Search

Business accelerators play a key role in the initial critical stages of assessment of commercial viability, offering mentorship provision of funding and protection of intellectual property (IP) for product development and refinement. However, little is known about the decision-making criteria and detailed analysis of the underlying criteria and interdependencies between the key factors used by accelerator organizations to fund start-ups. This article focuses on the decision-making criteria utilized by a leading £21M accelerator program, largely funded by the European Regional Development Fund for initial stage funding and IP protection for product and innovation commercialization. We incorporate a multimethodological interpretive-based approach based on Day's “Real-Win-Worth” framework to develop the interrelationships and ranking between the factors. The results highlight the significance and weighting attached to the factors associated with the technical competency of the proposer and evidence of demand existing for the product. We propose a new framework that models the key factor interrelationships offering additional insight to accelerator-based decision making. [ABSTRACT FROM AUTHOR]

Transfer learning enables to re-use knowledge learned on a source task to help learning a target task. A simple form of transfer learning is common in current state-of-the-art computer vision models, i.e., pre-training a model for image classification on the ILSVRC dataset, and then fine-tune on any target task. However, previous systematic studies of transfer learning have been limited and the circumstances in which it is expected to work are not fully understood. In this paper we carry out an extensive experimental exploration of transfer learning across vastly different image domains (consumer photos, autonomous driving, aerial imagery, underwater, indoor scenes, synthetic, close-ups) and task types (semantic segmentation, object detection, depth estimation, keypoint detection). Importantly, these are all complex, structured output tasks types relevant to modern computer vision applications. In total we carry out over 2000 transfer learning experiments, including many where the source and target come from different image domains, task types, or both. We systematically analyze these experiments to understand the impact of image domain, task type, and dataset size on transfer learning performance. Our study leads to several insights and concrete recommendations: (1) for most tasks there exists a source which significantly outperforms ILSVRC’12 pre-training; (2) the image domain is the most important factor for achieving positive transfer; (3) the source dataset should include the image domain of the target dataset to achieve best results; (4) at the same time, we observe only small negative effects when the image domain of the source task is much broader than that of the target; (5) transfer across task types can be beneficial, but its success is heavily dependent on both the source and target task types. [ABSTRACT FROM AUTHOR]

Transfer learning enables to re-use knowledge learned on a source task to help learning a target task. A simple form of transfer learning is common in current state-of-the-art computer vision models, i.e., pre-training a model for image classification on the ILSVRC dataset, and then fine-tune on any target task. However, previous systematic studies of transfer learning have been limited and the circumstances in which it is expected to work are not fully understood. In this paper we carry out an extensive experimental exploration of transfer learning across vastly different image domains (consumer photos, autonomous driving, aerial imagery, underwater, indoor scenes, synthetic, close-ups) and task types (semantic segmentation, object detection, depth estimation, keypoint detection). Importantly, these are all complex, structured output tasks types relevant to modern computer vision applications. In total we carry out over 2000 transfer learning experiments, including many where the source and target come from different image domains, task types, or both. We systematically analyze these experiments to understand the impact of image domain, task type, and dataset size on transfer learning performance. Our study leads to several insights and concrete recommendations: (1) for most tasks there exists a source which significantly outperforms ILSVRC’12 pre-training; (2) the image domain is the most important factor for achieving positive transfer; (3) the source dataset should include the image domain of the target dataset to achieve best results; (4) at the same time, we observe only small negative effects when the image domain of the source task is much broader than that of the target; (5) transfer across task types can be beneficial, but its success is heavily dependent on both the source and target task types. [ABSTRACT FROM AUTHOR]

Innovation is seen today as an indispensable practice if an organization wishes to achieve competitive advantage and sustainable success in the marketplace. The development of new products and services and new technologies that enable those products and services seem to be the crown jewel of innovation. The management of these technologies, more specifically the process through which technologies are transferred, has been the focus of scholarly debates and studies for several decades. Currently, there are no quantitative, comprehensive, and easily applicable models in the literature that researchers and practitioners can use to assess organizations’ technology transfer capabilities. This article has sought to delineate important factors and perspectives applicable to technology transfer, and to use those in order to assess an organization's capabilities, through the application of a hierarchical decision modeling model to create a technology transfer score. The research model is applied to a utility company in the Pacific Northwest of the United States, and the results provide insights on how to measure and interpret technology transfer capabilities as well as a starting point for technology managers to improve those capabilities. [ABSTRACT FROM AUTHOR]

How do global value chains (GVCs) affect a firm's innovation impact? Working as international communities, GVCs are rich in knowledge and resource interchange. The prior literature has focused on knowledge absorbing or internal knowledge transfer within firms for in-house innovation. However, scant research has focused on promoting innovation across interfirm networks in a GVC context. This article develops a new concept of interfirm transformative capacity as a firm's ability to transfer developed technology knowledge across time, space, and firm networks. In this article, we explore three mechanisms of interfirm transformative capacity in GVCs: 1) choosing knowledge intermediaries, 2) transmitting and maintaining knowledge over time and space, and 3) reactivating and synthesizing knowledge. We predict that engagement in a lead firm's GVC, structural holes, and network centrality will affect a focal firm's innovation impact as a result of interfirm transformative capacity. Integrator suppliers have higher innovation impacts than modular suppliers. To test these three mechanisms, we build a novel dataset from a five-year panel of Apple suppliers and their patent impacts. The results support our hypotheses related to the mechanisms of interfirm transformative capacity. Our theoretical elaboration and empirical findings have significant implications for innovation policy within, and governance of, GVCs. [ABSTRACT FROM AUTHOR]

Servitization by manufacturing firms has been a growing trend in recent years. However, the literature is unclear on whether servitization improves firm performance. Using survey data from 334 Chinese manufacturing firms, we examine how servitization affects firm performance. We also investigate the contingent role of platform-based innovation. Our results show that servitization enhances latent needs identification and worsens dominant logics conflict. Platform-based innovation is effective in mitigating harm from dominant logics conflict, but it also limits the transfer of latent customer needs. This article contributes to the literature by identifying underlying mediating effects between servitization and firm performance to explain how servitization affects firm performance. The findings also address the “service paradox.” [ABSTRACT FROM AUTHOR]

This article aims to understand the effect of firms’ application development activities when introducing new technologies from universities or public institutes on the success of commercialization and corporate growth. Technology itself does not guarantee corporate growth; instead, it is commercialization that contributes more to achieving a sustainable competitive edge, as it offers economic value. We collected samples on the commercialization performance of start-ups that adopted the technology of universities or public research institutes and measured the intensity of the firms’ application development activities as the gap between the technology readiness levels (TRLs) at the time of technology acquisition and the firms’ current TRL. Our findings show that the process of transferring technology to firms is mediated by the firms’ application development performance. Most notably, we first observed that firms’ investments in commercialization have no significant direct effect on commercialization success; instead, we found that this relationship is mediated by the TRL gap. This article suggests that it is necessary to consider the entire process—from the technology transfer to firms to successful commercialization—to form a mechanism under which universities and research institutes can create substantial additional value and allow their research to realize their full potential. [ABSTRACT FROM AUTHOR]

The smart specialization strategy (S3) introduces a new “policy-prioritization logic” according to which regional strategies should target interventions in those research and innovation domains in which the region has a competitive advantage. In other words, the adoption of an S3 strategy has shifted the regional innovation strategies’ focus from privileging exclusively high-tech sectors to including also those lower tech domains that are recognized as a source of regional competitive advantage. As a consequence, this shift calls for a rethinking of university–industry relationships for innovation, which have traditionally privileged high-tech sectors. In this light, this article aims to analyze how the introduction of the S3 has changed the role of universities by comparing patterns of academic engagement with local entrepreneurs in both low-tech and high-tech domains. To this purpose, we conduct an empirical study in the Region of Campania (South of Italy) to analyze differences in terms of university–industry channels and motivations in different S3 technological domains. [ABSTRACT FROM AUTHOR]

Bidirectional capacitive power transfer technology makes it possible for energy sharing among multiple electronic devices. This article proposes an asymmetrical bidirectional power conversion topology to satisfy different input and output characteristics’ requirements and increase load variation tolerance. A π–T (CLC–LCL) resonant topology is designed for this bidirectional conversion mode with constant output voltage characteristics in both directions. A hybrid power flow regulation strategy is proposed by integrating multiple zero-voltage switching soft-switching operating points switching and phase-shifted mode. This method overcomes the problem of mixing two different type topologies (push–pull and full bridge) with two different power regulation modes (switch soft-switching operating points and phase shift), which provide a way to make two different chargers compatible. Simulation and experimental results verified the proposed method. [ABSTRACT FROM AUTHOR]

This article analyzes the effects of three channels of technology transfer—global value chain participation (GVCP), foreign technology import (FTI), and domestic technology purchase (DTP)—on innovation efficiency by using 11 years of data from Chinese manufacturing industries. Empirical results show FTI and GVCP significantly facilitate innovation efficiency, whereas DTP decreases it. Further, absorptive capacity (AC) positively moderates the relationship between FTI and innovation efficiency. Moreover, in the high-tech industry, we find that GVCP positively affects innovation efficiency, and AC positively moderates the effect of GVCP on innovation efficiency. This finding is informative from a strategic perspective as firms can make strategic decisions in the selection of technology transfer channels. In addition, firms may also absorb and apply external technologies while developing AC to improve enterprises' ability according to their own organizational and strategic contexts in order to achieve and maintain competitive advantages. This finding provides convincing empirical evidence on the relationship between technology transfer channels and innovation efficiency and valuable lessons to other developing countries for the selection of technology transfer channels. [ABSTRACT FROM AUTHOR]

In this article, we propose BNAS-v2 to further improve the efficiency of broad neural architecture search (BNAS), which employs a broad convolutional neural network (BCNN) as the search space. In BNAS, the single-path sampling-updating strategy of an overparameterized BCNN leads to terrible unfair training issue, which restricts the efficiency improvement. To mitigate the unfair training issue, we employ a continuous relaxation strategy to optimize all paths of the overparameterized BCNN simultaneously. However, continuous relaxation leads to a performance collapse issue that leads to the unsatisfactory performance of the learned BCNN. For that, we propose the confident learning rate (CLR) and introduce the combination of partial channel connections and edge normalization. Experimental results show that 1) BNAS-v2 delivers state-of-the-art search efficiency on both CIFAR-10 (0.05 GPU days, which is $4\times $ faster than BNAS) and ImageNet (0.19 GPU days) with better or competitive performance; 2) the above two solutions are effectively alleviating the performance collapse issue; and 3) BNAS-v2 achieves powerful generalization ability on multiple transfer tasks, e.g., MNIST, FashionMNIST, NORB, and SVHN. The code is available at https://github.com/zixiangding/BNASv2 . [ABSTRACT FROM AUTHOR]

The existing simultaneous wireless power (SWP) and information transfer can only achieve serial signal transmission and it is hard to decouple the power and signal transmission without the additional circuits. To solve these problems, a SWP and multibit signals transfer technology is proposed in this article. Based on the working principle of the hybrid modulation waves pulsewidth modulation control technology, the power and multibit signals are modulated into the system through the modulation waves with different frequencies. Therefore, the signal transmission no longer depends on the power transmission, and the signal transmission rate is multiplied through the realization of parallel signals transmission. Subsequently, the parameter selection criterion of the signal channel is given by analyzing the signal transmission rate and the interference between the power and signal transmission. Finally, the proposed technology is verified with experimental results. The experimental results demonstrate that the power and two-bit signals can be transmitted stably. Compared to the serial transmission, the band signal transmission rate is increased by two times. [ABSTRACT FROM AUTHOR]

For implantable medical devices, it is of paramount importance to ensure uninterrupted energy supply to different circuits and subcircuits. Instead of relying on battery stored energy, harvesting energy from the human body and any external environmental sources surrounding the human body ensures prolonged life of the implantable devices and comfort of the patients. In this article, we present existing issues and challenges related to the state-of-the-art solutions used for harvesting energy to power implantable devices. In addition, the details on existing energy storage technologies and various wireless power transfer techniques incorporating external or internal energy sources and sensors have been discussed. The authors have outlined the performance and power constraints of existing biomedical devices and provided a brief overview of various power architectures found in the literature. This survey has been conducted on existing implantable solutions in terms of output voltage, current, device dimension, application, generated power, energy density, and so on. Finally, the advantages and drawbacks of different solutions have been discussed and compared. Therefore, this article can be considered as an expedient reference for researchers conducting research in the field of energy scavenging, internal energy storage, wireless power transfer techniques, and power management of implantable medical devices. [ABSTRACT FROM AUTHOR]

The papers in this special section focus on the ability to integrate evolving incubator models into the organization. Management and organizational science focusing on the complex process of R&D that leads to the subsequent commercialization has come a long way since the days of post-World War II industrial and research laboratories. In this context, the evolution of the incubator can be retraced from the foundational research in the 1950s and 1960s of Researchon-Research—that explored the R&D process in a number of settings to derive and test theories of organization behavior. Since then, we have seen a concentration of research and practice that have spanned a burgeoning range of incubators models and entities that, in some manner, exist to play a role to support and harness new technology-based start-ups. For more than half a century, these third-party entities in the entrepreneurial ecosystem have spread from the United States to Europe and are a well-established worldwide phenomenon. Business incubators have been largely viewed as an essential means to facilitating the process of knowledge transfer and tech-based commercialization through the provision of office infrastructure, laboratories and equipment, business support services, mentoring, and access to entrepreneurial finance and specialist networks for the founders of these embryonic start-ups. [ABSTRACT FROM AUTHOR]

University spin-offs (USOs) have attracted increasing attention due to being an important source of innovation, university income, and job creation. To support their creation and growth, universities have developed university ecosystems of diverse actors. However, existing research has found mixed results on the effectiveness of these support mechanisms. In particular, there is a lack of research illustrating the impact premarket support mechanisms can have on premarket USO development and their strategic outcomes such as timing of the USOs’ first market entry. The findings identify that both the university environment and ecosystem actors leave lasting positive and negative imprints upon USO founders. USO founders who had positive engagement with university ecosystem actors and perceived their university environment as supportive, developed entrepreneurial skills, and market knowledge earlier in the USO formation process. This resulted in greater market readiness and, consequently, contributed to an early first market entry. Conversely, USO founders who perceived barriers within their university environment and had negative experiences with ecosystem stakeholders developed more cautious and risk-averse behavior, contributing to a late market entry. This article contributes new knowledge into the antecedents of USO market entry timing through the novel lens of imprinting. [ABSTRACT FROM AUTHOR]

Misunderstanding of technology transfer and lack of a scientific viewpoint in this area are among the fundamental problems of some industries. A novel Delphi-decision-making trial and evaluation laboratory (DEMATEL)-elimination and choice expressing reality (ELECTRE) decision-making method, which benefited from both qualitative and quantitative methods, is introduced on gray numbers to solve the selection problem of technology providers systematically. The technology provider evaluation criteria are developed according to a literature review and experts’ opinions. Next, the Delphi method is used to evaluate the criteria. Gray DEMATEL is used to investigate the cause and effect relationship between the selected criteria, based on which the criteria were ranked. Finally, using the gray ELECTRE method, the priority of companies is determined. Based on the results of the illustrative case study, the two most robust causal criteria are management/executive capability and cost. Moreover, cost and support through organization policies played a leading role in the system. Also, out of nine technology providers, company 4 is superior to other companies, whereas most companies were of greater advantage to company 5. According to the comparison of weights by Spearman's correlation coefficients, the proposed method is as efficient as the available methods. This novel method can be used in uncertain conditions to select the most reliable company. [ABSTRACT FROM AUTHOR]

3-D printing is an additive manufacturing process, which enables products to be custom designed. Obtaining the evolution trend and identifying the potential R&D hotspots of 3-D printing technology are important strategic issues for both nations and enterprises. In this article, a combination method is proposed using the Viterbi algorithm to identify technological terms in patents, using the latent Dirichlet allocation model to capture potential technological topics in patents, and using the hidden Markov model to analyze the distribution and evolution pattern of the technological topics. Based on the results, technological topics and future R&D hotspots in 3-D printing are identified. Among the technological topics, extrusion devices and laser powder molding are gradually losing R&D; model acquisition, extrusion devices, rotary printing, and denture are the key topics of technological evolution that are difficult to overcome; laser focusing, print head, color printing, recyclable, feeding, control panel, and denture will become the focus of R&D in the near future. Finally, a comparison and experiments are given to demonstrate the plausibility of our method. [ABSTRACT FROM AUTHOR]

This article aims to investigate the factors influencing the electronic customer relationship management (e-CRM) assimilation in Southeastern European companies. With ICT innovation in mind, a conceptual model is developed, integrating three theories of innovative diffusion. Multivariate techniques, factor analysis, and cluster analysis are applied in order to reduce the number of variables and classify the cases. A discriminant analysis is taken in order to test the reliability of the clustering and to explore the clusters’ characteristics. The relationships between cluster categorization and life cycle phases, along with the size of the firm, are all checked against contingency coefficients. When comparing cluster categorization, the research results show that the significance of technical, organizational, and environmental factors changes depending on the phase of the e-CRM assimilation. The results additionally show that the degree of usage and diffusion of these technologies differs in small to medium enterprises compared to big enterprises. Owing to these findings, which have enabled us to provide insights into the ways in which contextual factors influence the e-CRM assimilation, theorists, and practitioners can see the necessary patterns of action. [ABSTRACT FROM AUTHOR]

Wireless rechargeable sensor networks (WRSNs) employ wireless charging transfer technology to provide dynamic energy replenishment for sensors. Previous related research considers enhancing the flexibility of WRSNs with the use of wireless charging vehicles (WCVs). Although this research prolongs network lifetime to some extent, there still exist two essential drawbacks: off-road limitation and traveling speed limitation. Benefiting from the intrinsic flexibility, high flight speed, low cost and small size, wireless charging drones (drones) have great potentials to be used to charge sensors in WRSNs. However, suffering from limited battery capacities, it is impossible to use drones for charging sensors in large WRSNs independently. To conquer the drawbacks of WCVs and drones, we propose a novel WRSN model which integrates WCVs, WCV-carried drones and independent drones to construct an efficient novel charging system. The proposed system, which aims to extend network lifetime and to minimize the time of charging scheduling on premise of satisfying charging requests with the cooperation of WCVs and drones. Thus, we develop a novel charging scheduling scheme, which divides the sensing region into three circular sub-regions for scheduling the WCVs or drones collaboratively. Sensor nodes located in the inner ring are charged by independent drones directly from the base station. Nodes located in the middle ring are charged by WCVs, while the outer ring charged by WCV-carried drones. Simulation results show that our scheduling scheme outperforms multiple WCVs which either employ temporal or distantial priority scheme and earliest-deadline first scheme with respect to the lifetime and network charging efficiency. [ABSTRACT FROM AUTHOR]

We consider the broad class of decentralized optimal resource allocation problems in wireless networks, which can be formulated as a constrained statistical learning problems with a localized information structure. We develop the use of Aggregation Graph Neural Networks (Agg-GNNs), which process a sequence of delayed and potentially asynchronous graph aggregated state information obtained locally at each transmitter from multi-hop neighbors. We further utilize model-free primal-dual learning methods to optimize performance subject to constraints in the presence of delay and asynchrony inherent to decentralized networks. We demonstrate a permutation equivariance property of the resulting resource allocation policy that can be shown to facilitate transference to dynamic network configurations. The proposed framework is validated with numerical simulations that exhibit superior performance to baseline strategies. [ABSTRACT FROM AUTHOR]

Dynamic wireless charging (DWC) is one of the most promising and economical power supply solutions for electric vehicles (EVs), which can address range anxiety and downsize the volume of onboard batteries. To solve the null power phenomenon and transfer power pulsation phenomenon in DWC scheme, this paper proposes a new magnetic coupler array to implement continuous charging and reduce the output power variation. The transmitters are placed with a proper interval to eliminate the adjacent coupling and enlarge the coil array’s effective length. ANSYS Maxwell is utilized to demonstrate the finite element analysis (FEA) of the DWC system. A larger receiver coil is designed to significantly mitigate power pulsation and increase the magnetic coupling as the receiver moves along the array. With the well-designed receiver coil, the output power pulsation is $\pm$ 1.52 $\%$. For the high-power capacity scenario, a multilevel connected DC-DC converter is designed on the receiver side to regulate the transfer power. Besides, a passivity-based control (PBC) is developed for the receiving DC-DC converter based on the Euler-Lagrange (EL) model, and a current sharing compensator is employed to balance each phase inductor current. Finally, a 500- $\text{W}$ prototype is constructed, and experimental results are represented to validate the proposed optimized coil array and the proposed PBC method. [ABSTRACT FROM AUTHOR]

This article proposes an enhanced timely and robust constant current (CC) control for wireless in-flight charging systems. The challenge for practical wireless in-flight charging systems is to maintain a CC output for hovering drones under circumstances of the continuous variation of coupling effect, various charging power requirements, and the parameter shifting, which is nearly unexplored in previous studies on wireless power transfer technologies. In order to address the issue, this article adopts the online-trained radial basis function neural network (RBFNN) to ensure the expected CC output for battery charging, which aims to handle negative impacts of the continuously varied coupling effect, the disturbance of parameters, and the change of charging current. In this article, both simulated and experimental results are given to verify the effectiveness of the proposed control scheme, wherein the accuracy of the controlled output current is within 5% and the average response time is less than 100 ms. It shows that the proposed dynamic-balancing robust current control is an ideal technical solution for wireless in-flight charging of drones by means of remarkable characteristics of the adopted RBFNN-based controller, namely, the increased rapidity and the enhanced robustness. [ABSTRACT FROM AUTHOR]

Governments and industries continuously engage in large-scale projects aiming to unlock economic potential for commercial and social benefits of a country and its people. In the case of developing countries, where the dynamics of innovation ecosystems differ from those in developed countries, the need for sociotechnical transition projects are often quite high. The aim of this article is to propose a standardized framework suitable for developing countries for the technology roadmapping of large projects aimed to bring about systemic sociotechnical transitions. The framework is developed through the integration of existing generic technology roadmapping approaches with transition management theory and complex systems theory and was illustrated against a successful sociotechnical transition program that is characterized by Big Fast Results (BFR) requirements. Data were collected from qualitative interviews as well as from the critical analysis of relevant documents. The proposed sociotechnical transition technology roadmap accurately captured the key elements of the BFR project it was benchmarked against and even identified key elements that were not included in the BFR methodology. An advantage of the proposed sociotechnical transition technology framework is the fact that it makes use of a standard technology roadmapping process with few customizations that are ideal for use in developing countries. [ABSTRACT FROM AUTHOR]

Many firms increasingly collaborate with universities to create innovation through the commercialization of scientific knowledge. However, a variety of mismatches between a firm's interests and the value delivered by university partners (e.g., common knowledge base, strategic goals, agreements on intellectual property management, etc.) can cause rough interactions and relationships. Therefore, managers demand an effective practice to support the management of university-industry collaborations (UICs). This article builds on the previous work on a matching quality assessment model by extending its applicability to propose an analytical framework for developing a strategic roadmap to guide relationship development toward effective UIC. [ABSTRACT FROM AUTHOR]

The misalignment between the transmitting (Tx) coil and the receiving (Rx) coil is an urgent problem in the drone charging system using wireless power transfer technology. The coupling coefficient is so low when the resonance coils is not aligned, which degrades the drone wireless charging performance. In this article, the optimization processes of the coupling mechanism are presented in detail. First, we illustrate the great superiority of unequally spaced Tx coil, compared with traditional uniformly spaced one. Then, a genetic algorithm to optimize the side lengths of the Tx coils is highlighted by discussing the main design considerations and methods in a comprehensive manner. Furthermore, the Rx coil with varying radius size per turn is proposed in terms of the distribution of magnetic field generated by the optimized Tx coil. Experimental results show that the optimized coils have better coupling performance due to small mutual inductance change under the condition of misalignment. Power transfer efficiencies in different misalignment situations are measured and the best power transfer efficiency improvement achieved by the proposed system can reach 56.23% in contrast with traditional coils. This drone charging system can transfer a maximum power of 100 W with the remarkable transfer efficiency of 92.41%. [ABSTRACT FROM AUTHOR]

Experts have difficulty assessing the economic value of university-originated technologies due to the high level of uncertainty associated with the commercialization of early stage and basic technologies. This article proposes a random forest approach to the valuation of university-originated technologies that integrates monetary value and patent value models for technology valuation. First, a technological characteristics-value matrix was constructed after defining a total of 23 indicators from the U.S. Patent and Trademark and Scopus databases and extracting the value of university-originated technologies from technology transaction databases. Second, a random forest model, an ensemble machine learning model based on a multitude of decision trees, was employed to assess the economic value of university-originated technologies. Finally, the performance of our approach was assessed using quantitative metrics. A case study of the technologies registered in the Office of Technology Licensing of Stanford University confirms, with statistically significant outcomes, that our method is valuable as a complementary tool for the valuation of university-originated technologies. [ABSTRACT FROM AUTHOR]

Embedded nonvolatile memory technologies, such as resistive random access memory (RRAM) and spin-transfer torque magnetic RAM (STT MRAM), are increasingly being researched for application in neuromorphic computing and hardware accelerators for artificial intelligence (AI). However, the stochastic write processes in these memory technologies affect their yield and need to be studied alongside process variations, which drastically increase the complexity of yield analysis using the Monte Carlo approach. Therefore, we propose an approach based on the Fokker–Planck equation for modeling the stochastic write processes in STT MRAM and RRAM devices. Moreover, we show that our proposed approach can reproduce the experimental results for both STT-MRAM and RRAM devices. [ABSTRACT FROM AUTHOR]

One of the challenges with the dynamic inductive power transfer (DIPT) technique is the electric vehicle detection (EVD) that helps the DIPT system to control the power supply of the transmitter. The EVD method applying auxiliary coils is a promising candidate because the flat shape of the auxiliary coils is suitable for the compact design. However, the EVD may fail when the metallic foreign object (MFO) is present. Therefore, the desire emerges in the integration design of the EVD and foreign object detection (FOD). The FOD can ensure the reliability of the EVD as well as the highly efficient operation of the DIPT system without MFOs. In this context, this paper proposes an integrated solution to the EVD and FOD well suited for DIPT systems. The integrated solution utilizes both passive coil sets (PCSs) and active coil sets (ACSs). Additionally, a novel detection resonant circuit (DRC) is proposed to realize EVD and FOD using the same coil sets and to amplify the measurement sensitivity. The operation mechanisms, the detection coil sets architecture, the design of the proposed resonant circuits and the detection procedure are detailed. Finally, a printed circuit board based prototype is built to validate the integrated functionality of the EVD and FOD in a DIPT prototype processing 1 kW output. Experiments considering the practical DIPT application scenarios are conducted, and the proposed detection method is able to achieve advantageously high sensitivity and no blind zone. [ABSTRACT FROM AUTHOR]

An uninterruptable wireless power supply is developed for a left ventricular assist device (LVAD), which is an implanted, electrically driven centrifugal pump used to treat advanced heart failure pathologies. The LVADs are currently powered through a cable driveline (DL) that exits the patient's body through the skin. Here, an innovative uninterruptable power supply is presented, based on a hybrid wireless/wired architecture: the electric power of 10 W is transferred by the near field wireless power transfer technology operating at 300 kHz, from outside the human body to an implanted subcutaneous receiver, and then by an internal cable to the LVAD. A significant improvement of the subcutaneous receiver is here proposed, based on the integration in the receiving coil of a battery pack providing the following benefits: no-break power supply for the LVAD; peak load management; enhanced system reliability; elimination of DL infection; enhancement of patient's quality of life. The proposed system is first designed and optimized, then a demonstrator is made and tested in a laboratory using the latest LVAD generation. The measured coil currents are used as sources for dosimetric analysis to assess the compliance of the system with the electromagnetic field safety and thermal regulations. [ABSTRACT FROM AUTHOR]

Finding new industry partners poses a challenge to many public research organizations. This article explores how statistical classification can support partner selection at the example of the Fraunhofer Society in Germany, Europe's largest public organization for applied research. We use internal cooperation data and feature sets based on unstructured data, i.e., text and industry codes, both of which describe business activities of firms. An important advantage of this data is that it is available for most companies in Germany, even small and medium enterprises, which allows for an almost complete screening of the market, in contrast to using other data sources, e.g., patents. In addition, we also include economic variables linked to firms, as turnover, number of employees/managers and firm age. We report the performance of various classification techniques such as logistic regression, support vector machines, and random forests in our dataset for diverse combinations of feature sets. Results show that simple methods with fewer parameters remain competitive in comparison to complex ones. Overall, the performance of most classifiers is high enough to support the decision process of finding new industry partners for public research. [ABSTRACT FROM AUTHOR]

When considering a joint venture or merger, it is essential for firms to explore innovators with complementary technology to compensate for any internal limitations in R&D resources. In this article, we provide a framework for exploring the technology complementarity between enterprises in a quantitative manner based on text-mining patent data. A hierarchical latent Dirichlet allocation topic model identifies the technology topics hidden in patent documents along with the hierarchical structure of those topics. The technology complementarity between broad classes of technology and their subclassifications across enterprises is then measured with an improved formulation. An empirical study on three-dimensional printing illustrates the validity, reliability, and practicality of this method and the measurement formula used, endorsed by technical experts. This method can be used to identify R&D opportunities, to find appropriate acquisition targets and potential collaborators, and to support managerial decision-making with quantified information on technology complementarity. [ABSTRACT FROM AUTHOR]

This article uses text mining techniques to determine the time lag of knowledge transfer between research activity and technology development in bioremediation, complementing these with advanced visualization techniques in order to extract patterns that could be of interest for decision making in this field. The emergence patterns in this field have been identified and a method based on subject-action-object (SAO) semantic structure is proposed for characterizing such patterns, using 2-word tuples. Our results show that technology developments in heavy metal bioremediation swiftly follow scientific advances, as opposed to developments in bioremediation of organic chemical components. The science mapping reveals three distinct areas: 1) heavy metal remediation and phytoremediation; 2) aerobic and anaerobic remediation of chemical elements; and 3) bioremediation techniques for treating specific contamination sources such as oil. The emergence analysis points at activities involving energy recovery by bioremediation, and shows an increasing amount of technologies involving specific strains of microorganisms, which could gain significant traction in this field in an estimated time horizon of ten years. Our SAO approach, tested on the data sample corresponding to these strains, proves to be useful for characterizing the emerging technologies when applied to instrumental concepts. [ABSTRACT FROM AUTHOR]

Firms engaged in university–industry collaboration (UIC) desire to advance knowledge and new technologies with their academic partners, however the performance of knowledge creation and transfer is greatly affected by complementarity and compatibility between the firm and academic institute. In practice, many firms assess the appropriateness of academic partners using a broad range of criteria, with high subjectivity. This paper therefore seeks to address this challenge by introducing a systematic approach to help firms determine the appropriateness of academic partners, represented as “matching quality” in this paper. Three facets of matching quality are examined: complementarity of research ability, compatibility in business mindset, and compatibility in relationships. This paper proposes a framework and model for assessing matching quality, along with a case study to demonstrate how the proposed model works. [ABSTRACT FROM AUTHOR]

Based on the evolutionary game theory, this paper explores the evolutionary stable strategy (ESS) of university–industry collaborative innovation. By utilizing MATLAB software, the analysis of the influence of agent coupling degree, group size, and government policies on the stable evolution of university–industry collaborative innovation is taken into account, while simulating. Our findings indicate that, first, agent coupling degree is positively and nonlinearly related to the stable evolution of university–industry collaborative innovation; second, when an enterprise's size is equal to university's, group size is more significant to the stable evolution of university–industry collaborative innovation than agent coupling degree; third, when a group size of an enterprise and university is different, the combinations of agent coupling degree and group size have various effects on the stable evolution of enterprise group and university group; and fourth, government policies have a significantly positive impact on the stable evolution of university–industry collaborative innovation, and incentive measures are more effective than punitive ones. Research conclusions have developed the relationship among university–industry collaborative innovation agents, expanded the boundary of the stability study of collaborative innovation, and provided new ideas for improving the effectiveness of government innovation policies. [ABSTRACT FROM AUTHOR]

Computing capacity has always been on the upward climb due to the constant technological improvements in semiconductor manufacturing and packaging industry. This growth in computing capability is usually accompanied by a steep rise in heat flux density associated with the electronic component (CPUs or GPUs, for example). High-performance computing (HPC) data centers often employ several of these high-performance devices for crunching their enormous artificial intelligence (AI) or scientific computing workloads. State-of-the-art air cooling technologies throttle after a certain heat flux level and would require bigger heat sinks driving enormous airflow through it, which may not be desirable from a data center operation standpoint. Hence, a lookout for advanced thermal management techniques is quite imperative. In this article, a commercially available intel overclockable CPU i9-9900k was exercised with high performance workloads while employing and evaluating three different cooling technologies namely air-cooled, cold plates, and two-phase (2P) immersion. The high heat carrying capacity associated with cold plates and 2P immersion techniques outperformed the air-cooled solution by constantly yielding higher CPU clock rates up to 41% utilizing cold plates and 51% employing 2P immersion. The performance of cold plates and 2P immersion was evaluated at different functional points (different coolant operating temperatures and 2P coolant types for example) to understand how the technologies would respond in a legacy or modern data center operation. It was again observed that both 2P immersion technology and cold plates provided the least thermal resistance (as low as 0.247 °C/W) path to heat transfer and therefore provided higher computational performance and efficiency compared to air cooling. Further improvement in performance is not limited by cooling but by packaging constraints. Advanced packaging techniques coupled with minimizing the interfacial resistance can reap enhanced performance and energy efficiency gains using these modes of cooling. This article demonstrates the potential increase in virtual machine (VM) performance that can be attained using cold plates or 2P immersion. Authors recommend adoption of liquid cooling in early chip design and architecture phases. Finally, the article provides a generic overclocking methodology that the industry can use to qualify postturbo server class chip performance. [ABSTRACT FROM AUTHOR]

If we have any hope of achieving sustainability, we, as researchers, need to develop a new perspective on universities and multistakeholder engagement. Stakeholder theory teaches us that engaging stakeholders in specific transactions and interactions can foster the sustainable development. But solving the grand challenge of sustainability requires more. Universities need to see themselves as part of a great network of stakeholders where interactions, knowledge, and data management go beyond the entrepreneurial university paradigm. Hence, in this article, we present Oztel's concept of a fourth-generation university as a launch-point for broadening the discourse on higher education's third mission. We establish a new stream of inquiry through four main propositions for research and technology management. Further, we call for interested scholars to pioneer this field with an inexhaustive list of potential avenues for future research. University managers, researchers, and policymakers should find great insights into the future evolution of sustainable development and its soon-to-be intrinsic place in the fabric of teaching, research, and society. [ABSTRACT FROM AUTHOR]

The constantly increasing availability of data, the rapid expansion in computational and storage capacities of information technology systems, and algorithmic advances in Machine Learning (ML) and Artificial Intelligence (AI) are making a huge impact in the manufacturing industry for improving efficiency, operations and throughput. The semiconductor industry, being one of the most data-intensive industries, has seen the diffusion of several ML-based technologies in recent years. [ABSTRACT FROM AUTHOR]

Deep-learning algorithms have produced promising results, however, domain adaptation remains a challenge. In addition, excessive training time and computing resource requirements need to be addressed. Deep-learning algorithms face a domain adaptation issue when the data distribution of a target domain differs from that of the source domain. The emerging concept of broad learning shows potential in addressing the domain adaptation and training time issues. An adaptive unsupervised broad transfer learning (AUBTL) algorithm is proposed to tackle the cross-domain problems. The proposed algorithm utilizes a sparse autoencoder and random orthogonal mapping to extract and augment the feature space. Then, it initializes the weights of a classifier by solving a ridge regression problem. The logit ranking strategy is applied to develop a transfer estimator to evaluate and sample data in the target domain for an adaptive transfer. Based on the sampled data, AUBTL optimizes the hyperparameter space. The performance of the AUBTL algorithm is validated with three benchmark datasets, including 20 transfer tasks. The computational results demonstrated the efficiency and accuracy of the proposed algorithm over other deep-learning algorithms considered in this research. [ABSTRACT FROM AUTHOR]

Acoustic droplet ejection (ADE) uses the acoustic energy produced by a focused ultrasound beam to provide a noncontact, highly precise, automatic, and cost-effective liquid transfer method for life science applications. The reported minimum precision of the current acoustic liquid transfer technology is 1 nL. Since precision improvement always brings valuable results in biological research, it is highly necessary to develop pico-liter precision liquid transfer technology. In this work, we developed a 40-MHz ultrahigh -frequency focused ultrasound transducer with a large aperture of 7 × 7 mm2 and a wide bandwidth of 76.4%. The designed transducer can successfully eject pico-liter droplets, and the droplet ejection accuracy ranges from 28 to 439 pL. The effects of the acoustic parameters, including excitation amplitude, pulsewidth, and frequency, on the size of the ejected droplet were studied. A wide range of ejected droplet sizes could be obtained by adjusting the acoustic parameters, thereby making liquid transfer flexible. The flexible pico-liter liquid transfer based on the wide-bandwidth, high-frequency ultrasound transducer is easier to achieve automatically, and thus it has broad prospects in biological research and industrial applications. [ABSTRACT FROM AUTHOR]

The research and production of computer graphics imagery and animation at The Ohio State University started with the artistic work of Prof. Charles Csuri. He developed the Computer Graphics Research Group in response to the award of a National Science Foundation Grant in 1974, and the group transferred its technology to a commercial production effort, Cranston/Csuri Productions, Inc., in 1981. CGRG evolved into the Advanced Computing Center for the Arts and Design in 1987. This article provides an historical review of the significant activities of these groups. [ABSTRACT FROM AUTHOR]

The use of wireless power transfer technology in unmanned aerial vehicle (UAV) systems is rapidly improving in different application areas. The main limitations of electric-powered UAVs are their charging range and stability in dynamic conditions. To solve these, this article proposes a hybrid control method based on dc-link and switched capacitor control with a fixed operating frequency. The dc-link voltage is regulated by the boost power factor correction converter to smoothen the output voltage, and the switched capacitor is used to dynamically compensate for self-inductance variations under dynamic coupling conditions. The output voltage stabilization and near-unity power factor of the input impedance can be achieved by applying the hybrid control method. Therefore, efficiency improvement can be achieved. Furthermore, the proposed control method can also be applied to systems with large coupling variations. To verify the proposed control method, a 500-W prototype was set up in the laboratory, and the experimental results proved that the control method could ensure a stable output voltage with horizontal and vertical offsets of 200 and 62 mm. respectively. Additionally, the results show that the maximum efficiency of the proposed control method is 91.9%, demonstrating that the proposed method effectively improves the efficiency. [ABSTRACT FROM AUTHOR]

How do we match technology transfer sources to recipients? In this paper, we approach this question by addressing gaps in two distinct streams of the literature. In the technology-transfer literature, the integrated pretransfer matching framework is in need of empirical validation, which calls for an empirical approach that can cope with its configurational, equifinal, causally asymmetric, and multinomial features. In the qualitative comparative analysis (QCA) literature, the emerging variant multivalue QCA (mvQCA), one of the few techniques to tackle the above four features is in need of multivalue theory applications. Thus, these two seemingly unrelated literature hold the key to fill each other's gap. To address the gaps, we leveraged the unique public research institute (PRI) to small and medium-sized enterprise (SME) transfer context in Singapore. We collected 188 responses from three stakeholders [i.e., research scientists and engineers, public research institutes (PRIs), and small and medium-sized enterprises (SMEs)] who are involved in technology transfer. Our analysis suggested one necessary and six sufficient PRI-SME configurations that enhance the commercialization potential of the transferred technology. Furthermore, we identified three sufficient PRI-SME configurations that hinder the commercialization of technology transfer activities. Not only does our paper establishes the validity of pretransfer matching framework, but also introduces the mvQCA method to management scholars. [ABSTRACT FROM AUTHOR]

Open innovation (OI) has captured increasing interest over recent decades as this approach is linked to higher level organizational ambidextrous strategy, and has become a prerequisite for achieving competitive advantages. Assessing firms’ propensity for OI has, however, turned out to be an increasingly challenging endeavor. This is particularly true for small and medium-sized enterprises (SMEs) due to the myriad of factors that influence these companies’ capability for innovation. To overcome this challenge, this paper sought to integrate two operational research/management science techniques—cognitive mapping and the Choquet integral (CI) (a nonadditive measure and information aggregator)—to identify and prioritize relevant criteria for evaluating SMEs’ propensity for OI, and improving their organizational ambidexterity. To facilitate a real-world application, information was first collected from SME managers and entrepreneurs who agreed to participate in face-to-face group meetings, allowing realism to be incorporated in the decision-making process. The results were validated by both the panel members and project director of COTEC Portugal—a leading think-and-action network for advancing technology diffusion and business innovation cooperation. The findings include that cognitive mapping facilitates the identification and understanding of cause-and-effect relationships between the determinants of OI in SMEs. The CI, in turn, introduces realism into the construction of value functions and the respective assessments of SMEs. The limitations and implications of the proposed system are also discussed. [ABSTRACT FROM AUTHOR]

The evolution of the incubator is retraced from the foundational research in the 1950s and 1960s of Research-on-Research—that explored the R&D process in a number of contexts to derive and test theories of organizational behavior. In this paper, we review the next phase of this evolution, starting with seminal incubation research in the early 1970s and, we reflect on the prolific growth of incubation research and practice, as well as new challenges that lie ahead for the next phase of incubation. The incubation field has expanded to include a wide range of models and a plethora of spin-off terminology that have been used interchangeably, with the emergence of the most recent model—the accelerator. Our historical critical review of the evolution of the incubation field is used to ask: in this current era of technological change, what can we learn from the past? We reflect on this question in an attempt to help direct researchers in their endeavors to contribute to the next generation of incubation research. This paper presents lessons learned from prior research and provides a range of indicative areas for further empirical studies at four levels of analysis. [ABSTRACT FROM AUTHOR]

In the past few years, with the rapid development of CPU-GPU heterogeneous computing, the issue of task scheduling in the heterogeneous cluster has attracted a great deal of attention. This problem becomes more challenging with the need for efficient co-execution of tasks on the GPUs. However, the uncertainty of heterogeneous cluster and the interference caused by resource contention among co-executing tasks can lead to the unbalanced use of computing resource and further cause the degradation in performance of computing platform. In this article, we propose a two-stage task scheduling approach for streaming applications based on deep reinforcement learning and neural collaborative filtering, which considers fine-grained task division and task interference on the GPU. Specifically, the Learning-Driven Workload Parallelization (LDWP) method selects an appropriate execution node for the mutually independent tasks. By using the deep Q-network, the cluster-level scheduling model is online learned to perform the current optimal scheduling actions according to the runtime status of cluster environments and characteristics of tasks. The Interference-Aware Workload Parallelization (IAWP) method assigns subtasks with dependencies to the appropriate computing units, taking into account the interference of subtasks on the GPU by using neural collaborative filtering. For making the learning of neural network more efficient, we use pre-training in the two-stage scheduler. Besides, we use transfer learning technology to efficiently rebuild task scheduling model referring to the existing model. We evaluate our learning-driven and interference-aware task scheduling approach on a prototype platform with other widely used methods. The experimental results show that the proposed strategy can averagely improve the throughout for distributed computing system by 26.9 percent and improve the GPU resource utilization by around 14.7 percent. [ABSTRACT FROM AUTHOR]

Societies today have come closer together as a result of the development of communication technologies and the greater ease with which business is transacted. Consequently, technological tools and devices are being transferred from industrial societies to other countries at dazzling speeds. The transfer of technology to non-industrialized societies with their own histories, civilizations, and backgrounds has led to the emergence of cultural issues and conflicts that are specific to the modern period and that had not existed prior to this period. In this article, we have reviewed the ideas of Don Ihde, a leading contemporary philosopher who has done much work in the field of the philosophy of technology and the relationship between technology and society. Using Don Ihde's work, we have striven to solve the problem of technology and culture interaction in the global periphery. We start by giving an account of Don Ihde's general view of the major relation between culture and technology, and proceed to an analysis of Ihde's ideas on the problem of cultural conflict in non-industrial countries. Having provided a critique of Ihde's ideas, we conclude that his answers to the problem are useful but incomplete. [ABSTRACT FROM AUTHOR]

Domain adaptation has been a fundamental technology for transferring knowledge from a source domain to a target domain. The key issue of domain adaptation is how to reduce the distribution discrepancy between two domains in a proper way such that they can be treated indifferently for learning. In this paper, we propose a novel domain adaptation approach, which can thoroughly explore the data distribution structure of target domain. Specifically, we regard the samples within the same cluster in target domain as a whole rather than individuals and assigns pseudo-labels to the target cluster by class centroid matching. Besides, to exploit the manifold structure information of target data more thoroughly, we further introduce a local manifold self-learning strategy into our proposal to adaptively capture the inherent local connectivity of target samples. An efficient iterative optimization algorithm is designed to solve the objective function of our proposal with theoretical convergence guarantee. In addition to unsupervised domain adaptation, we further extend our method to the semi-supervised scenario including both homogeneous and heterogeneous settings in a direct but elegant way. Extensive experiments on seven benchmark datasets validate the significant superiority of our proposal in both unsupervised and semi-supervised manners. [ABSTRACT FROM AUTHOR]

This paper proposes a quadrature-shaped pickup coil, which can realize an angular-misalignment insensitive omnidirectional wireless charging system. Recently, the three-dimensional wireless charging has shown promising prospects for removing the performance deterioration caused by the positional misalignment, which can transfer power to arbitrary spatial objectives. In the conventional design, the strength and the direction of the induced magnetic field can be adjusted by controlling the excitation currents of transmitting coils. However, the existence of angular misalignment caused by the self-rotation of the pickup unit still can result in a sharp decrease of the transmitted power. Accordingly, this paper proposes and implements a novel pickup coil topology to compensate such huge fluctuations in power transfer, which aims to ensure an angular misalignment insensitive omnidirectional wireless charging. Simulated and experimental results are both given to verify the feasibility of the proposed quadrature-shaped topology. [ABSTRACT FROM AUTHOR]