The cluster analysis in the aluminium industry with K-means method: an application for Bahrain

Qahtani, Haitham Al; Sankar, Jayendira P. Article The cluster analysis in the aluminium industry with Kmeans method: an application for Bahrain Cogent Business & Management Provided in Cooperation with: Taylor & Francis Group Suggested Citation: Qahtani, Haitham Al; Sankar, Jayendira P. (2024) : The cluster analysis in the aluminium industry with K-means method: an application for Bahrain, Cogent Business & Management, ISSN 2331-1975, Taylor & Francis, Abingdon, Vol. 11, Iss. 1, pp. 1-19, https://doi.org/10.1080/23311975.2024.2361475 This Version is available at: https://hdl.handle.net/10419/326310 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Cogent Business & Management ISSN: 2331-1975 (Online) Journal homepage: www.tandfonline.com/journals/oabm20 The cluster analysis in the aluminium industry with K-means method: an application for Bahrain Haitham Al Qahtani & Jayendira P. Sankar To cite this article: Haitham Al Qahtani & Jayendira P. Sankar (2024) The cluster analysis in the aluminium industry with K-means method: an application for Bahrain, Cogent Business & Management, 11:1, 2361475, DOI: 10.1080/23311975.2024.2361475 To link to this article: https://doi.org/10.1080/23311975.2024.2361475 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published online: 04 Jun 2024. Submit your article to this journal Article views: 953 View related articles View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oabm20 OperatiOns ManageMent | research article Cogent Business & ManageMent 2024, VoL. 11, no. 1, 2361475 The cluster analysis in the aluminium industry with K-means method: an application for Bahrain haitham al Qahtania and Jayendira p. sankarb auniversity of technology Bahrain, salmabad, Bahrain; bCollege of administrative and Financial sciences, university of technology Bahrain, salmabad, Bahrain ABSTRACT this study examines the utilization of the K-means clustering method to analyze Bahrain’s aluminum industry. in addition, this study emphasizes the importance of clustering in understanding productivity, quality, and competitiveness within the sector. Data collection involved rigorous cleaning of diverse sources to ensure accuracy. By employing the K-means algorithm, this study successfully identified distinct clusters within the dataset, offering insights into industry dynamics. in addition, it proposes a roadmap for cluster development, providing actionable recommendations for stakeholders to enhance competitiveness and sustainability. Overall, this research advances knowledge of clustering techniques and informs strategic decision-making in Bahrain’s aluminum industry. 1. Introduction clustering analysis has become very important through the influential work of porter across ten different countries (Kim et al., 2023). in addition, Kalicanin and gavric (2014) mentioned that clustering aims to group objectives based on factors influencing competitive advantage. productivity is a crucial factor in any industry, with the optimized use of resources, such as labor, materials, and machines to generate goods and services effectively (Jain et al., 2016; prakash et al., 2017). Moreover, in the context of the aluminum industry in Bahrain, it is one of the key drivers of productivity and economic growth. importantly, clustering analysis studies in the aluminum industry in Bahrain will gain prominence in understanding the dynamics that affect a nation’s competitive strength. Furthermore, clustering involves grouping firms, innovation centers, and information hubs to foster collaboration and explore novel market strategies, methods, and products (Ye etal., 2021; Zhao etal., 2023). therefore, this study investigates the impact of clustering within Bahrain’s aluminum industry and explores its implications on productivity and product quality. © 2024 the author(s). Published by informa uK Limited, trading as taylor & Francis group CONTACT Jayendira P. sankar [email protected].bh College of administrative and Financial sciences, university of technology Bahrain, salmabad, Bahrain. https://doi.org/10.1080/23311975.2024.2361475 this is an open access article distributed under the terms of the Creative Commons attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. the terms on which this article has been published allow the posting of the accepted Manuscript in a repository by the author(s) or with their consent. ARTICLE HISTORY received 12 March 2024 revised 1 May 2024 accepted 22 May 2024 KEYWORDS Bahrain aluminium industry; K-means cluster analysis; gap analysis; assessment of linkages; road map; positioning REVIEWING EDITOR Diego corrales-garay, Universidad rey Juan carlos, spain SUBJECTS Business, Management, and accounting; production, Operations, and information Management; Management of technology and innovation; Development studies; economics and Development; industry and industrial studies 2 h. al Qahtani anD J. p. sanKar traditional manufacturing strategies involve multiple machines, and extensive labor leads to tooling expenditure, material handling expenditure, space requirements, and high production costs (heragu & ekren, 2009; pereira etal., 2019). in contrast, owing to heavy competition and rapid technological growth, manufacturing industries must adopt innovative strategies to control costs and increase overall productivity (huang, 2023; hwang & Kim, 2022). therefore, the central goal of this study is to use an analysis-based approach to minimize waste and enhance production quality and value addition. in addition, this study involved a meticulous examination of the current manufacturing process, identifying the challenges, and implementing strategies for streamline operations. Moreover, this research conducted a study in the aluminum industry in Bahrain, focusing on producing high-quality outputs. predominantly, the context of the aluminum industry in Bahrain emphasizes the relevance of clustering studies in collaborative environments and fostering economic connections. Furthermore, clustering in the aluminum industry involves grouping firms that produce similar products and share inputs, such as labor and technology, horizontally and vertically (Madsen et al., 2003; Yang & gu, 2021). Moreover, the critical characteristics of clusters in Bahrain’s aluminum industry include diverse participants, geographical proximity, and economic interdependence regarding economic status and activities. regarding the cluster models, Karaca (2018) and Vargas-hernández et al. (2020) state that the pure cluster model emphasizes the benefits of geographical proximity, the industrial complex model identifies stable relationships based on commercial ties, and the social network model focuses on economic activities and social integration of roles within networks and institutions. numerous studies have utilized clustering techniques across diverse fields, showcasing their effectiveness and versatility. Despite variations in objectives and datasets, clustering principles remain consistent in education (liu, 2022), healthcare (Yang et al., 2023), image processing (Mittal et al., 2022), social network analysis (sharma, 2011), financial services (cook & pandit, 2012), environmental science (Koo et al., 2023), manufacturing (Unterberger & Müller, 2021), transportation and logistics (rivera et al., 2016), agriculture (Dureti et al., 2023), and so on have all benefited from clustering in uncovering insights and aiding decision-making. interestingly, studies across different domains often adopt similar clustering approaches. holý et al. (2017) used hierarchical clustering in retail to analyze consumer preferences, while schulz et al. (2020) employed k-means clustering in biomedical research for disease subtype classification. Despite varied contexts, these studies share a common thread in leveraging clustering for data exploration and pattern recognition. Further, this suggests potential for clustering techniques in indices and policy evaluation, where they can extract insights from complex datasets encompassing economic, social, and policy-related variables. therefore, this K-mean clustering application framework study represents a novel approach to enhancing decision-making and policy formulation in these domains. the study of the clusters in the Bahrain aluminum industry offers economic advantages, such as supply chain cost, reducing transportation, creating a large market at a lower price, enhancing access to a qualified workforce, promoting cooperation, encouraging competition, and fostering specialization. Various clustering analysis methods have been proposed. the location quotient (lQ) method, input-output approaches, system dynamics approaches, dynamic models, and the porter Diamond Model have been utilized in clustering studies to assess economic development, competitiveness, and competition between clusters. clustering analysis of Bahrain’s aluminum industry is a multifaceted exploration of economic interdependence, geographical proximity, and collaborative advantages. this study aims to provide insights into how clustering can contribute to increased productivity, enhanced product quality, and overall competitiveness in the dynamic landscape of the aluminum industry in Bahrain. First, it aims to address the theoretical gap in clustering the aluminum industry in Bahrain using the K-means method, which has not been done before in any part of the research. second, there is a methodological gap in using the K-means method for specific clustering of the aluminum industry in Bahrain and its rationale. third, the practical gap can be filled with practical recommendations for policymakers to implement clustering initiatives effectively in Bahrain’s aluminum industry. Fourth, there is a contextual gap in no earlier studies within Bahrain’s broader economic development goals, policy frameworks, or industry dynamics. the remainder of this paper is organized as follows. section 2 reviews the relevant literature. section 3 describes the research methodology and section 4 presents the results and findings. cOgent BUsiness & ManageMent 3 section 5 presents the discussion and implications of this study. Finally, section 6 concludes the paper and presents the limitations and avenues for future research. 2. Review of literature clustering is a holistic approach that emphasizes collaboration, shared resources, and interconnectedness to enhance competitiveness (pereira etal., 2023; Zhao etal., 2023). the characteristics of industrial clusters include physical closeness, value creation through interconnected production, and a shared business environment (Kim et al., 2023). therefore, the dynamic interplay between cooperation and competition is pivotal for fostering innovation within industrial clusters (Bagherzadeh et al., 2022; König, 2023). the literature review was a thorough review of the literature between 2001 and 2023. to ensure the reliability of the literature, all articles were retrieved from the scopus database. a study of Brazilian industrial clustering on intellectual property (ip) usage in manufacturing organizations in developing countries highlighted the importance of footwear firms in combating counterfeit and pirated goods threats and the growing importance of ip protection in global competitiveness (cavalheiro & Brandao, 2017). Kadokawa (2013) revealed that the location decision of new manufacturing plants with advantages in Japan’s industrial clustering is a key location factor, particularly in high-tech industries. similarly, libaers and Meyer (2011) point out the impact of industrial clustering on small technology-based firms, with distinctions between serial and non-serial innovators, which have more effective implications for economic development. guo and guo (2011) exposed chinese industrial clusters through knowledge-spanning mechanisms that synthesize learning opportunities and absorptive capacity perspectives to understand technological learning behavior across cluster types, cognitive subgroups, and innovation types. chen (2011) revealed the competitive advantage of production systems, such as taiwan’s industrial clustering, highlighting the role of lead firms’ relational capabilities fostered by cluster embeddedness in effectively governing suppliers. additionally, park et al. (2009) mentioned that inter-industrial knowledge flows are crucial for developing knowledge clusters and a national innovation system of Korean industrial clusters using two types of knowledge flows, embodied and disembodied. clustering in the British broadcasting industry’s impact on firm growth and entry survival rates has significant positive clustering on firm growth and entry survival areas akin to high-tech industries, such as computing and biotechnology (cook et al., 2001). likewise, the connection between emotions and product shapes in automobiles, sofas, and kettles uses four fundamental dimensions: trend, emotion, complexity, potency, and significant shape features (ssFs) (hsiao & chen, 2006). in addition, the success of spinoffs in the global fashion design industry, their role in clustering design firms in selected cities, and insignificant effects from migration flows and localization economies (Wenting, 2008). correspondingly, to identify the sub-sectors of turkish textile patents, the study of patent count data and fuzzy-based clustering to detect technology trends, such as classic or dated, through k-means clustering (Dereli & Durmuşoğlu, 2009). a study on timely sales identification represents the hybrid model by merging the fuzzy neural network and k-means cluster, which increases the forecast accuracy in small regions. in addition, this study reiterates that the hybrid model is more effective than the alternative (chang et al., 2009). in addition, the risk identification and perception in fisheries systems in Faroes, iceland, greece, and the UK through clustering highlighted the supporting social science theories and the subjective nature of risk (tingley et al., 2010). another thing to remember is that clustering occupational accidents in the italian wood-processing industry revealed that K-means clustering identifies the standard sequence of events leading to accidents and devises preventive measures (palamara etal., 2011). similarly, a case-mix project in the netherlands on healthcare products for hospitals clustered the product structure with cost homogeneity (Westerdijk et al., 2012). a study by Jun and park (2013) on apple’s technological innovation by analyzing patent applications using clustering of patents identifies vacant technology domains and social network analysis for the future. Furthermore, clustering regional businesses in the northern spanish region for principle components analysis provided superior cluster solutions (argüelles et al., 2014). Boix et al. (2015) studied the spatial patterns of the location and co-location of clusters in 16 european countries and revealed a 4 h. al Qahtani anD J. p. sanKar highly clustered creative belt from southern england and southern germany. similarly, it brings out innovation from multiple sectors in Korean industrial convergence by clustering technology-driven new value generators, service-integrated social business generators, policy-driven environmental enhancers, and technology enhancers (geum et al., 2016). Zhou etal. (2017) showed insightful clustering of monthly household electricity consumption patterns in Jiangsu province in china using a fuzzy c-mean clustering model for energy efficiency strategies in the power industry. Fuzzy c-mean clustering was used to evaluate supplier sustainability in the plastic pipe industry; it offers managerial insights into outsourcing decisions and resilience strategies to minimize costs and maximize sustainability during disruptions (Jabbarzadeh et al., 2018). additionally, the combination of the length, recency, frequency, monetary, and periodicity (lrFMp) model and K-means clustering in iranian fintech companies is supported by marketing intelligence and strategic decision-making in the business-to-business (B2B) fintech industry (sheikh et al., 2019). likewise, using latent Dirichlet allocation and K-means clustering for university selection and visualizing cooperation networks on blockchain has brought significant challenges, such as selecting research teams, evaluating university competitors, and accessing technology (ran et al., 2020). in connection with the Fourth industrial revolution, finding the relationship between green economics and digital technology development in Ukraine through clustering analysis revealed the alignment of digitalization efforts with national innovation strategies and mitigating environmental risks (plantec etal., 2021). specifically, edina et al. (2021) revealed industry 4.0 investments and their impact on hungarian food companies’ business performance using K-means clustering to find process innovation problems in the increasing use of industry 4.0 tools. in addition, a study on the influential factors affecting china’s electricity industry emissions (ceei) using K-means clustering reveals the aiding ceei control and promotion of low-carbon transformation (he et al., 2022). correspondingly, industrial clusters in china’s new energy enterprises revealed enhanced profitability by reducing production inefficiency and new energy industry development (sun etal., 2022). clustering is vital in advancing sustainable development goals in healthcare, energy, transportation, and manufacturing (Oyewole & thopil, 2023). in addition, a study of china’s low-carbon development of the power sector for neutrality goals and carbon peaking using K-means clustering revealed emission reduction by aiding provincial-level emission reduction strategies for global green transformation (Wang et al., 2023). another thing to remember is a study by singhal et al. (2023), which categorizes online fashion consumers based on their perceptions and relationships with bands through social media using K-means cluster analysis, emphasizing a pioneering approach to understanding consumer behavior in online fashion. table 1 represents the overview of reviewed sources, including author with year, focus of study, methodology, and key findings. clustering emphasizes shared resources and collaboration, which fosters competitiveness through interconnected production and leveraging physical proximity in industrial clusters. additionally, the interplay between competition and cooperation drives innovation. Further, literature reviews spanning 2001–2023, notably on Japanese manufacturing and Brazilian ip usage, underscore the impact of clustering on global competitiveness. Moreover, studies on taiwanese and chinese clusters highlight knowledge exchange roles, whereas British broadcasting and fashion design showcase diverse applications. in addition, techniques, such as fuzzy clustering enhance energy efficiency and supplier sustainability analysis. Furthermore, in the context of industry 4.0, clustering aids in emission control and aligns with sustainability goals. crucially, clustering advances sectors, such as low-carbon transitions, shaping consumer behavior, energy, and healthcare. thus, no specific literature on clustering analysis in the aluminum industry using K-means in Bahrain shows a literature gap. therefore, the purpose of this study is to fill the literature gap and provide solid recommendations for the effective clustering of the aluminum industrial sector in Bahrain. 3. Methodology Many studies have used a methodology similar to that used in this study, but these data mining methods are still the most commonly used in the natural sciences. Furthermore, clustering algorithms were cOgent BUsiness & ManageMent 5 Table 1. overview of reviewed sources. author and year Focus of study Methodology Key findings argüelles et al. (2014)Clustering regional businesses in northern spain Principle components analysis superior cluster solutions for regional businesses Boix et al. (2015)spatial patterns of clusters in european countries geographic analysis identification of highly clustered creative belt Cavalheiro and Brandao (2017) iP usage in Brazilian manufacturing organizations Qualitative analysis importance of iP protection in combating counterfeit goods, enhancing global competitiveness Chang et al. (2009)Hybrid model for timely sales identification integration of fuzzy neural network and k-means cluster increased forecast accuracy in small regions Chen (2011)Competitive advantage of taiwan’s industrial clustering Case study analysis Role of lead firms’ relational capabilities in governing suppliers Cook et al. (2001)impact of clustering on firm growth in British broadcasting industry statistical analysis significant positive clustering effects on firm growth and entry survival rates Dereli and Durmuşoğlu (2009)identification of sub-sectors of turkish textile patents Fuzzy-based clustering Detection of technology trends through k-means clustering edina et al. (2021)impact of industry 4.0 investments on Hungarian food companies K-means clustering identification of process innovation problems geum et al. (2016)innovation in Korean industrial convergence Clustering of technology-driven value generators identification of innovation sectors in convergence guo and guo (2011)Knowledge-spanning mechanisms in Chinese industrial clusters synthesis of learning opportunities understanding technological learning behavior across cluster types He et al. (2022)Factors affecting China’s electricity industry emissions K-means clustering aid in emission control and low-carbon transformation Hsiao and Chen (2006)Connection between emotions and product shapes in various industries Conceptual analysis identifying fundamental dimensions and shape features Jabbarzadeh et al. (2018)evaluation of supplier sustainability in plastic pipe industry Fuzzy c-mean clustering Managerial insights for outsourcing decisions and resilience strategies Jun and Park (2013)analysis of apple’s technological innovation using patent clustering Patent analysis and social network analysis identification of vacant technology domains and future trends Kadokawa (2013)Location decision of new manufacturing plants in Japan’s industrial clusters Case study analysis Key location factor, particularly in high-tech industries Libaers and Meyer (2011)impact of industrial clustering on small technology-based firms Literature review Differential implications for economic development between serial and non-serial innovators oyewole and thopil (2023)Role of clustering in advancing sustainable development goals Conceptual analysis Contribution to sustainability in various sectors Palamara et al. (2011)Clustering occupational accidents in italian wood-processing industry K-means clustering identifying standard sequence of events leading to accidents Park et al. (2009)inter-industrial knowledge flows in Korean industrial clusters network analysis Crucial for developing knowledge clusters and national innovation systems Plantec et al. (2021)Relationship between green economics and digital technology in ukraine Clustering analysis alignment with national innovation strategies and environmental risk mitigation Ran et al. (2020)Clustering for university selection and cooperation networks on blockchain Latent Dirichlet allocation and K-means clustering Challenges in research team selection and technology access sheikh et al. (2019)Clustering of iranian fintech companies LRFMP model and K-means clustering support for strategic decision-making in B2B fintech industry singhal et al. (2023)Categorization of online fashion consumers based on social media perceptions K-means cluster analysis understanding consumer behavior in online fashion sun et al. (2022)Profitability enhancement in China’s new energy enterprises statistical analysis Reduction of production inefficiency and industry development tingley et al. (2010)Risk identification in fisheries systems through clustering Qualitative analysis Highlighting subjective nature of risk and supporting social science theories Wang et al. (2023)Low-carbon development of China’s power sector K-means clustering support for provincial-level emission reduction strategies Wenting (2008)Role of spinoffs in clustering design firms in global fashion industry Case study analysis insignificant effects from migration flows and localization economies (Continued) 6 h. al Qahtani anD J. p. sanKar used to group data with common properties. clustering is a well-known method used in economics to ensure the correctness of clustering according to the number of employees and local units, and the K-means clustering method was used in the original data mining. additionally, data relevant to the aluminum manufacturing industry in Bahrain on the geographical distribution of manufacturing units, workforce demographics, sales figures, and production quantities were collected. specifically, this study utilized reliable and accurate data from databases, industry reports, and government publications. the data were then cleaned to remove and fix incomplete, duplicate, incorrectly formatted, corrupted, and incorrect data because they were collected from multiple sources. Furthermore, this study used the classification of statistical regional units in Bahrain to define the geographical region for analysis and to concentrate on aluminum manufacturing activities in Bahrain. Moreover, the datasets were selected to highlight the uniqueness of Bahrain’s aluminum manufacturing sector, such as geographical distribution, market demand, and production volume. in addition, the parameters were fixed for the number of clusters (K value), and the distance metric was used to measure the similarity between the data points. the following steps were followed to ensure the correct clustering of the aluminum manufacturing sector in Bahrain. 1. Bahrain statistics for each region published by the institutions and the number of local units in the sector were obtained. 2. an algorithm was developed for the k-means method (data-mining method). 3. the solution to the written algorithm determines the clusters to be set up. author and year Focus of study Methodology Key findings Westerdijk et al. (2012)Case-mix project in the netherlands healthcare industry statistical analysis Clustered product structure with cost homogeneity Zhou et al. (2017)Clustering of household electricity consumption patterns Fuzzy c-mean clustering insights for energy efficiency strategies in the power industry Source: self-made by the author(s) based on the available data. Table 1. Continued. Figure 1. Five key steps of framework analysis. Source: self-made by the author(s) based on the available data. cOgent BUsiness & ManageMent 7 the Davies-Bouldin index (DBi) metric, introduced by Davies and Bouldin (1979), serves as a tool for evaluating clusters. it assesses the internal validity of clustering by analyzing how effectively it has been executed by computing various statistical features derived from the dataset. in essence, the DBi value should ideally approach zero or be close to non-negative values, indicating a higher quality of the obtained cluster and facilitating the judgment of its goodness. dxy x y ii i n ii , () = () = ∑ 1 2 - Figure 1 represents the five key steps of framework analysis with familiarization framework, indexing, charting, and interpretation. the study implemented strategies to optimize effectiveness and overcome potential limitations. Firstly, enhancing data collection methods is crucial, utilizing both traditional approaches and advanced data mining techniques to comprehensively analyze the aluminium manufacturing sector in Bahrain, focusing on aspects like geographical distribution, market demand, and production volume. secondly, implementing differential privacy offers a rigorous approach to safeguarding data privacy while enabling organizations to analyze sensitive data. additionally, ensuring the robustness of K-means clustering through methods, such as intra-cluster and inter-cluster similarity, cluster validation, and determining the optimal number of clusters (k) is essential for accurate partitioning of datasets. Moreover, mitigating bias in clustering analysis by employing techniques, such as K-means clustering, evaluation metrics, and euclidean distance helps ensure the accurate partitioning of datasets into distinct clusters. By incorporating multiple clustering algorithms and validation techniques, potential limitations are mitigated, enhancing the reliability of study outcomes and facilitating informed decision-making processes based on valuable insights extracted from the data. 3.1. Data Data was gathered using a combination of methodologies commonly employed in natural sciences research. specifically, data mining methods, particularly clustering algorithms, were utilized to analyze and group data with shared characteristics. the primary focus was on the aluminum manufacturing industry in Bahrain. Further, the dataset encompasses various facets of the aluminum manufacturing sector, including but not limited to the geographical distribution of manufacturing units, workforce demographics, sales figures, and production quantities. in addition, these variables were chosen to characterize Bahrain’s aluminum manufacturing landscape comprehensively. also, the data was sourced from diverse outlets, including databases, industry reports, and government publications. these repositories ensured the reliability and accuracy of the data used in the study. 4. Clustering analysis and results in this study, K-means clustering, proposed by MacQueen in 1967 (MacQueen, 1967), was employed to determine the effectiveness and efficiency of partitioning the datasets. in addition, the algorithm iteratively assigns each data point to one of k clusters based on the similarity of their attributes. this study continues until convergence when the clusters exhibit maximum intra-cluster and minimum inter-cluster similarities. second, data mining is critical to modern data analysis to extract valuable insights from large datasets. Furthermore, clustering analysis facilitates the grouping of objects with characteristics similar to those of a fundamental data-mining technique. hence, this study aims to identify patterns and relationships within datasets that provide valuable information for decision-making processes. therefore, this study utilized the K-means clustering method as a nonhierarchical approach to partition the dataset into distinct clusters. according to Kangallı et al. (2014), five methods are employed in stacker clustering: single link, full link, average link, ward method, and central method, all based on different principles. tan et al. (2022) used the divisor hierarchical method to form clusters by dividing the large clusters into small clusters based on their similarity. in addition, swarndepp and pandya (2016), in partition-type clustering, such as K-means, iteratively adjust cluster centers until points are closest, forming distinct clusters based on the 14 h. al Qahtani anD J. p. sanKar through a systematic approach involving stakeholders that can lead to the growth and competitiveness of Bahrain’s aluminum cluster, which scant the study by prebanić and Vukomanović (2023). Further, the study displayed the positioning of the aluminum cluster in Bahrain, including future product portfolios, market regions, and value chain coverage, including strategic positioning aligned with profitability, market intelligence, and regional focus, which scant the study by OecD (2019). this study provides insights into applying clustering techniques, particularly K-means, in various contexts, aiding decision-making processes and strategic planning for cluster development in Bahrain’s aluminum industry. Based on empirical and detailed research, this study proposes countermeasures for cluster analysis in the aluminum industry using the K-means method in Bahrain. in addition, rich resources and complex network structures in the cluster area can provide better resources and information for enterprises. First, to address the theoretical gap, this study provides actionable insights and recommendations that will inform the strategic decision-making process, promote innovation, and drive the sustainable growth of the aluminum sector in Bahrain. hence, the feasibility of comprehensive research on Bahrain’s aluminum sector is vital for identifying growth opportunities and fostering innovation by analyzing market trends, competitors, and regulations to pinpoint areas for innovation and growth, considering emerging technologies and sustainability. therefore, these tailored recommendations will promote sustainable growth and align with industry objectives, involving stakeholder collaboration. presenting findings and recommendations will facilitate collective strategy development. continuous monitoring and adaptation will drive growth and innovation in Bahrain’s aluminum sector. second, data collection involved gathering and clearing relevant metrics to address the methodological gap through clustering analysis, which was performed using a developed K-means algorithm evaluated through intra-cluster similarity and the Davies-Bouldin index. third, the study addressed the practical gap by analyzing factors, such as manufacturing distribution and sales, and aimed to offer insights for industry enhancement that will guide strategic decisions and support Bahrain’s economic growth. hence, combining sustainability and technology is reshaping Bahrain’s industries, driven by global initiatives like the sDgs through integrating eco-conscious practices to fulfill social responsibilities and bring cost savings and competitive advantages by technologies, such as ai, iot, blockchain, and 3D printing to enhance efficiency and customer experiences, offering significant competitive edges when combined with sustainable principles. also, embracing sustainability and technology can spur economic growth, attract investment, foster entrepreneurship, and create jobs, necessitating government policies and public-private partnerships for a prosperous future. Fourth, this study addressed the contextual gap by applying K-means clustering to understand Bahrain’s aluminum manufacturing sector by determining optimal cluster numbers, evaluating clustering quality, and examining linkages within the industry, aiming to inform strategic decision-making and investments. Finally, the study addressed the literature gap by exploring data mining methodologies, mainly clustering algorithms, in analyzing Bahrain’s aluminum sector. it also covered data collection, cleaning, K-means clustering, algorithm steps, and evaluation metrics, and assessed the industry cluster, proposing a development roadmap for strategic positioning and growth opportunities. 6. Conclusion this study thoroughly examines clustering analysis in Bahrain’s aluminum industry, utilizing the K-means method to address gaps in theory, methodology, practice, context, and literature. Furthermore, the study emphasizes the significance of clustering in boosting productivity, product quality, and competitiveness within the sector and fostering collaboration, innovation, and economic connections. additionally, the study achieved a DBi value of 0.918, indicating the clustering quality of the dataset. this lower DBi suggests better clustering, with more distinct clusters and denser packing within each cluster. also, by applying the K-means algorithm, the study successfully identified distinct clusters within the dataset, offering insights into the industry’s structure and dynamics and suggesting growth and strategic development opportunities. also, the study proposes a roadmap for cluster development, providing actionable recommendations for policymakers, industry stakeholders, and researchers to cOgent BUsiness & ManageMent 15 enhance competitiveness and sustainability. Overall, it contributes to existing knowledge by applying advanced clustering techniques and offers valuable insights for shaping strategic decisions and fostering sustainable development in Bahrain’s aluminum sector amid evolving technological and market landscapes. 6.1. Limitations and future research the study on Bahrain’s aluminium industry presents limitations and avenues for future research. First, this study’s focus on Bahrain’s aluminium industry restricts the generalizability of findings to broader regional or sectoral contexts. this limitation highlights the need for further research to explore clustering dynamics across diverse industries and regions. second, the absence of dynamic industry factors, such as technological advancements and global market trends in the analysis limits the study’s comprehensiveness. Future research should aim to incorporate these factors to provide a more nuanced understanding of industry evolution and clustering dynamics. third, data limitations and biases raise concerns about transparency and reliability. a thorough examination of data limitations is necessary to ensure the credibility of findings and interpretations. in future research. First, comparative analyses with industries or sectors in other countries or within Bahrain itself could yield valuable insights into universal clustering trends and variations. exploring similarities and differences across contexts can enrich understanding and inform policy decisions. second, longitudinal studies to examine the impact of external factors, such as economic conditions and trade policies, on industry clustering dynamics could provide a more comprehensive understanding. long-term analysis allows for the observation of trends and patterns over time, offering insights into the resilience and adaptability of industries. third, employing alternative clustering methods to complement existing approaches and enhance the accuracy and reliability of results. experimenting with different methodologies can offer new perspectives and improve the robustness of findings. Finally, enhancing data collection processes can contribute to the reliability and validity of research outcomes. employing advanced data collection techniques and ensuring data quality is essential for producing credible insights and conclusions. addressing these limitations and pursuing suggested future research directions will enable scholars to advance their understanding of industry clustering dynamics, ultimately leading to more robust and applicable findings with broader implications. Ethical approval ethical approval for this study was obtained from the ethics committee of the research center at the University of technology Bahrain. this study adhered to the ethical principles outlined in the irB-hsBs informed consent guidelines of the University of technology Bahrain. Consent for publication not applicable. Author contributions haQ designed the study and was responsible for the quality assurance of the study results. Jps was responsible for the quality assurance of the interpretation of the results. haQ was mainly responsible for analyzing the data for this manuscript, and Jps drafted the manuscript in close collaboration with haQ. Both authors participated in editing, reading, and approving the final manuscript. Disclosure statement no potential conflict of interest was reported by the author(s). 16 h. al Qahtani anD J. p. sanKar About the authors Dr. Haitham Al Qahtani, a highly educated professional, holds a phD in chemical engineering from the prestigious University of tufts in Boston, a Master of energy engineering from the esteemed University of leeds in the UK, and a Bachelor of petroleum engineering from the renowned University of southern california in the Usa. his academic journey has equipped him with a deep understanding of his field, which he now applies in his role as the Vice president of academic affairs at the University of technology Bahrain. his wealth of academic and professional experience is evident in his previous roles as a project leader during the establishment of the British college of Bahrain, group strategic planning & project Developer at the Yusuf Bin ahmed Kanoo group of companies, and Vice president for knowledge-based services at the economic Development Board. Dr. Jayendira P. Sankar received his phD from the University of Madras, india. he worked for ten years in india and in Bahrain for seven years. Overall, seventeen years of academic experience, senior Fellow hea-UK and academic Fellow cip D-UK with many publications in various high-impact journals and conferences. the current research interests include Development economics: economic growth, poverty, inequality, market failure, fiscal, economic and social conditions, healthcare, education, employment sector, industrial and social infrastructure; human resource Management: employee welfare, work-family balance, consumer behaviour, retail marketing, corporate social responsibility, sustainability, and teaching-learning and assessments. Funding no funding was received for conducting this study. ORCID haitham al Qahtani http://orcid.org/0009-0009-0817-4746 Jayendira p. sankar http://orcid.org/0000-0001-8435-2123 Data availability statement the data that support the findings of this study are available from the corresponding author Jps upon reasonable request. References argüelles, M., Benavides, c., & Fernández, i. (2014). a new approach to the identification of regional clusters: hierarchical clustering on principal components. Applied Economics, 46(21), 1–19. https://doi.org/10.1080/0003684 6.2014.904491 Bagherzadeh, M., ghaderi, M., & Fernandez, a. s. (2022). coopetition for innovation – the more, the better? an empirical study based on preference disaggregation analysis. European Journal of Operational Research, 297(2), 695–708. https://doi.org/10.1016/j.ejor.2021.06.010 Boix, r., hervás-Oliver, J. l., & De Miguel-Molina, B. (2015). Micro-geographies of creative industries clusters in europe: From hot spots to assemblages. Papers in Regional Science, 94(4), 753–773. https://doi.org/10.1111/ pirs.12094 cavalheiro, g. M. D. c., & Brandao, M. (2017). assessing the ip portfolio of industrial clusters: the case of the Brazilian footwear industry. Journal of Manufacturing Technology Management, 28(8), 994–1010. https://doi.org/10.1108/ JMtM-10-2016-0137 chang, p. c., liu, c. h., & Fan, c. Y. (2009). Data clustering and fuzzy neural network for sales forecasting: a case study in printed circuit board industry. Knowledge-Based Systems, 22(5), 344–355. https://doi.org/10.1016/j.knosys.2009.02.005 chen, l. c. (2011). the governance and evolution of local production networks in a cluster: the case of taiwan’s machine tool industry. GeoJournal, 76(6), 605–622. https://doi.org/10.1007/s10708-009-9317-2 cook, g. a. s., & pandit, n. r. (2012). clustering and the location of multinational enterprises: an exploration of financial services in london. in The regional economics of knowledge and talent: Local advantage in a global context (pp. 281–299). edward elgar. https://doi.org/10.4337/9781781953549.00019 cook, g. a. s., pandit, n. r., & swann, g. M. p. (2001). the dynamics of industrial clustering in British broadcasting. Information Economics and Policy, 13(3), 351–375. https://doi.org/10.1016/s0167-6245(01)00041-5 Davies, D. l., & Bouldin, D. W. (1979). a cluster separation measure. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1(2), 224–227. https://doi.org/10.1109/tpaMi.1979.4766909 cOgent BUsiness & ManageMent 17 Dereli, t., & Durmuşoğlu, a. (2009). classifying technology patents to identify trends: applying a fuzzy-based clustering approach in the turkish textile industry. Technology in Society, 31(3), 263–272. https://doi.org/10.1016/j.techsoc.2009.06.007 Du, M., & Wu, F. (2022). grid-based clustering using boundary detection. Entropy, 24(11), 1606. https://doi.org/10.3390/ e24111606 Dureti, g. g., tabe-Ojong, M. p., & Owusu-sekyere, e. (2023). the new normal? cluster farming and smallholder commercialization in ethiopia. Agricultural Economics, 54(6), 900–920. https://doi.org/10.1111/agec.12790 edina, e., József, p., Miklós, n., & Judit, O. (2021). the role of industry 4.0 technologies in the innovation activities of food manufacturing companies. Statisztikai Szemle, 99(10), 978–996. https://doi.org/10.20311/stat2021.10. hu0978 ergun, M., Uyguçgil, h., & atalik, Ö. (2020). creating a geodemographic classification model within geo-marketing: the case of eskişehir province. Bulletin of Geography. Socio-Economic Series, 47(47), 45–61. https://doi.org/10.2478/ bog-2020-0003 gergely, B., & Vargha, a. (2021). how to use model-based cluster analysis efficiently in person-oriented research. Journal for Person-Oriented Research, 7(1), 22–35. https://doi.org/10.17505/jpor.2021.23449 geum, Y., Kim, M. s., & lee, s. (2016). how industrial convergence happens: a taxonomical approach based on empirical evidences. Technological Forecasting and Social Change, 107, 112–120. https://doi.org/10.1016/j.techfore.2016.03.020 guo, B., & guo, J. J. (2011). patterns of technological learning within the knowledge systems of industrial clusters in emerging economies: evidence from china. Technovation, 31(2–3), 87–104. https://doi.org/10.1016/j.technovation.2010.10.006 hadhri, W., arvanitis, r., & M’henni, h. (2016). Determinants of innovation activities in small and open economies: the lebanese business sector. Journal of Innovation Economics & Management, 21(3), 77–107. https://doi.org/10.3917/ jie.021.0077 hanawalt, e., & rouse, W. (2017). assessing location attractiveness for manufacturing automobiles. Journal of Industrial Engineering and Management, 10(5), 817–852. https://doi.org/10.3926/jiem.2321 he, Y., Xing, Y., Zeng, X., Ji, Y., hou, h., Zhang, Y., & Zhu, Z. (2022). Factors influencing carbon emissions from china’s electricity industry: analysis using the combination of lMDi and K-means clustering. Environmental Impact Assessment Review, 93, 106724. https://doi.org/10.1016/j.eiar.2021.106724 heragu, s. s., & ekren, B. (2009). Materials handling system design. in Environmentally conscious materials handling (pp. 1–26). John Wiley & sons. https://doi.org/10.1002/9780470432730.ch1 holý, V., sokol, O., & Černý, M. (2017). clustering retail products based on customer behaviour. Applied Soft Computing, 60, 752–762. https://doi.org/10.1016/j.asoc.2017.02.004 hong, s. J., Kwon, i. W. g., & li, J. (2014). assessing the perception of supply chain risk and partnerships using importance-performance analysis model: a case study of sMes in china and Korea. Supply Chain Forum, 15(2), 110–125. https://doi.org/10.1080/16258312.2014.11517344 hsiao, K. a., & chen, l. l. (2006). Fundamental dimensions of affective responses to product shapes. International Journal of Industrial Ergonomics, 36(6), 553–564. https://doi.org/10.1016/j.ergon.2005.11.009 huang, X. (2023). the roles of competition on innovation efficiency and firm performance: evidence from the chinese manufacturing industry. European Research on Management and Business Economics, 29(1), 100201. https://doi. org/10.1016/j.iedeen.2022.100201 hwang, W. s., & Kim, h. s. (2022). Does the adoption of emerging technologies improve technical efficiency? evidence from Korean manufacturing sMes. Small Business Economics, 59(2), 627–643. https://doi.org/10.1007/ s11187-021-00554-w ikotun, a. M., ezugwu, a. e., abualigah, l., abuhaija, B., & heming, J. (2023). K-means clustering algorithms: a comprehensive review, variants analysis, and advances in the era of big data. Information Sciences, 622, 178–210. https://doi.org/10.1016/j.ins.2022.11.139 Jabbarzadeh, a., Fahimnia, B., & sabouhi, F. (2018). resilient and sustainable supply chain design: sustainability analysis under disruption risks. International Journal of Production Research, 56(17), 5945–5968. https://doi.org/10.1080 /00207543.2018.1461950 Jain, r., gupta, s., Meena, M. l., & Dangayach, g. s. (2016). Optimisation of labour productivity using work measurement techniques. International Journal of Productivity and Quality Management, 19(4), 485–510. https://doi. org/10.1504/iJpQM.2016.10000353 Jun, s., & park, s. s. (2013). examining technological innovation of apple using patent analysis. Industrial Management & Data Systems, 113(6), 890–907. https://doi.org/10.1108/iMDs-01-2013-0032 Kadokawa, K. (2013). a search for an industrial cluster in Japanese manufacturing sector: evidence from a location survey. GeoJournal, 78(1), 85–101. https://doi.org/10.1007/s10708-011-9433-7 Kalicanin, D., & gavric, O. (2014). the importance of clusters as drivers of competitive advantage of companies. Ekonomika Preduzeca, 62(3–4), 164–172. https://doi.org/10.5937/ekopre1404164K Kangallı, s. g., Uyar, U., & Buyrokoğlu, s. (2014). OecD Ülkelerinde ekonomik Özgürlük: Bir Kümeleme analizi. Uluslararası Alanya İşletme Fakültesi Dergisi, 6(3), 95–109. 18 h. al Qahtani anD J. p. sanKar Karaca, Z. (2018). the cluster analysis in the manufacturing industry with K-mean method: an application for turkey. Eurasian Journal of Economics and Finance, 6(3), 1–12. https://doi.org/10.15604/ejef.2018.06.03.001 Kim, D. h., Kim, s., & lee, J. s. (2023). the rise and fall of industrial clusters: experience from the resilient transformation in south Korea. The Annals of Regional Science, 71(2), 1–23. https://doi.org/10.1007/ s00168-022-01170-6 Kim, h., hwang, s. J., & Yoon, W. (2023). industry cluster, organizational diversity, and innovation. International Journal of Innovation Studies, 7(3), 187–195. https://doi.org/10.1016/j.ijis.2023.03.002 König, t. (2023). Between collaboration and competition: co-located clusters of different industries in one region— the context of tuttlingen’s medical engineering and metal processing industries. Regional Science Policy & Practice, 15(2), 288–325. https://doi.org/10.1111/rsp3.12581 Koo, g. p. Y., Zheng, h., aik, J. c. l., tan, B. Y. Q., sharma, V. K., sia, c. h., Ong, M. e. h., & ho, a. F. W. (2023). clustering of environmental parameters and the risk of acute ischaemic stroke. International Journal of Environmental Research and Public Health, 20(6), 4979. https://doi.org/10.3390/ijerph20064979 libaers, D., & Meyer, M. (2011). highly innovative small technology firms, industrial clusters and firm internationalization. Research Policy, 40(10), 1426–1437. https://doi.org/10.1016/j.respol.2011.06.005 liberti, l., lavor, c., Maculan, n., & Mucherino, a. (2014). euclidean distance geometry and applications. SIAM Review, 56(1), 3–69. https://doi.org/10.1137/120875909 liu, r. (2022). Data analysis of educational evaluation using K-means clustering method. Computational Intelligence and Neuroscience, 2022, 3762431. https://doi.org/10.1155/2022/3762431 MacQueen, J. (1967). some methods for classification and analysis of multivariate observations. in Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability (Vol. 1, pp. 281–297). Madsen, e. s., smith, V., & Dilling-hansen, M. (2003). industrial clusters, firm location and productivity – some empirical evidence for Danish firms. in Working Paper 03-26 (03-26, issue May 2014). https://pure.au.dk/ws/ files/32304323/03-26_esmvs.pdf Mittal, h., pandey, a. c., saraswat, M., Kumar, s., pal, r., & Modwel, g. (2022). a comprehensive survey of image segmentation: clustering methods, performance parameters, and benchmark datasets. Multimedia Tools and Applications, 81(24), 35001–35026. https://doi.org/10.1007/s11042-021-10594-9 nidhi, & patel, K. a. (2016). an efficient and scalable density-based clustering algorithm for normalize data. Procedia Computer Science, 92, 136–141. https://doi.org/10.1016/j.procs.2016.07.336 novillo-Villegas, s., ayala-andrade, r., lopez-cox, J. p., salazar-Oyaneder, J., & acosta-Vargas, p. (2022). a roadmap for innovation capacity in developing countries. Sustainability, 14(11), 6686. https://doi.org/10.3390/su14116686 OecD (2019). Measuring distortions in international markets: the aluminium value chain. in OECD trade policy papers (pp. 1–121). OecD publishing. https://doi.org/10.1787/c82911ab-en Oyewole, g. J., & thopil, g. a. (2023). Data clustering: application and trends. Artificial Intelligence Review, 56(7), 6439–6475. https://doi.org/10.1007/s10462-022-10325-y palamara, F., piglione, F., & piccinini, n. (2011). self-organizing map and clustering algorithms for the analysis of occupational accident databases. Safety Science, 49(8–9), 1215–1230. https://doi.org/10.1016/j.ssci.2011.04.003 park, J., lee, h., & park, Y. (2009). Disembodied knowledge flows among industrial clusters: a patent analysis of the Korean manufacturing sector. Technology in Society, 31(1), 73–84. https://doi.org/10.1016/j.techsoc.2008.10.011 pereira, D., leitão, J., Oliveira, t., & peirone, D. (2023). proposing a holistic research framework for university strategic alliances in sustainable entrepreneurship. Heliyon, 9(5), e16087. https://doi.org/10.1016/j.heliyon.2023.e16087 pereira, t., Kennedy, J. V., & potgieter, J. (2019). a comparison of traditional manufacturing vs additive manufacturing, the best method for the job. Procedia Manufacturing, 30, 11–18. https://doi.org/10.1016/j.promfg.2019.02.003 plantec, Q., le Masson, p., & Weil, B. (2021). impact of knowledge search practices on the originality of inventions: a study in the oil & gas industry through dynamic patent analysis. Technological Forecasting and Social Change, 168, 120782. https://doi.org/10.1016/j.techfore.2021.120782 prakash, a., Jha, s. K., prasad, K. D., & singh, a. K. (2017). productivity, quality and business performance: an empirical study. International Journal of Productivity and Performance Management, 66(1), 78–91. https://doi.org/10.1108/ iJppM-03-2015-0041 prebanić, K. r., & Vukomanović, M. (2023). exploring stakeholder engagement process as the success factor for infrastructure projects. Buildings, 13(7), 1785. https://doi.org/10.3390/buildings13071785 ran, c., song, K., & Yang, l. (2020). an improved solution for partner selection of industry-university cooperation. Technology Analysis & Strategic Management, 32(12), 1478–1493. https://doi.org/10.1080/09537325.2020.1786044 rivera, l., gligor, D., & sheffi, Y. (2016). the benefits of logistics clustering. International Journal of Physical Distribution & Logistics Management, 46(3), 242–268. https://doi.org/10.1108/iJpDlM-10-2014-0243 schulz, M. a., chapman-rounds, M., Verma, M., Bzdok, D., & georgatzis, K. (2020). inferring disease subtypes from clusters in explanation space. Scientific Reports, 10(1), 12900. https://doi.org/10.1038/s41598-020-68858-7 sharma, s. (2011). improved Bsp clustering algorithm for social network analysis. Bonfring International Journal of Software Engineering and Soft Computing, 1(1), 15–20. https://doi.org/10.9756/BiJsesc.1003 sheikh, a., ghanbarpour, t., & gholamiangonabadi, D. (2019). a preliminary study of fintech industry: a two-stage clustering analysis for customer segmentation in the B2B setting. Journal of Business-to-Business Marketing, 26(2), 197–207. https://doi.org/10.1080/1051712X.2019.1603420 cOgent BUsiness & ManageMent 19 singhal, V., singh, a. B., ahuja, V., & gera, r. (2023). consumer segmentation in the fashion industry using social media: an empirical analysis. Journal of Information and Organizational Sciences, 47(2), 399–419. https://doi. org/10.31341/jios.47.2.9 sun, J., Fan, p., Wang, K., & Yu, Z. (2022). research on the impact of the industrial cluster effect on the profits of new energy enterprises in china: Based on the Moran’s i index and the fixed-effect panel stochastic frontier model. Sustainability, 14(21), 14499. https://doi.org/10.3390/su142114499 swarndepp, s. J., & pandya, s. (2016). an overview of partitioning algorithms in clustering techniques. International Journal of Advanced Research in Computer Engineering & Technology, 5(6), 1943–1946. tan, a. J. J., chong, c. Y., & aleti, a. (2022). e-sc4r: explaining software clustering for remodularisation. Journal of Systems and Software, 186, 111162. https://doi.org/10.1016/j.jss.2021.111162 thakare, Y., & Bagal, s. (2015). performance evaluation of K-means clustering algorithm with various distance metrics. International Journal of Computer Applications, 110(11), 12–16. https://doi.org/10.5120/19360-0929 tingley, D., Ásmundsson, J., Borodzicz, e., conides, a., Drakeford, B., rúnar eðvarðsson, i., holm, D., Kapiris, K., Kuikka, s., & Mortensen, B. (2010). risk identification and perception in the fisheries sector: comparisons between the Faroes, greece, iceland and UK. Marine Policy, 34(6), 1249–1260. https://doi.org/10.1016/j.marpol.2010.05.002 Unterberger, p., & Müller, J. M. (2021). clustering and classification of manufacturing enterprises regarding their industry 4.0 reshoring incentives. Procedia Computer Science, 180, 696–705. https://doi.org/10.1016/j.procs.2021.01.292 Vargas-hernández, J. g., Vargas gonzalez, O. c., & Morones servín, J. s. (2020). a theoretical approach to the concept of the cluster. Problems of Management in the 21st Century, 15(1), 56–67. https://doi.org/10.33225/10.33225/ pmc/20.15.56 Wang, W., tang, Q., & gao, B. (2023). exploration of cO2 emission reduction pathways: identification of influencing factors of cO2 emission and cO2 emission reduction potential of power industry. Clean Technologies and Environmental Policy, 25(5), 1–15. https://doi.org/10.1007/s10098-022-02456-1 Wenting, r. (2008). spinoff dynamics and the spatial formation of the fashion design industry, 1858–2005. Journal of Economic Geography, 8(5), 593–614. https://doi.org/10.1093/jeg/lbn030 Westerdijk, M., Zuurbier, J., ludwig, M., & prins, s. (2012). Defining care products to finance health care in the netherlands. The European Journal of Health Economics, 13(2), 203–221. https://doi.org/10.1007/s10198-011-0302-6 Yang, F., & gu, s. (2021). industry 4.0, a revolution that requires technology and national strategies. Complex & Intelligent Systems, 7(3), 1311–1325. https://doi.org/10.1007/s40747-020-00267-9 Yang, W. c., lai, J. p., liu, Y. h., lin, Y. l., hou, h. p., & pai, p. F. (2023). Using medical data and clustering techniques for a smart healthcare system. Electronics, 13(1), 140. https://doi.org/10.3390/electronics13010140 Ye, D., Zheng, l., & he, p. (2021). industry cluster innovation upgrading and knowledge evolution: a simulation analysis based on small-world networks. SAGE Open, 11(3), 215824402110316. https://doi.org/10.1177/21582440211031604 Yuan, c., & Yang, h. (2019). research on K-value selection method of K-means clustering algorithm. J, 2(2), 226–235. https://doi.org/10.3390/j2020016 Zhao, l., liang, Y., & tu, h. (2023). how do clusters drive firm performance in the regional innovation system? a causal complexity analysis in chinese strategic emerging industries. Systems, 11(5), 229. https://doi.org/10.3390/ systems11050229 Zhou, K., Yang, s., & shao, Z. (2017). household monthly electricity consumption pattern mining: a fuzzy clustering-based model and a case study. Journal of Cleaner Production, 141, 900–908. https://doi.org/10.1016/j. jclepro.2016.09.165