Partial online cycle elimination in inclusion constraint graphs

The purpose of this paper is to evaluate and compare past models of transfer lines. After a discussion of the nature of transfer lines and their stoppages, the assumptions and derivations of seven analytical models are compared. The major discrepancies in assumptions concern whether failures are time dependent or operation dependent. The comparison of derivations includes a translation of all results to a common set of symbols. The predictions of the models at various banking levels are observed to vary significantly. Model validation is discussed in two phases. First the validity of the assumptions is tested by real data from a transfer line. The major problem is the failure of the real data to satisfy the assumptions of the model. Finally, the predictions of the analytical models are compared with a simulation model which uses the actual data. The difference between analytical and simulation models was found to be significant.

Recent advances in Natural Language Processing (NLP) have produced state of the art results on several sequence to sequence (seq2seq) tasks. Enhancements in embedders and their training methodologies have shown significant improvement on downstream tasks. Word vector models like Word2Vec, FastText & Glove were widely used over one-hot encoded vectors for years until the advent of deep contextualized embedders. Protein sequences consist of 20 naturally occurring amino acids that can be treated as the language of nature. These amino acids in combinations with each other makeup the biological functions. The choice of vector representation and architecture design for a biological task is highly dependent upon the nature of the task. We utilize unlabelled protein sequences to train a Convolution and Gated Recurrent Network (CGRN) embedder using Masked Language Modeling (MLM) technique that shows significant performance boost under resource constraint setting on two downstream tasks i.e., F1-score(Q8) of 73.1% on Secondary Structure Prediction (SSP) & F1-score of 84% on Intrinsically Disordered Region Prediction (IDRP). We also compare different architectures on downstream tasks to show the impact of the nature of biological task on the performance of the model.

It is well recognized that models in the life sciences can be sensitive to small variations in their model functions, a phenomenon known as ‘structural sensitivity’. Conventionally, modellers test for sensitivity by varying parameters for a specific formulation of the model functions, but models can show structural sensitivity to the choice of functional representations used: a particularly concerning problem when system processes are too complex, or insufficiently understood, to theoretically justify specific parameterizations. Here we propose a rigorous test for the detection of structural sensitivity in a system with respect to the local stability of equilibria, the main idea being to project infinite dimensional function space onto a finite dimensional space by considering the local properties of the model functions. As an illustrative example, we use our test to demonstrate structural sensitivity in the seminal Rosenzweig–MacArthur predator–prey model, and show that the conventional parameter-based approach can fail to do so. We also consider some implications that structural sensitivity has for ecological modelling: we argue that when the model exhibits structural sensitivity but experimental results remain consistent it may indicate that there is a problem with the model construction, and that in some cases trying to find an ‘optimal’ parameterization of a model function may simply be impossible when the model exhibits structural sensitivity. Finally, we suggest that the phenomenon of structural sensitivity in biological models may help explain the irregular oscillations often observed in real ecosystems.

This paper presents a framework for analyzing the performance of multithreaded programs using a model called a constraint graph. We review previous constraint graph definitions for sequentially consistent systems, and extend these definitions for use in analyzing other memory consistency models. Using this framework, we present two constraint graph analysis case studies using several commercial and scientific workloads running on a full system simulator. The first case study illustrates how a constraint graph can be used to determine the necessary conditions for implementing a memory consistency model, rather than conservative sufficient conditions. Uisng this method, we classify coherence misses as either required or unnecessary. We determine that on average one half of all load instructions which suffer cache misses due to coherence activity are unnecessarily stalled because the original copy of the cache line could have been used without violating the memory consistency model. The second case study demonstrates the effects of memory consistency constraints on the fundamental limits of instruction level parallelism, compared to previous estimates which did not include multiprocessor constraints. Using this method we determine the fundamental performance differences of various memory consistency models for processors which do not perform consistency-related speculation.

Given an edge-weighted graph and a set of known seed vertices of interest, a network scientist often desires to understand the graph relationships to explain connections between the seed vertices. If the size of the seed set is 2, shortest path calculations are an attractive computational kernel to explore the connections between the two vertices. When the seed set is 3 or larger (say up to 1,000s) Steiner minimal tree – min-weight acyclic connected subgraph (of the input graph) that contains all the seed vertices – is an attractive generalization of shortest weighted paths. In general, computing a Steiner minimal tree is NP-hard, but decades ago several polynomial-time algorithms were designed and proven to yield Steiner trees whose total weight is bounded within 2 times the minimal Steiner tree. Despite its rich theoretical literature, works related to parallel Steiner minimal tree computation and their scalable implementations are rather scarce. In this paper, we present a parallel 2-approximation Steiner minimal tree algorithm (with theoretical guarantees) and its MPI-based distributed implementation. In place of distance computation between all pairs of seed vertices, an expensive phase in many approximation algorithms, the solution we employ, exploits Voronoi cell computation. Also, this approach has higher parallel efficiency than others that involve minimum spanning tree computation on the entire graph. Furthermore, our distributed design exploits asynchronous processing and a message prioritization scheme to accelerate convergence of distance computation, employs techniques to avoid inefficient distributed spanning tree computation on the entire graph, and harnesses a combination of vertex and edge centric processing to offer fast time-to-solution. We demonstrate scalability and performance of our solution using real-world graphs with up to 128 billion edges and 512 compute nodes (8K processes), show the ability to find Steiner trees with up to 10K seed vertices in under one minute, and present in-depth analyses that highlight the benefits of our design choices. Using four real-world graphs and three seed sets for each, we compare our solution with the state-of-the-art exact Steiner minimal tree solver, SCIP-Jack, and two sequential algorithms with the same approximation bound as our algorithm. Our distributed solution comfortably outperforms these related works on graphs with 10s million edges and offers decent strong scaling – up to 90% efficient. We empirically show that, on average, the total distance (sum of edge weights) of the Steiner tree identified by our solution is 1.0527 times greater than the Steiner minimal tree (i.e., the optimal solution) – well within the theoretical bound of less than equal to 2.

Natural-draft flow and heat transfer in a plancha-type biomass cookstove

Theoretical studies of a spark-ignited supercharged hydrogen engine

Wave propagation in granular chains is subject to dispersive effects as well as nonlinear effects arising from the Hertzian contact law. This enables the formation of wideband pulses, which is a desirable feature in the context of diagnostic and therapeutic ultrasound applications. However, coupling of the ultrasonic energy from a chain of spheres into biological tissue is a big challenge. In order to improve the energy transfer efficiency into biological materials, a matching layer is required. A prototype device is designed to address this by using six aluminum spheres and a vitreous carbon matching layer. The matching layer and the precompression force are selected specifically to maximize the acoustic pressure in water and its bandwidth. The designed device generates a train of wideband ultrasonic pulses from a narrow-band input with a center frequency of 73 kHz. An analytical model is created to simulate the behavior of a matching layer as a flexible thin plate clamped from the edges. This model is then verified using free-field hydrophone measurements in water, which successfully predict the increased bandwidth by generation of harmonics. The shapes of the measured and predicted waveforms are compared by calculating the normalized cross-correlation, which shows 83% similarity between both. Since the generation of harmonics is of interest in this study, the total harmonic distortion (THD) and the − 6 -dB bandwidth of the signals are used to analyze signal fidelity between the hydrophone measurements and the model predictions. The acoustic signals in water have a root-mean-square THD of 73%, and the model predicts a root-mean-square THD of 78%. The − 6 -dB bandwidths of individual pulses measured by a hydrophone and predicted with the model are 280 and 252 kHz, respectively. At these high ultrasonic frequencies, it is an experimental demonstration of resonant chains operating in water with a matching layer.

A variety of mechanistic and statistical methods to forecast seasonal influenza have been proposed and are in use; however, the effects of various data issues and design choices (statistical versus mechanistic methods, for example) on the accuracy of these approaches have not been thoroughly assessed. Here, we compare the accuracy of three forecasting approaches—a mechanistic method, a weighted average of two statistical methods and a super-ensemble of eight statistical and mechanistic models—in predicting seven outbreak characteristics of seasonal influenza during the 2016–2017 season at the national and 10 regional levels in the USA. For each of these approaches, we report the effects of real time under- and over-reporting in surveillance systems, use of non-surveillance proxies of influenza activity and manual override of model predictions on forecast quality. Our results suggest that a meta-ensemble of statistical and mechanistic methods has better overall accuracy than the individual methods. Supplementing surveillance data with proxy estimates generally improves the quality of forecasts and transient reporting errors degrade the performance of all three approaches considerably. The improvement in quality from ad hoc and post-forecast changes suggests that domain experts continue to possess information that is not being sufficiently captured by current forecasting approaches.

Purpose The purpose of this paper is to present a methodology for research through game design and discuss how simulation games can be used to bridge the gap between operational exercises and simulation or analytical modelling and to provide guidelines on how simulation games can be designed for different research purposes in the context of humanitarian logistics. Design/methodology/approach This paper combines a literature review on gaming as a research method with an analysis of requirements for humanitarian logistics research methods. Starting from this theoretical framework, the authors develop a design thinking approach that highlights how games can be used for different research purposes. To illustrate the approach, the authors develop two different game set-ups that are of increasing fidelity and complexity. Finally, the authors discuss the results of the evaluation of both approaches, reflect on the design choices and provide recommendations for research and practice. Findings Gaming is a suitable research method to explore and analyse behaviour and decisions in emergent settings that require team work and collaborative problem solving. Especially when safety and security concerns may hinder access and experimentation on site, gaming can offer a realistic and engaging quasi-experimental environment. The aspects of engagement and realism also make gaming a suitable tool to combine training and research. Originality/value Although the use of games has attracted some attention in commercial supply chain management and crisis response, there is no systematic overview of gaming as a research method in humanitarian logistics. This paper is set to make a headway in addressing this gap by proposing a concrete approach to design games for humanitarian logistics research.

A systematic analysis of design choices in short-term taxi demand prediction models

This study presents a low-order broadband noise prediction tool for centrifugal plug fans, based on analytical models using inputs from CFD simulations. Firstly, the fan used as a reference in this investigation was tested in a double reverberant room where the aerodynamic performance and the acoustic signature were measured simultaneously. Secondly, CFD RANS simulations of the fan in the test rig were performed. The outcome of the simulations is then used to feed Amiet’s analytical broadband noise model, along with wall-pressure spectral and correlation length models. Amiet’s model has been transposed to a rotating blade, adapted to the geometry of the fan. Finally, the model predictions are compared with the measurements.

Laser-assisted machining (LAM) could help to significantly reduce the machining forces and improve machinability of hard to machine materials. The physical ground explanation for machinability improvement of the LAM stay largely unknown. In the current study, a 3D model is developed for the temperature field calculation induced by laser preheating. Taking preheating temperature field as initial input, an analytical milling force model is developed based on Oxley’s contact mechanics. Model validations are conducted by comparing the milling forces between the model predictions and experimental measurements in the LAM of Inconel 718 (IN 718). A good agreement is found. Also, the LAM forces are compared with traditional milling. The reduced specific cutting energy and reduced milling forces in LAM show the improved machinability of IN 718. The proposed analytical model provides a new method for the LAM process optimization.

An analytical model of a wheat breeding program

The process of the bubble growth and detachment from a wall with uniform up°ow parallel to the wall has been studied in this thesis. Experiments have been designed and performed in such a way that the assumptions of an existing analytical model were met. The aim of this study was to validate the predictions and to quantify the hydrodynamic lift force on a boiling bubble. This kind of detailed modeling of detachment can in principle be combined with numerical modeling of °ow with heat transfer in evaporator tubes, e.g. ri°ed tubes, with the aid of commercially available packages. The ¯ndings can be used to predict boiling characteristics in complicated geometries, such as ri°ed tubes, for conditions that occur during di®erent operation regimes of conventionally ¯red power plants. In the corresponding EC-project, all major European electricity companies join e®orts to design a power plant operating under high-temperature steam conditions to achieve 5% increase of the net e±ciency. The model presented yields analytical expressions for the forces acting on the bub- ble. A fully closed solution of the added mass forces involved in motion and growth of bubbles footed on the wall was obtained. Validation and quanti¯cation experiments corresponding to this model are described in this thesis. Nucleate boiling experiments were performed using demineralized water at near- saturated °ow boiling conditions. Both the thermal boundary condition and the approaching °ow (inlet) condition have been prescribed. The bubble substrate tem- perature and approach velocity were both constant. A microscale heater and a Wheat- stone bridge were used to maintain a constant wall temperature around an arti¯cial cavity. This cavity was used as the nucleation site where bubble would appear on a vertical part of the wall. This bubble generator has been given a special shape to minimize the velocity boundary layer thickness at the location of the arti¯cial cavity. The bubble generator intrudes a pipe and positions the arti¯cial site at the center of the pipe. As a result, a (nearly) homogeneous liquid velocity pro¯le approaches the bubble at the boiling site. Extensive image processing analysis enabled the determination of bubble geometrical parameters and to make a 3D reconstruction of the bubble volume history. The main non-hydrodynamic force components in the direction perpendicular to the vertical wall were determined from the measured quantities. The sum of these forces should equal the hydrodynamic force that is independently assessed using model predictions and measured geometrical parameter histories. This comparison was made, and used to draw conclusions on how much the deformations in°uence comparison with the model predictions and how the model should be ex- tended. In addition to model validation, temperature and power measurements have been used in a heat transfer analysis. It is shown that a signi¯cant part (60-70%) of the heat needed to make a bubble of certain volume in °ow boiling experiments arises from the superheated liquid layer in front of the heating element in respect to the heat delivered through the micro layer beneath the bubble foot from the electrical heater. These ¯ndings are con¯rmed by comparing the heater area and e®ective area, and by investigating the area of the bubble in°uence. The main conclusion drawn is that the assumption of a truncated sphere shape of the bubble does not correspond to actual bubble shapes at early times of bubble growth. The bubble is °attened parallel to the wall in this stage of growth. Therefore the volume equivalent radius does not yield an accurate representation of the actual frontal area of the bubble. That is the reason why the prediction of the hydrodynamic lift force is not good for this case. In the second half of the growth time the bubble becomes elongated away from the wall. It is not growing any more, but it moves away from the wall and pushes the surrounding liquid. This results in the negative hydrodynamic force. Before the de- tachment the neck is formed and this phenomenon is not considered by the model. Deformations should be accounted for by introducing more than one geometrical pa- rameter to describe the shape. Other recommendations given in this thesis consider the improvement of the experimental set-up and using some other measuring tech- niques. A brief description of how the model can be combined with commercially available packages has been provided. Boiling is the most e±cient, yet least understood, phase change process. The work presented in this thesis leads to an increased understanding of the physical phe- nomenon of the bubble growing and detaching in °ow boiling. The obtained knowl- edge should be used to improve modeling of this process and to be applied in industry.