924 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Inf & Commun T echnol ISSN: 2252-8776 911 en vironments. A comprehensi v e re vie w of related w orks is presented in section 6. Finally , the paper concludes with a summary of k e y ndings, current limitations, and proposed future research directions in section 7. This paper aims to pro vide a comprehensi v e analysis of semantic interoperability in IoT syst ems within the conte xt of Industry 4.0. It re vie ws the current state of interoperability technologies, highlights the challenges posed by semantic heterogeneity , and identies open research problems that hinder seamless data e xchange and inte gration in industrial IoT en vironments. The main contrib utions of this study are summarized as follo ws, - T o dene and clarify k e y concepts related to IoT interoperability , including semantic interoperability , se- mantic technologies, and their underlying models and frame w orks. - T o systematically re vie w and cate gorize e xisting semantic interoper ability processing strate gies based on recent research contrib utions. - T o identify and analyze the major challenges and limitations f aced by IoT and Industry 4.0 systems due to insuf cient semantic interoperability . - T o discuss open research challenges and future research directions for enhancing semantic interoperability in IoT -based Industry 4.0 ecosystems. 2. RESEARCH METHOD This study emplo ys a semi-systematic lit erature re vie w m ethodology to in v estig ate semantic int er - operability within the conte xt of Industry 4.0 and the IoT . A semi-systematic re vie w is appropriate for con- ceptually broad and emer ging elds where research outcomes are heterogeneous and quantitati v e data may be limited [16]. This approach aims to identify , analyze, and synthes ize conceptually signicant patterns within the e xisting literature through meta-narrati v e synthesis. It w as chosen for this research as it enables a concise, conte xtually rele v ant, and criti cal o v ervie w of the state of kno wledge on semantic interoperability in IoT for Industry 4.0. The re vie w process follo ws the six-step frame w ork proposed by T emplier and P are, which includes the formulation of research questions, literature search, study screening, quality appraisal, data e xtraction, and syn- thesis as illustrated in Figure 1 [17]. The research questions (RQ1–RQ4) were designed to e xplore denitions, e xisting approaches, challenges, and future directions related to semantic interoperability . In the second phase, a comprehensi v e literature search w as conducted across multiple academic databases including Scopus, W eb of Science, ScienceDirect, DO AJ, and Google Scholar . K e yw ords such as “semantic interoperability , “IoT , “In- dustry 4.0, “ontology , and “semantic frame w orks” were used in v arious combinations with Boolean operators to rene the search results. Bot h published and unpublished articles were considered to capture a comprehen- si v e picture of the current research landscape. Additionally , interoperability-related concepts in adjacent areas, such as fog computing and industrial automation, were also e xplored to conte xtualize the ndings. Figure 1. Implemented research methodology steps During the third phase, a screening procedure w as implemented based on predened inclusion criter ia, such as selected studies be published between 2015 and 2024 and rele v ance to interoperability denitions, conceptual models, technological frame w orks, and application conte xts within industrial IoT systems. Studies were e xcluded if the y were non-Engli sh, lack ed rele v ance to semantic aspects, or were duplicates. Studies were initially screened by title and abstract, follo wed by full-te xt screening. The fourth phase in v olv ed a quality appraisal of the selected studies, assessing thei r research design, methodology , and ndings for academic rigor and rele v ance. Semantic inter oper ability in IoT for Industry 4.0: Re vie w , taxonomy ... (De vamekalai Na gasundar am) Evaluation Warning : The document was created with Spire.PDF for Python.
912 ISSN: 2252-8776 In the fth phase, data e xtraction w as conducted on the nalized set of studies. K e y information such as denitions, models, interoperability dimensions, technologies, and challenges w as systematically recorded. This process identied the foundational elements of semantic interoperability rele v ant to IoT and Industry 4.0. The nal phase in v olv ed data analysis and synthesis through content analysis techniques commonly applied in narrati v e re vie ws [16]. This f acilitated the identication of themes, trends, and challenges in the liter - ature, forming the basis for the taxonomy , challenges, and future research directions proposed in this paper .The nal dataset included 70 peer -re vie wed articles, with emphasis on recent contrib utions from 2020 to 2024. These studies span multiple domains including smart manuf acturing, healthcare, smart cities, and industrial automation, ensuring broad co v erage of semantic interoperability challenges and solutions. While the semi-systematic approach pro vides a rich conceptual o v ervie w , it may not capture all quan- titati v e metrics or unpublished industrial implementations. Additionally , the r eliance on English-language sources may e xclude rele v ant re gional studies. Despite these limitations, the methodology of fers a rob ust foun- dation for understanding the current landscape and guiding future research in semantic interoperability for Industry 4.0. 3. RESUL TS AND DISCUSSION 3.1. IoT inter operability This section addresses the rst research question by pro viding an o v ervie w of interoperability within the IoT ecosystem, particularly in the conte xt of Industry 4.0. In IoT systems, interoperability refers to the ability of heterogeneous de vices, platforms, and applications de v eloped by dif ferent manuf acturers or v endors to seamlessly communicate, e xchange, and utilize data wi thin a unied en vironment [18]. It ensures that de vices operating on distinct hardw are architectures, softw are protocols, and communication s tandards can ef fecti v ely collaborate and deli v er inte grated services [19], [20]. Achie ving interoperability in industrial IoT systems requires the adoption of standardized communica- tion protocols, data formats, and inte grati on frame w orks [21]. Common lightweight communication protocols such as mess age queuing telemetry transport (MQTT), constrained application protocol (CoAP), and HTTP are widely emplo yed to enable reliable data e xchange among resource-constrained IoT de vices [22]. Simi- larly , standardized data serial ization formats lik e JSON, XML, and Y AML f acilitate syntactic compatibility and simplify data processing across disparate systems [23]. Be yond de vice-to-de vice communication, interoperability also e xtends to the inte gration of di v erse subsystems, including edge de vices, cloud platforms, data analytics tools, and enterprise applications [10]. This necessitates the use of inte gration frame w orks, middle w are, and standardized APIs that allo w seamless data e xchange and operational coordination across heterogeneous infrastructures. Interoperability is essential for realizing the full potential of IoT -based Industry 4.0 en vironments, as it enables di v erse systems to cooper - ate and deli v er cohesi v e, scalable, and adapti v e industrial solutions. Figure 2 illustrates these four fundamen- tal dimensions of interoperability in IoT systems, highlighting the relationships and inte gration requirements among them. Figure 2. The dimensions of interoperability [24] Interoperability within the IoT conte xt encompasses multiple dimensions, each addressing a di stinct aspect of inte gration. According to Santos et al. [25], these include technical, syntactic, semantic, and or - g anizational interoperability . A clear understanding of these interoperability types is crucial for the ef fecti v e implementation and management of interoperable IoT systems. The four primary dimensions are summarized as follo ws, Int J Inf & Commun T echnol, V ol. 15, No. 2, June 2026: 909–924 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Inf & Commun T echnol ISSN: 2252-8776 913 - T echnical Interoperability: The ability of de vices, systems, and applications to communicate and e xchange data using common netw orking protocols, communication standards, and data serialization formats. It estab- lishes the foundational connecti vity necessary for de vice-le v el inte gration in IoT en vironments [26]. - Syntactic Interoperability: The capacity of disparate systems to e xchange data in a structured and recogniz- able format, ensuring that the transmitted data can be correctly parsed and understood by recei ving systems. This is typically achie v ed through standardized data formats such as JSON, XML, and RDF [27]. - Semantic Interoperability: The ability of systems and applications to interpret the meaning of e xchanged data consistently and meaningfully . Semantic interoperability ensures that data semantics are preserv ed across heterogeneous systems, enabling accurate and conte xt-a w are information e xchange [28]. - Or g anizational Interoperability: The capability of dif ferent or g anizations, b usiness processes, and go v er - nance structures to ef fecti v ely collaborate and e xchange information across IoT systems, supported by com- patible policies, standards, and b usiness objecti v es [29]. In summary , achie ving comprehensi v e interoperability across these dimensions is crucial for ensuring the seamless inte gration and ef cient operation of industrial IoT systems. It enables di v erse de vices, plat- forms, and or g anizations to cooperate in real time, f acilitating scalable, intelligent, and automated Industry 4.0 en vironments. The subsequent section will specically e xamine semantic interoperability , its technological enablers, and its pi v otal role in o v ercoming inte gration barriers within IoT -based industrial systems. 3.2. T axonomy of semantic inter operability in IoT f or Industry 4.0 Semantic interoperability is a crucial component in Industry 4.0 IoT ecosystems, as it ensures that de vices, systems, and applications can e xchange, interpret, and process data with a shared, unambiguous un- derstanding of its meaning [30]. It enables heterogeneous de vices to interact autonomously and allo ws ap- plications to le v erage data from di v erse sources without ambiguity or the need for human interv ention [28], [31], [32]. Importantly , semantic interoperability e xtends be yond merely dening information models or align- ing data transport formats; it in v olv es the consistent and meaningful interpretation of e xchanged data across distrib uted systems and kno wledge frame w orks. In recent years, signicant adv ancements ha v e emer ged in semantic technologies tailored for Industry 4.0 and related domains such as healthcare and smart cities [33]-[35]. F or e xample, Elkhodr et al. [36] pro- posed a blockchain-inte grated semantic IoT middle w are that le v erages ontology-po wered conte xt a w areness and secure data e xchange, addressing both semantic alignment and trust in healthcare IoT deplo yments. Ad- ditionally , NGSI-LD an ETSI-standardized information model and API has been widely adopted across smart industry and digital twin projects, enabling conte xtualized semantic data interchange between platforms [37]. Among widely adopted standards is the semantic sensor netw ork ( SSN) ontology , which pro vides a formal, machine-interpretable frame w ork for describing sensors, observ ations, and related metadata [38]. Lik e wise, the open platform communications unied architecture (OPC U A) has been recognized as a k e y Industry 4.0 standard, of fering a unied data model and service set for seamless, platform-independent data e xchange across industrial systems [39]. These ef forts highlight the importance of semantic ontologies in enabling standardized and scalable industrial data ecosystems. Gi v en the comple xity and di v ersity of industrial en vironments, achie ving semantic interoperability requires a multidimensional approach that considers v arious architectural, technological, and modeling aspects [40]. T o systematically analyze the current l andscape and guide future research, a taxonomy comprising v e k e y dimensions proposed in this section: netw ork model, ontology , middle w are, data model, and information model. These dimensions, as illustrated in Figure 3, were deri v ed from a synthesis of recent literature and reect the most inuential f actors shaping semantic interoperability frame w orks in Industry 4.0 [41]. The netw ork model dimension cate gorizes semantic interoperability solutions based on their deplo y- ment architecture. F og-based models perform semantic process ing at the edge of the netw ork, closer to data sources [42]. These models are particularly suited for latenc y-sensiti v e applications, such as real-time mon- itoring and control in industrial automation, as the y of fer reduced data transmission delays and impro v ed re- sponsi v eness [43]. Cloud-based models centralize semantic operations in cloud infrastructures, beneting from scalable computing resources and supporting lar ge-scale data aggre g ation and reasoning tasks [43]. Ho we v er , the y may introduce latenc y and bandwidth o v erhead, making them less suitable for time-critical applications. W eb-based models le v erage web technologies and standards to enable semantic data e xchange across dis- trib uted systems, of fering adv antages for interoperability across or g anizational boundaries and inte gration with e xternal services [44]. Semantic inter oper ability in IoT for Industry 4.0: Re vie w , taxonomy ... (De vamekalai Na gasundar am) Evaluation Warning : The document was created with Spire.PDF for Python.
914 ISSN: 2252-8776 Figure 3. T axonomy of semantic interoperability [24] Ontologies serv e as the backbone of semantic interoperability by pro viding formal repres entations of domain kno wledge [45]. The taxonomy distinguishes between lightweight, hea vyweight, and domain- specic ontologies. Light weight ontologies are designed for simplicity and ef cienc y , making them suitable for resource-constrained de vices such as sensors and embedded systems [45]. Hea vyweight ontologies of fer rich semantic e xpress i v eness and support comple x reasoning, typically used in centralized systems or cloud en vironments where computational resources are ab undant [46]. Domain-specic ontologies are tailored to particular industrial sectors, such as manuf acturing or healthcare, capturing specialized terminology and rela- tionships that enhance semantic precision and conte xtual rele v ance. This classication reects the trade-of f between semantic richness and system performance, where lightweight ontologies enable f ast processing b ut may lack depth, while hea vyweight and domain-specic ontologies pro vide detailed semantic co v erage at the cost of increased comple xity . Middle w are plays a pi v otal role in managing communication and semantic inte gration between di v erse IoT components [47]. The taxonomy includes distrib uted and centralized middle w are architectures. Distrib uted middle w are decentralizes semantic services across multiple nodes, enhancing scalability , f ault tolerance, and e xibility , which is particularly suitable for lar ge and dynamic industrial en vironments [48]. Centralized mid- dle w are consolidates semantic processing in a single location, simplifying management and deplo yment b ut potentially limiting scalability and resilience [49]. The distinction between these architectures is based on design choices that directly impact system performance, maintainability , and adaptability [48]. Distrib uted middle w are is increasingly f a v ored in Industry 4.0 due to its alignment with decentralized and autonomous system requirements. The data model dime n s ion addresses ho w semantic data is structured and managed within IoT s ystems. It includes dynamic, static, and real-time models. Dynamic models support schema e v olution and accommodate changes in data structures o v er time, essential for en vironments where de vices and services are frequently updated or recongured [50]. Static models rely on x ed schemas and predened data structures, of fering simplicity b ut limited e xibility [15]. Real-time models enable immediate semantic interpretation of streaming data, which is critical for time-sensiti v e applications such as predicti v e maintenance and real-time analytics [15]. This classication is justied by the need to balance e xibility , performance, and comple xity in data handling, with dynamic and real-time models being particularly rele v ant for Industry 4.0 [51]. Information models dene the formal languages and standards used to represent semantic data. The taxonomy includes resource description frame w ork (RDF), RDF schema (RDF-S), and web ontology language (O WL) [52]. RDF pro vides a foundational model for representing information about resources in the semantic web . RDF-S e xtends RDF by of fering basic constructs for describing groups of related resources and their properties [52]. O WL of fers adv anced capabilities for dening and reasoning o v er comple x ontologies, sup- porting higher le v els of semantic e xpressi v eness [53]. These s tandards are widely adopted in semantic web technologies and pro vide the syntactic and semantic foundation for interoperability . The classication reects Int J Inf & Commun T echnol, V ol. 15, No. 2, June 2026: 909–924 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Inf & Commun T echnol ISSN: 2252-8776 915 the progression from basic data repres entation to more adv anced semantic modeling and reasoning capabilities, allo wing systems to choose the appropriate le v el of e xpressi v eness based on their requirements [53]. The taxonomy presented in Figure 3 of fe rs a comprehensi v e frame w ork for understanding the s truc- tural and functional components of semantic interoperability in Industry 4.0. Each c lassication dimension w as selected based on its pre v alence in the literature and its practical rele v ance to industrial IoT deplo yments. By or g anizing the landscape into these v e cate gories, the taxonomy f acilitates comparati v e analysis, high- lights e xisting g aps, and supports the de v elopment of rob ust, scalable, and adapti v e semantic interoperability solutions. In addition to these, adv anced semantic technologies, such as articial intelligence (AI), m achine learning, and natural language processing (NLP) ha v e g ained traction for automating data mapping, ontology alignment, and semantic annotation processes in IoT en vironments [54], [55]. F or instance, Linardatos et al. [56] applied machine learning to automate the classication of sensor data from heterogeneous sources, impro ving semantic alignment and enhancing interoperability in Industry 4.0 conte xts. Another recent study proposed a frame w ork for automatically detecting and classifying data streams within manuf acturing processes to impro v e operational ef cienc y and semantic inte gration in production systems [57]. Cross-platform and cross-domain interoperability are equally vital for Industrial IoT applica tions, particularly in scenarios requiring data e xchange between i ndependent systems or or g anizations [58]. Further ef forts ha v e introduced technologies and tools for enhancing cross-domain semantic interoperabili ty . F or e x- ample, Abb uru [59] proposed a method for inte grating multi-source IoT data by combining ontologies with machine learning techniques for automated data mapping and con v ersion. Similarly , Da vies and Fisher [60] demonstrated ho w embedded semantic models and annotations within industrial applications impro v e IoT data consistenc y , scalability , and operational ef cienc y . Balakrishna et al. [61] highlighted the importance of se- mantic models for optimizing industrial IoT applications’ scalability and sustainability in Industry 4.0. Despite these adv ancements, seamless, scalable, and dynamic semantic interoperability remains an open research challenge. As recent studies emphasize, the lack of uni v ersally adopted semant ic frame w orks and the limited maturity of real-time semantic al ignment mechanisms continue to hinder the full inte gration of heterogeneous de vices, platforms, and s ervices in high-frequenc y , data-intensi v e Industry 4.0 en vironments [62]. Ongoing research into semantic data annotation, reasoning, disco v ery , and visualization is critical to addressing these barriers and enabling truly autonomous, interoperable industrial ecosystems. 3.3. Obstacles to achie v e semantic inter operability in Industry 4.0 Achie ving semantic interoperability in Industry 4.0 IoT en vironments presents se v eral persistent chal- lenges due to the sheer di v ersity of de vices, communication protocols, data formats, and operational conte xts [59]. Numerous studies ha v e identied critical obstacles that hinder seamless semantic interoperability in het- erogeneous industrial systems [63]-[65]. One of the primary challenges lies in the incompatibility of communication protocols and data for - mats across de vices and platforms. Although standardized protocols such as MQTT , CoAP , and OPC U A ha v e g ained traction, the lack of uni v ersally accepted semantic data models and ontol og i es limits interoperability and impedes cross-platform inte gration [41], [66]. Conse q ue n t ly , data generated by di v erse IoT de vices of- ten remains conned within isolated silos, complicating data aggre g ation, semantic alignment, and real-time analytics in Industry 4.0 en vironments [67]. De vice heterogeneity represents another signicant barrier , as industrial IoT systems typically com- prise de vices from mult iple manuf acturers, each emplo ying di stinct data models, terminologies, and conte xtual interpretations [68]. This inconsistenc y in semantic s tructures leads to dif culties in achie ving a common un- derstanding of e xchanged data, thereby af fecting system interoperability and inte gration at the semantic layer . The scale and v ariety of data produced by IoT de vices in Industry 4.0 applications further e xacerbate interoperability challenges. The enormous v olume of heterogeneous, real-time, and unstructured data demands ef cient semantic data modeling, kno wledge management, and inte gration techniques capable of supporting dynamic, scalable, and conte xt-a w are interoperability solutions [69], [70]. Pri v ac y and security concerns also constitute critical obstacles to semantic interoperability . Industri al IoT systems frequently handle sensiti v e operational, or g anizational, and personal data, necessitating rob ust pri v ac y-preserving mechanisms and secure semantic data e xchange protocols [71]. W ithout adequate security models and access control mec hanisms inte grated into semantic interoperability frame w o r ks, data condential- ity , inte grity , and trust cannot be ensured in interconnected industrial ecosystems. Semantic inter oper ability in IoT for Industry 4.0: Re vie w , taxonomy ... (De vamekalai Na gasundar am) Evaluation Warning : The document was created with Spire.PDF for Python.
916 ISSN: 2252-8776 T o address these challenges, a range of approaches ha v e been proposed, including the adoption of ontology-based frame w orks, semantic web technologies, and machine learning-assisted semantic mapping techniques [32], [69], [72], [73]. These solutions aim to enhance semantic compatibility among heterogeneous platforms, automate data mapping and translation processes, and establish common kno wledge representation frame w orks to f acilitate seamless data e xchange, inte gration, and reasoning within Industry 4.0 en vironments. In summary , while notable progress has been made in de v eloping semantic interoperability frame- w orks and tools, achie ving scalable, secure, and dynamic semantic inte gration across heterogeneous industrial IoT systems remains an unresolv ed research challenge. Addressing these barriers requires the continued ad- v ancement of ontology engineering, real-time semantic annotation techniques, and AI-dri v en interoperability solutions tailored to the demands of Industry 4.0. 3.4. Challenges caused by shortcomings of semantic inter operability in IoT f or Industry 4.0 Se v eral operati o na l and strat e gic challenges within Industry 4.0 IoT ecosystems ha v e been identied as direct consequences of insuf cient semantic interoperability [74]. These shortcomings hinder the scalability , ef cienc y , accessibilit y , and economic viability of industrial IoT deplo yments. The principal challenges are outlined as follo ws: - Restricted scalability , The inte gration of ne w IoT syst ems, de vices, and applications at scale is signicantly constrained when semantic interoperability is lacking [75]. The absence of standardized semantic frame- w orks leads to compatibility issues, making it dif cult to incorporate adv anced or heterogeneous de vices into e xisting systems without e xtensi v e custom inte gration ef forts [68]. This limitation ulti mately restricts the scalability and e xibility of Industry 4.0 infrastructures. - Inef cient data storage and resource utilization: Industrial IoT systems generate v ast v olumes of hetero- geneous data from distrib uted de vices. W ithout adequate semantic interoperability , ef fecti v e data sharing across applications and platforms becomes dif cult, resulting in redundant or siloed storage of o v erlapping data [66]. This inef cienc y increases storage costs and complicates lar ge-scale data management in Industry 4.0 en vironments. - V endor lock-in: A lack of semantic interoperabil ity forces industries to adopt proprietary IoT systems and de vices from single v endors, as inte gration with alternati v e systems is often comple x and costly [58]. This v endor dependenc y restricts system e xibility , hinders the adoption of competiti v e technologies, and im- pedes long-term inno v ation by creating monopolistic tendencies within the industrial IoT mark et. - Reduced system accessibility and Data Sharing: Interoperability limitations result in closed, siloed IoT ecosystems where data and services cannot be easily accessed or shared across platforms and applications [76]. This restrict ed accessibility diminishes the potential for inte grated, cross-or g anizational collaboration, limiting the operational and strate gic benets of Industry 4.0 architectures. - T echnological uncertainty and instability: The absence of unied semantic interoperability frame w orks increases the risk of technological fragmentation, where v endors f ail to deli v er agreed-upon services or maintain consistent functionality across de vices [77]. Industrial operators are then forced to adopt unreliable or unstable solutions due to incompatibility constraints, undermining operational continuity and trust in IoT systems. - Increased operational costs: The cost of deplo ying and maintaining Industry 4.0 IoT ecosystems escalates in the absence of semantic interoperability [65]. Industries are frequently unable to adopt more af ford- able, adv anced IoT solutions without fully replacing e xisting incompatible systems. This lack of modular upgradeability dri v es higher operational e xpenses and reduces the economic feasibility of long-term IoT deplo yments. In summary , semantic interoperability decienci es introduce substantial barriers to the scalability , ef - cienc y , and cost-ef fecti v eness of Industry 4.0 IoT systems. Addressing these challenges is essential for enabling e xible, scalable, and inte grated industrial ecosystems capable of supporting dynamic, data-dri v en operations. The follo wing sections re vie w current solutions and propose future research directions for o v ercoming these limitations. 3.5. Recent r esear ch eff orts to ward achie ving semantic inter operability Semantic interoperability has emer ged as a crucial component of IoT frame w orks for Industry 4.0 applications [78]. V arious s tudies and research projects ha v e proposed frame w orks, ontologies, and distrib uted architectures to address semantic interoperability challenges within heterogeneous IoT ecosystems [20], [61]. Int J Inf & Commun T echnol, V ol. 15, No. 2, June 2026: 909–924 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Inf & Commun T echnol ISSN: 2252-8776 917 Ontology-based models, fog computing-assisted semantic architectures, and lightweight semantic frame w orks are among the widely e xplored approaches. Iong and Smys [79] proposed a fog-assisted semantic frame w ork designed to enhance interoperabil ity among IoT de vices by inte grating fog computing principles with semantic technologies. The frame w ork in- troduces a distrib uted computing infrastructure where fog nodes perform localized data aggre g ation, ltering, modeling, and semantic annotation before forw arding processed data to cloud serv ers for archi v al and adv anced analytics in Figure 4. The semantic model within this frame w ork utilizes standardized representations based on semantic web technologies such as RDF and O WL, f acilitating seamless data e xchange and interoperabil- ity across heterogeneous de vices. Additionally , Iong and Smys proposed data prioritization algorithms within the fog layer to optimize data transmission ef cienc y and reduce service delays, thereby supporting scalable, interoperable Industry 4.0 systems. Figure 4. F og-assisted semantic frame w ork Rahman and Hussain [80] introduced a lightweight ontology model (LiO-IoT) to support sema ntic interoperability for commonly encountered IoT components such as sensors, actuators, and radio frequenc y identication (RFID) systems. The ontology focuses on minimizing comple xity by adopting a simplied se- mantic representation, impro ving processing ef cienc y in constrained IoT en vironments. Ho we v er , the pro- posed model lacks dynamic semantic adaptability , a limitation subsequently addressed in Rahman’ s later w ork [73], which introduced a li ghtweight dynamic ontology frame w ork. This dynamic ontology inte grates machine learning techniques for automatically identifying and incorporating ne w attrib utes and concepts into the ontol- ogy , f acilitati n g real-time semantic adaptation. The frame w ork emplo ys clustering algorithms to detect no v el patterns within data streams, though the authors ackno wledged that clustering-induced delays may impact sys- tem response times in time-sensiti v e industrial applications. Further e xtending semantic interoperability solutions, Rahman and Hussain [81] proposed a fog-based semantic frame w ork that migrates semantic processing tasks traditionally handled at the cloud le v el to dis- trib uted fog nodes. This hierarchical fog computing architecture consists of Le v el-2 (L2-F og) nodes respon- sible for initial data collection, ltering, and aggre g ation, and Le v el-1 (L1-F og) nodes task ed with higher - le v el semantic modeling, annotation, and decision-making in Figure 5. Semantic annotation is achie v ed using lightweight O WL-based ontologies managed via a lightweight middle w are layer . By shifting semantic reason- ing and data processing closer to data sources, this frame w ork reduces netw ork utilization, ener gy consumption, and service latenc y while impro ving interoperability across heterogeneous IoT de vices. Ho we v er , the frame- w ork relies on static ontologies, limiting its ability to dynamically accommodate emer ging de vices or e v olving semantic conte xts. Semantic inter oper ability in IoT for Industry 4.0: Re vie w , taxonomy ... (De vamekalai Na gasundar am) Evaluation Warning : The document was created with Spire.PDF for Python.
918 ISSN: 2252-8776 Figure 5. The frame w ork of the suggested model Gyrard and Ser rano [82] proposed a unied semantic engine for IoT and smart city applications that inte grates semantic web technologies, big data analytics, and IoT middle w are. The engine comprises three layers: a data layer for collecting and preprocessing sensor data, a semantic layer for standardizing and an- notating data using ontologies, and an application layer for de v eloping conte xt-a w are services. The authors v alidated their approach through a smart parking use case, demonst rating the engine’ s ability to process and analyze real-time sensor data, enabling intelligent parking management decisions. Although focused on smart city deplo yments, the engine’ s scalable, ontology-dri v en architecture of fers v aluable insights for Industry 4.0 semantic interoperability frame w orks. Collecti v ely , thes e studies highlight the di v erse m ethodologies proposed to address semantic inter - operability challenges in industrial IoT en vironments. T able 1 sho ws the capabilities and limitation analysis of the related w ork discussed. Although ontology-based models, fog-assisted frame w orks, and lightweight dy- namic ontologies ha v e adv anced interoperability capabilities, limitations persist in achie ving real-time semantic adaptability , dynamic ontology generation, and standardized cross- p l atform inte gration. Continued research is necessary to de v elop scalable, secure, and dynamic s emantic interoperability frame w orks capable of supporting the comple x, data-intensi v e requirements of Industry 4.0 applications. T able 1. Analysis of related w ork Related w ork research reference Attrib utes [79] [80] [73] [81] [82] Support real time No No Y es No No Dynamic interoperable No No Y es No No Suitable for small scale Y es Y es No Y es No Suitable for lar ge scale Y es No Y es Y es Y es Latenc y Medium High High Medium High Ener gy consumption Medium Medium High Medium High Netw ork usage Medium High Medium Medium High Deplo yment cost High Lo w High High Medium F og based Y es No No Y es No 3.6. Comparati v e analysis and discussion Semantic interoperability within IoT ecosystems is a cornerstone for achie ving seamles s data e x- change and inte gration in Indust ry 4.0 en vironments. While earlier studies ha v e e xpl o r ed the impact of seman- tic technologies such as ontologies, middle w are, and semantic web services, the y ha v e not e xplicitly addressed the inuence of dynamic semantic alignment and cross-domain adaptability in real-time industrial conte xts. This study in v estig ated the classication and ef fecti v eness of semantic interoperability frame w orks, re v ealing that the lack of uni v ersally accepted semantic models and real-time adaptability remains a signicant g ap in e xisting research. Int J Inf & Commun T echnol, V ol. 15, No. 2, June 2026: 909–924 Evaluation Warning : The document was created with Spire.PDF for Python.