Search

Humans possess a remarkable capacity to recognize and manipulate abstract structure, which is especially apparent in the domain of geometry. Recent research in cognitive science suggests neural networks do not share this capacity, concluding that human geometric abilities come from discrete symbolic structure in human mental representations. However, progress in artificial intelligence (AI) suggests that neural networks begin to demonstrate more human-like reasoning after scaling up standard architectures in both model size and amount of training data. In this study, we revisit empirical results in cognitive science on geometric visual processing and identify three key biases in geometric visual processing: a sensitivity towards complexity, regularity, and the perception of parts and relations. We test tasks from the literature that probe these biases in humans and find that large pre-trained neural network models used in AI demonstrate more human-like abstract geometric processing.

Value Stream Mapping (VSM) is a Lean method for waste reduction, crucial across industries. In the era of the fourth industrial revolution, Small and Medium-sized Enterprises (SMEs) are adopting smart, sustainable practices. The focus has shifted from traditional VSM to Smart Sustainable VSM (SS-VSM). This study introduces an SS-VSM framework, integrating lean, sustainable, and smart techniques. A fuzzy-based Preference Selection Index (FPSI) method selects the optimal SS-VSM strategy, considering environmental, economic, and social factors. Sensitivity analysis tests the approach's robustness. The results offer a conceptual framework for implementing and managing VSM, aiding SMEs in enhancing performance through SS-VSM practices.

Artificial systems currently outperform humans in diverse computational domains, but none has achieved parity in speed and overall versatility of mastering novel tasks. A critical component to human success, in this regard, is the ability to redeploy and redirect data passed between cognitive subsystems (via abstract feature representations) in response to changing task demands. However, analyzing shared representations is difficult in neural systems with distributed nonlinear coding. This work presents a simple but effective approach to this problem. In experiments, the proposed model robustly predicts behavior and performance of multitasking networks on natural language data (MNIST) using common deep network architectures. Consistent with existing theory in cognitive control, representation structure varies in response to (a) environmental pressures for representation sharing, (b) demands for parallel processing capacity, and (c) tolerance for crosstalk. Implications for geometric (dimension, curvature), functional (automaticity, generalizability, modularity), and applied aspects of representation learning are discussed.

Applying machine learning algorithms to automatically infer relationships between concepts from large-scale collections of documents presents a unique opportunity to investigate at scale how human semantic knowledge is organized, how people use it to make fundamental judgments ("How similar are cats and bears?"), and how these judgments depend on the features that describe concepts (e.g., size, furriness). However, efforts to date have exhibited a substantial discrepancy between algorithm predictions and human empirical judgments. Here, we introduce a novel approach to generating embeddings for this purpose motivated by the idea that semantic context plays a critical role in human judgment. We leverage this idea by constraining the topic or domain from which documents used for generating embeddings are drawn (e.g., referring to the natural world vs. transportation apparatus). Specifically, we trained state-of-the-art machine learning algorithms using contextually-constrained text corpora (domain-specific subsets of Wikipedia articles, 50+ million words each) and showed that this procedure greatly improved predictions of empirical similarity judgments and feature ratings of contextually relevant concepts. Furthermore, we describe a novel, computationally tractable method for improving predictions of contextually-unconstrained embedding models based on dimensionality reduction of their internal representation to a small number of contextually relevant semantic features. By improving the correspondence between predictions derived automatically by machine learning methods using vast amounts of data and more limited, but direct empirical measurements of human judgments, our approach may help leverage the availability of online corpora to better understand the structure of human semantic representations and how people make judgments based on those.

The social and economic development of the islands is generally dependent on the interconnection level with the mainland obtained through maritime transport services. When connecting many islands, the route planning is essential and typically a variety of constraints must be considered. Various optimization methods have been established to improve cost-efficiency but today environmental concerns, like the reduction of CO2 emissions, have become mandatory. This paper proposes a vessel scheduling optimization model able to simultaneously consider compulsory and optional tasks and vehicle-dependent profits. The algorithm was applied to seven islands of the Tyrrhenian Sea located in front of Sicily, named “Aeolian Islands”. Considering the regional requirements in terms of minimum number of routes and maximum fare prices for each season, this research compared the optimal vessels option obtained maximizing the profit with the one obtained minimizing the emissions. In particular, we have conducted three different analyses, in the first one we have considered only the mandatory routes while the second one was carried out identifying a series of potentially attractive additional activities based on historical demand data provided by the company that currently manages the service. Finally, the third analysis addresses a scenario where older fuel-powered vessels were replaced with hybrid electric ones.

Value Stream Mapping (VSM) is a standard Lean tool for identifying and reducing waste. It is used to create value and improve the performance of any industry. The value of the VSM has greatly expanded from conventional to smart over time; hence, researchers and practitioners in this sector are paying more emphasis. Comprehensive review research is needed to understand VSM-based smart, sustainable development from a triple-bottom-line perspective. The primary goal of this research is to look at various insights from the historical literature that can help with the adoption of smart, sustainable development through VSM. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) with a fifteen-year time range (2008–2022) is being considered to study various insights and gaps in value stream mapping. Analysis of the significant outcomes presents an eight-point study agenda: year, national context, research method, sector, wastes, VSM type, applied tools, and analysis indicators. The critical finding implies that empirical qualitative research dominates the research sector. Effective VSM implementation requires balancing the three sustainable dimensions of economic, environmental, and social through digitalization. The circular economy should also bolster research on the intersection of sustainability applications and new digital paradigms such as industry revolution 4.0.

One of the key sectors for the green transition of European countries is construction, that is more and more asked to evolve towards innovative ecological binders and green cost-effective processes. The construction sector is highly energy intensive, and the cement production is one of the main sources of environmental pollution in the world. In this regard, GeoPolymers (GP) seem to be promising for a sustainable replacement of cementitious materials. Therefore, the Life Cycle Assessment (LCA) of the industrial production of different formulations of Geopolymer Concrete (GC) was performed in this study after scaling up the Life Cycle Inventory (LCI) from a laboratory scale to an industrial one. Based on LCA results, the Global Warming Potential (GWP) indicator demonstrated a lower greenhouse gas emission of the proposed GC production in respect to the CC manufacturing process, while no significant difference was observed in the GWP scores of the considered GC formulations when referring to the functional unit. Nevertheless, the usage of the innovative GC formulations, on an industrial scale, would avoid a significant reduction of sand and kaolin extracted, with a consequent decrease on the environmental impact. Finally, the economic assessment showed that the combined production of thermal and electrical energy by a cogeneration system could provide a significant cost reduction when the percentage of electricity fed into the public network is higher than 39%.

Interstitial lung diseases (ILDs) are some of the first and most serious complications of connective tissue diseases (CTDs). However, the pathogenesis of CTD-related ILDs (CTD-ILDs) is still unclear and their treatment often depends on functional and radiographic disease progression as well as on patient age and comorbidities. It can be difficult to manage CTD-ILDs due to their heterogeneous nature, the lack of robust therapeutic data, and the few well-defined outcome measures. This review focuses on cyclophosphamide due to its crucial role in the treatment of systemic sclerosis-related ILD, particularly in the case of patients with progressive ILD. This narrative review was performed using PubMed, Medline, and Cochrane Library databases to retrieve English language papers published between 2000 and April 2020 concerning the treatment of CTD-ILDs with cyclophosphamide.

Industry 4 [...]

Finding the locations of departments or machines in a workspace is classified as a Facility Layout Problem. Good placement of departments has a relevant influence on manufacturing costs, work in process, lead times and production efficiency. This paper analyses the problem of allocating departments with restrictions in terms of unequal area and rectangular shape within a facility, in order to minimize the sum of material handling costs taking into account the satisfaction of the aspect ratio requested. In particular, we propose for the first time a Firefly Algorithm based on the slicing structure encoding. The proposed method was tested comparing the results obtained from other authors on the same literature instance. The results confirm the effectiveness of the Firefly Algorithm in solving the Facility Layout Problem by generating the best solutions with respect to those provided by previous researches.

In a sustainable management of logistics, transportation plays a crucial role. Traditionally, the main purpose was to solve the Vehicle Routing Problem minimizing the cost associated with the travelled distances. Nowadays, the economic profit cannot be the only driver for achieving sustainability and environmental issues have to be also considered. In this paper, to satisfy the intricate limits involved in real vehicle routing problem, the study has been structured considering different types of vehicles in terms of maximum capacity, velocity and emissions, asymmetric paths, vehicle-client constraints and delivery time windows. The firefly algorithm has been implemented to solve the vehicle routing problem and the TOPSIS technique has been applied to integrate economic and environmental factors. Finally, to prove the effectiveness of the proposed approach, a numerical example has been proposed using data provided by a logistic company located in Sicily.

Recent developments in manufacturing processes and automation have led to the new industrial revolution termed “Industry 4.0”. Industry 4.0 can be considered as a broad domain which includes: data management, manufacturing competitiveness, production processes and efficiency. The term Industry 4.0 includes a variety of key enabling technologies i.e., cyber physical systems, Internet of Things, artificial intelligence, big data analytics and digital twins which can be considered as the major contributors to automated and digital manufacturing environments. Sustainability can be considered as the core of business strategy which is highlighted in the United Nations (UN) Sustainability 2030 agenda and includes smart manufacturing, energy efficient buildings and low-impact industrialization. Industry 4.0 technologies help to achieve sustainability in business practices. However, very limited studies reported about the extensive reviews on these two research areas. This study uses a systematic literature review approach to find out the current research progress and future research potential of Industry 4.0 technologies to achieve manufacturing sustainability. The role and impact of different Industry 4.0 technologies for manufacturing sustainability is discussed in detail. The findings of this study provide new research scopes and future research directions in different research areas of Industry 4.0 which will be valuable for industry and academia in order to achieve manufacturing sustainability with Industry 4.0 technologies.

Traditional approaches to vibration risk assessment are often carried out manually on the basis of reference data, whereas an automated approach, can yield more valuable information that can be beneficial for improving the health and safety conditions of the workers. In particular, the exposure to prolonged vibration is a potential cause of pathologies such as the hand-arm vibration syndrome (HAVS), therefore national and international regulations establish recommended limits to the workers’ exposure to vibration within an allowable daily limit. However, evaluating the maximum allowable operating time is not straightforward, therefore reference vibration values are provided by the manufacturer. These values are frequently unreliable, since the effective vibration intensity generated by mechanical machines largely depends upon several specific factors such as maintenance, operating conditions, etc. A more effective approach would rather involve equipping the worker with suitable instruments to monitor and analyse in real time the vibration exposure. This is particularly difficult in open-field operations as mechanical harvesting, where wireless sensing and communicating technologies can be effectively employed in general framework of the Internet of Things (IoT) to develop small monitoring devices at affordable cost. In such context, the present research proposes an innovative system aimed at estimating the hand-arm exposure to vibration according to the Standard EN ISO 5349-1:2004. In particular, the proposed system is based on a referenced monitoring system, based on the employment of a compact wearable unit to be attached to the waist of the operator and a fixed station for data storage and analysis. The paper describes how this information can be exploited in a decision context to effectively schedule the working shifts for a team of workers in high exposure operations such as olive harvesting with mechanical shakers. .

The paradigm of Industry 4.0 allows to increase the efficiency and effectiveness of the production. Companies that will implement advanced solutions in production systems will increase their level of competitiveness and will be able reach high market shares. The present paper is focused on the development of advanced digital solutions to be implemented on a close power loop test bench designed to test high power transmissions for naval unit. In particular, the test configuration consists of a back-to-back connection between two identical mechanical reducers. Since the efficiency of these systems are very high, it is not necessary to use large electric motors, thus managing to contain the operating costs of the testing phase. The particular test bench allows to size the electric motor simply based on the dissipated power by the kinematic mechanisms. By means of suitable sensors installed on the test bench it is possible to extrapolate countless technical data. The implementation of Industry 4.0 enabling technologies allows to evaluate the increase in efficiency compared to traditional systems in terms of reduction of noise and vibrations, efficiency of lubrication, reduction of consumption, installation and maintenance cost of the entire system. [ABSTRACT FROM AUTHOR]

Polysomnography has developed from our understanding of sleep and its associated physiologic processes. This important tool extends the clinical examination into dynamic states that typically do not permit intrusive inspection. The two critical components of polysomnography are the determination of sleep-wake stage and identification of related bodily processes. In this article, the authors review the current standards for clinical polysomnography and discuss technical considerations that influence the accuracy of recorded data.

Saccharomyces cerevisiae Rad51 is structurally similar to Escherichia coli RecA. We investigated the role of S. cerevisiae RAD51 in DNA damage-associated unequal sister chromatid exchanges (SCEs), translocations, and inversions. The frequency of these rearrangements was measured by monitoring mitotic recombination between two his3 fragments, his3-Delta5' and his3-Delta3'::HOcs, when positioned on different chromosomes or in tandem and oriented in direct or inverted orientation. Recombination was measured after cells were exposed to chemical agents and radiation and after HO endonuclease digestion at his3-Delta3'::HOcs. Wild-type and rad51 mutant strains showed no difference in the rate of spontaneous SCEs; however, the rate of spontaneous inversions was decreased threefold in the rad51 mutant. The rad51 null mutant was defective in DNA damage-associated SCE when cells were exposed to either radiation or chemical DNA-damaging agents or when HO endonuclease-induced double-strand breaks (DSBs) were directly targeted at his3-Delta3'::HOcs. The defect in DNA damage-associated SCEs in rad51 mutants correlated with an eightfold higher spontaneous level of directed translocations in diploid strains and with a higher level of radiation-associated translocations. We suggest that S. cerevisiae RAD51 facilitates genomic stability by reducing nonreciprocal translocations generated by RAD51-independent break-induced replication (BIR) mechanisms.

The biological significance of DNA damage-induced gene expression in conferring resistance to DNA-damaging agents is unclear. We investigated the role of DUN1-mediated, DNA damage-inducible gene expression in conferring radiation resistance in Saccharomyces cerevisiae. The DUN1 gene was assigned to the RAD3 epistasis group by quantitating the radiation sensitivities of dun1, rad52, rad1, rad9, rad18 single and double mutants, and of the dun1 rad9 rad52 triple mutant. The dun1 and rad52 single mutants were similar in terms of UV sensitivities; however, the dun1 rad52 double mutant exhibited a synergistic decrease in UV resistance. Both spontaneous intrachromosomal and heteroallelic gene conversion events between two ade2 alleles were enhanced in dun1 mutants, compared to DUN1 strains, and elevated recombination was dependent on RAD52 but not RAD1 gene function. Spontaneous sister chromatid exchange (SCE), as monitored between truncated his3 fragments, was not enhanced in dun1 mutants, but UV-induced SCE and heteroallelic recombination were enhanced. Ionizing radiation and methyl methanesulfonate (MMS)-induced DNA damage did not exhibit greater recombinogenicity in the dun1 mutant compared to the DUN1 strain. We suggest that one function of DUN1-mediated DNA damage-induced gene expression is to channel the repair of UV damage into a nonrecombinogenic repair pathway.