Feng Li

Peripheral actors are a recognised source of novelty, but how they assemble viable value configurations before central actors, platforms or policy institutions take notice remains poorly understood. This paper develops Qiaomiao innovation as a mode of peripheral innovation in which actors recognise relationally latent resources, configure connections across separately governed domains, convert costs or constraints into value sources, and layer complementary value streams into ecosystem-level value architectures. Three qualitative cases from China, spanning food processing, agrivoltaics and e-commerce logistics, show how peripheral actors work across categories that dominant systems often treat as separate: meat and by-products, energy and grazing, fulfilment and advertising. Using abductive inference to the best explanation, the paper compares five candidate explanations drawn from creative marginality, frugal innovation, bricolage and effectuation, business model and ecosystem research, and organisational ingenuity. Each explains part of the pattern, but none fully captures how peripheral position enables actors without architectural authority to integrate these elements into a distinctive configurational logic. The study contributes to research on peripheral innovation by explaining how such configurations become economically viable before central recognition, and by showing how partial visibility can create both imitation barriers and accountability risks.

This study examines how parallel innovation ventures (PIVs), ventures characterized by parallel innovation pathways, shared technological foundations, and adjacent yet non-substitutable market segments, innovate and develop in an early-stage, fragmented industry without direct competition, deep cooperation, or centralized orchestration. Drawing on a multiple-case study of seven assistive technology ventures, we conduct phenomenon-based research to develop a framework that explains how PIVs advance their own innovation efforts while collectively contributing to market emergence through: (1) independent capability deployment, (2) innovation resonance activation, and (3) differentiated business model enactment. In particular, we theorize innovation resonance as a relational reinforcement mechanism rooted in PIVs’ structural alignment, amplifying their innovation efforts and collective momentum while anchoring relational proximity. We further show that innovation resonance is activated through reciprocal PIV-engagement, which, as a foundational relational capability in the PIV context, also enables the deployment of other capabilities and shapes PIVs’ business models. Together, these insights offer a novel perspective on relational innovation dynamics in emerging industries, with useful implications for innovation, policy, and practice.

Sustainability research is vast and complex, making it difficult to grasp the big picture. To help researchers understand the field better, we created a comprehensive knowledge map of sustainability in management literature by utilising citation‐based clustering and text mining techniques to analyse 25,737 sustainability‐related papers across 22 academic disciplines, as categorised by the Chartered Association of Business Schools (ABS)'s Academic Journal Guide. We identify 23 research streams, with the top three being corporate social responsibility, operations & supply chain and tourism. Each research stream has varying degrees of definition coverage, levels of analysis, levels of interdisciplinarity and connections to the 17 United Nations' Sustainable Development Goals (SDGs). Goals #9, #10, #12 and #13 receive more attention than Goals #3, #14 and #15. Our findings suggest a need for more interdisciplinary collaboration between researchers, and we propose four research directions to achieve this.

A number of recent studies have examined the impact of advanced technologies on organizations. However, many (particularly those in developing countries) still face challenges when it comes to the adoption of mature technologies and have also continued to repeat many of the mistakes of early adopters, primarily in relation to automated workflow systems. The current paper analyses a case study of a public organization in the developing country of Saudi Arabia, with the aim of understanding its resistance to change brought about by the implementation of a mature technology, i.e., automated workflow systems. The study undertook semi-structured interviews with employees to establish the nature of this resistance, identifying their preference for familiar processes and systems, alongside their unwillingness to embrace the new system. Furthermore, the study highlighted a number of issues experienced during the implementation of automated workflow systems, including job security; changes in laws and rules; an inability to understand, and/or trust, the technology; the perceived risks and costs associated with change; and the transformation of business processes. It also cited factors related to organizational structure and power, and the discomfort involved in making difficult decisions. This study, therefore, aims to assist organizations to create a sound foundation for change prior to the adoption of more advanced technologies.

The rapid development of cloud computing and Internet of Things technology has caused the explosive growth of data scale and the high complexity of data models, leading to cluster research facing many new problems and challenges, such as massive data and new computing environments. This paper deeply studies the parallel CKLDC-means clustering algorithm based on Hadoop platform, and gives the algorithm design method and strategy. Experiments on multiple data sets of different sizes show that the designed parallel clustering algorithm has a good acceleration ratio and expansion rate, and is suitable for the analysis and mining of massive data.

In response to urban rail transit capacity shortage, a capacity adaption optimization model for the new mixed train operation mode is proposed in the supernormal operation of urban rail transit networks. First, the characteristics of express-local train and long-short route slow train are analyzed to obtain the main influencing factors of capacity adaption. Moreover, various train types are simultaneously classified and recombined in the adaption train operation mode to maximize transportation capacity for passengers. A nonlinear mixed-integer programming model is established to maximize the total number of trains, improving train load rate and reducing the total passenger travel time. In addition, the actual case of Beijing Metro Line 1 is solved to demonstrate the feasibility and effectiveness of the proposed model. The optimal passenger travel time-saving rates of the morning and evening peak hours are 12.1% and 11.5%, respectively. The optimal utilization rates of capacity in the morning and evening peak hours are 79.5% and 84.5%, respectively. Meanwhile, the optimal load factor within the range of 0.7–0.9 would benefit the utilization of train resources and the passengers’ service level. The optimized results show that the proposed capacity adaption optimization model benefits for high service level and practical significance and rationality for operators.

Laparoscopic surgery, as a representative minimally invasive surgery (MIS), is an active research area of clinical practice. Automatic surgical phase recognition of laparoscopic videos is a vital task with the potential to improve surgeons’ efficiency and has gradually become an integral part of computer-assisted intervention systems in MIS. However, the performance of most methods currently employed for surgical phase recognition is deteriorated by optimization difficulties and inefficient computation, which hinders their large-scale practical implementation. This study proposes an efficient and novel surgical phase recognition method using an attention-based spatial–temporal neural network consisting of a spatial model and a temporal model for accurate recognition by end-to-end training. The former subtly incorporates the attention mechanism to enhance the model’s ability to focus on the key regions in video frames and efficiently capture more informative visual features. In the temporal model, we employ independently recurrent long short-term memory (IndyLSTM) and non-local block to extract long-term temporal information of video frames. We evaluated the performance of our method on the publicly available Cholec80 dataset. Our attention-based spatial–temporal neural network purely produces the phase predictions without any post-processing strategies, achieving excellent recognition performance and outperforming other state-of-the-art phase recognition methods.

The world's most valuable public companies today are built on digital platforms. While American digital platforms (ADPs) have successfully dominated many international markets, some Chinese digital platforms (CDPs) have managed to survive and thrive in China, with European digital platforms (EDPs) largely absent both at home and abroad. Using comprehensive longitudinal data over 6 years, this study examines the platform competition between ADPs and CDPs in China, and explores how digital platforms challenge traditional ways of thinking about strategy and international competition. Some ADPs initially dominated the Chinese market, but some CDPs used their institutional advantages in China to offset the competitive advantages of ADPs based on superior resources and capabilities and strong market positions. Their competition evolved over three distinctive episodes so far, and CDPs used successive temporary advantages to achieve sustainable competitive advantages dynamically and accomplished radical changes cumulatively through incremental changes. A new dynamic equilibrium model of platform competition – ‘the spiral model’ – between global and native digital platforms is developed, which highlights a trajectory that is different from the dominant models of change in the literature. Limitations of the study and new areas for future research are highlighted.