This paper examines the evolution of urban industrial symbiosis (UIS) through a business ecosystem lens. UIS, defined as the synergistic integration of urban and industrial systems where waste and by-products are repurposed as industrial inputs, has demonstrated strong potential to address climate change, urbanization challenges, and supply chain vulnerabilities. Existing UIS research has primarily focused on material flows, network structures, and policy arrangements. However, limited attention has been paid to how coordination mechanisms evolve as ecosystem interdependence and complexity increase over time. This omission is consequential because even powerful coordinators face structural limits when coordinating expanding interdependencies among heterogeneous actors and resource flows, creating a coordination gap. Drawing on case studies of Ulsan (South Korea), Händelö (Sweden), EcoPark (Hong Kong), Metro Vancouver (Canada), Kalundborg (Denmark), HS Orka (Iceland), and Kawasaki (Japan), the paper develops a process model of UIS ecosystem coordination that links generation-processing-auxiliary (G-P-A) subsystems, coordination tensions and mechanisms, and governance pathways across three analytical phases. The study advances UIS research by shifting from exchange-oriented perspectives toward dynamic ecosystem coordination and theorizing how coordination mechanisms evolve under increasing ecosystem complexity. It further extends business ecosystem theory by explaining how increasingly distributed coordination capacity sustains ecosystem resilience as growing interdependence strains centralized orchestration.
Strategic R&D management requires foresight tools that are dynamic, uncertainty-aware, and transparent, yet most existing approaches remain static, deterministic, and difficult to interpret. This study introduces PATTERN-X (Patent Analytics Through Temporal Network Evaluation and Reasoning with eXplainability), an explainable AI framework integrating temporal patent network analysis, Bayesian deep learning, and Retrieval-Augmented Generation (RAG). Using 4167 European Patent Office patents in the Intelligent Lighting Systems (ILS) domain (1981–2024), the framework is empirically validated within the European ILS innovation ecosystem. The study addresses three research questions: RQ1 examines how temporal network analysis can reveal the evolution of technology innovation ecosystems; RQ2 investigates how probabilistic deep learning can quantify predictive uncertainty for risk-aware R&D decision-making; and RQ3 explores how generative AI can translate quantitative forecasts into evidence-based strategic narratives. Addressing RQ1, temporal co-classification network analysis reveals ecosystem evolution from 3 to 16 technology clusters and a shift in centrality toward digital control and vehicle-integrated technologies. Addressing RQ2, a Bayesian LSTM achieves an MAE of 0.89, an 88% strategic hit rate, and well-calibrated uncertainty estimates (PICP = 0.91 against a nominal 95% interval, reflecting a slight undercoverage typical of variational approximations), enabling risk-sensitive evaluation of convergence opportunities. Addressing RQ3, a CPC-guided RAG system generates evidence-grounded strategic narratives with high faithfulness (0.85; κ = 0.82), translating probabilistic forecasts into auditable managerial explanations. Together, these results demonstrate how dynamic analysis, uncertainty quantification, and explainable generative AI can enhance technology convergence forecasting and strategic innovation management.
Frugal business models have emerged as innovative alternatives for serving low-income markets through affordable, sustainable, and robust solutions, supporting organizational strategies under conditions of resource constraints. Initially developed in the context of emerging economies, these models have increasingly gained relevance in developed countries as well, driven by growing concerns regarding resource efficiency and sustainable consumption. Despite the rising academic interest in frugal innovation, research on frugal business models remains fragmented, with limited consolidation of their defining characteristics, dominant patterns, and broader impacts. To address this gap, this study conducts a systematic literature review using the ProKnow-C method. The analysis of a final portfolio of 74 articles identified 23 illustrative cases of frugal products and services and enabled the mapping of recurring patterns, strategic logics, and challenges associated with frugal business models. The findings indicate that these models can support sustainable innovation strategies; however, further empirical research is required to assess their impacts across both emerging and developed market contexts. Overall, this study contributes to the literature by synthesizing existing knowledge and outlining avenues for future research in this domain.
The transition toward sustainability in infrastructure sectors requires interventions that address systemic lock-ins rather than isolated technological improvements. However, innovation ecosystem research offers only limited guidance on how public-sector-driven ecosystems can be intentionally steered toward sustainability under conditions of structural inertia. This study employs Donella Meadows’ leverage points framework to identify and categorize potential systemic interventions in the Dutch road infrastructure innovation ecosystem. Using an action-oriented research design conducted over 1.5 years with public clients, contractors, and knowledge institutions, the study identifies 25 interventions across five thematic areas. The findings show that interventions predominantly target system design and feedback structures, the strengthening of governance, foresight, and integration capabilities rather than focusing on parameter-level change. This highlights how transformations in public-sector innovation ecosystems emerge through the gradual reconfiguration of rules, information flows, and coordination mechanisms. The study contributes to innovation ecosystem research by demonstrating how leverage points can be used to guide and structure ecosystem-level intervention strategies and support deliberate steering of public-sector-driven innovation ecosystems toward sustainability.
Generative artificial intelligence (GAI) is increasingly used by employees to support creative idea generation, yet less is known about its psychological consequences for employees’ subsequent willingness to remain engaged with those ideas. Drawing on psychological ownership theory, we examine whether GAI usage is associated with lower creative psychological ownership (CPO)—the extent to which employees perceive creative ideas as “theirs”—and, in turn, weaker intentions to elaborate, champion, and implement those ideas. We further propose that this relationship varies across human–GAI collaboration modes, with the negative relationship between GAI usage and CPO being more pronounced under GAI-centric collaboration than under human-centric or symbiotic collaboration. Across two scenario-based experiments (Studies 1–2), two task-based studies involving actual GAI interaction (Studies 3–4), and a multi-wave field study (Study 5), the findings provide converging, although not uniform, evidence for the proposed relationships. A mini meta-analysis further indicates an overall negative association between GAI usage and CPO and positive associations between CPO and downstream innovation-related intentions. Overall, the findings identify a potential psychological cost of GAI-assisted idea generation. They also show how collaboration modes may shape employees’ perceived ownership of ideas and their intentions to continue developing and advancing them.
