Dynamic Batch Research Articles

Dynamic batch sentencing mechanisms for yield-based product acceptance determination with the simple linear profiles

Acceptance sampling plan has been extensively used in batch sentencing to provide the manufacturer and the customer a general benchmark to meet their predetermined needs on the batch quality. This paper develops a flexible sampling procedure, based on the SpkA index, for simple linear profiles by switching inspection rules. The plan parameters of the two suggested types of quick switching sampling (QSS) systems, satisfying the desirable quality levels and constraining the manufacturer’s and the customer’s risks, are derived by solving an optimization model. The comparisons between the suggested systems and the existing sampling plans are discussed, in terms of the discriminatory power and the average sample number to show the better performance of the suggested systems. Finally, the suggested QSS systems are applied in the electronics industry.

Dynamic hard pruning of Neural Networks at the edge of the internet

Neural Networks (NN), although successfully applied to several Artificial Intelligence tasks, are often unnecessarily over-parametrized. In edge/fog computing, this might make their training prohibitive on resource-constrained devices, contrasting with the current trend of decentralizing intelligence from remote data centres to local constrained devices. Therefore, we investigate the problem of training effective NN models on constrained devices having a fixed, potentially small, memory budget. We target techniques that are both resource-efficient and performance effective while enabling significant network compression. Our Dynamic Hard Pruning (DynHP) technique incrementally prunes the network during training, identifying neurons that marginally contribute to the model accuracy. DynHP enables a tunable size reduction of the final neural network and reduces the NN memory occupancy during training. Freed memory is reused by a dynamic batch sizing approach to counterbalance the accuracy degradation caused by the hard pruning strategy, improving its convergence and effectiveness. We assess the performance of DynHP through reproducible experiments on three public datasets, comparing them against reference competitors. Results show that DynHP compresses a NN up to 10 times without significant performance drops (up to 3.5% additional error w.r.t. the competitors), reducing up to 80% the training memory occupancy.

Research on Village Planning and Rural Architectural Design Based on Discrete Dynamic Modeling Technology

With the advent of the era of big data, the combination of science and technology and urban and rural planning has become the focus of many countries. The improvement of village planning and the redesign of rural buildings can promote the rapid development of villages and strengthen rural cohesion. Based on the above situation, this paper proposes a dynamic programming algorithm combined with discrete dynamic modeling technology to improve rural planning. Firstly, the dynamic programming algorithm is used to reconstruct the village layout and optimize the original village model. The dynamic monitoring technology is used to update the dynamic data in real time to provide specific information for the follow-up rural architectural design. Secondly, the dynamic modeling technology is used to build the building model, which can calculate the building location and building angle of the village. In order to further improve village development, we also put forward the concept of green building design. The performance of traditional modeling technology and discrete dynamic modeling technology in the green building design model is compared. The results show that the discrete dynamic modeling technology can improve the overall performance of rural buildings and improve the operation efficiency of the system in the batch design of green buildings. The village layout improved by dynamic planning reduces the specific travel distance of villagers and provides effective help for rural construction and economic development. Compared with traditional modeling technology, dynamic modeling technology has a shorter workflow and less time. Discrete dynamic modeling technology can realize dynamic batch architecture design and has higher applicability than traditional algorithms.

Anonymous and Efficient Authentication Scheme for Privacy-Preserving Distributed Learning

Distributed learning is proposed as a promising technique to reduce heavy data transmissions in centralized machine learning. By allowing the participants training the model locally, raw data is unnecessarily uploaded to the centralized cloud server, reducing the risks of privacy leakage as well. However, the existing studies have shown that an adversary is able to derive the raw data by analyzing the obtained machine learning models. To tackle this challenge, the state-of-the-art solutions mainly depend on differential privacy and encryption techniques (e.g., homomorphic encryption). Whereas, differential privacy degrades data utility and leads to inaccurate learning, while encryption based approaches are not effective to all machine learning algorithms due to the limited operations and excessive computation cost. In this work, we propose a novel scheme to resolve the privacy issues from the anonymous authentication approach. Different from the two types of existing solutions, this approach is generalized to all machine learning algorithms without reducing data utility, while guaranteeing privacy preservation. In addition, it can be integrated with detection schemes against data poisoning attacks and free-rider attacks, being more practical for distributed learning. To this end, we first design a pairing-based certificateless signature scheme. Based on the signature scheme, we further propose an anonymous and efficient authentication protocol which supports dynamic batch verification. The proposed protocol guarantees the desired security properties while being computationally efficient. Formal security proof and analysis have been provided to demonstrate the achieved security properties, including confidentiality, anonymity, mutual authentication, unlinkability, unforgeability, forward security, backward security, and non-repudiation. In addition, the performance analysis reveals that our proposed protocol significantly reduces the time consumption in batch verification, achieving high computational efficiency.

Influence Mechanism of Cyanobacterial Extracellular Polymeric Substances on the Water Quality in Dynamic Water Supply System

As a kind of high-organic-content contamination source, extracellular polymeric substances (EPS) secreted by cyanobacteria have become an important factor restricting the safety of supply water. In the dynamic batch mode water supply system, cyanobacterial EPS accelerated the decay rate of residual chlorine, resulting in a 21–26 times increase of the total viable bacteria count within 72 h. The water turbidity exceeded upper limit of the standards for drinking water quality within 4 h, with an increase of 306–332% within 72 h. The biological stability was reduced with BDOC and AOC increased by 41.4–43.8% and 331–396%, respectively. The main cause is that cyanobacterial EPS act as nutrients and metabolic energy for microorganisms, promoting their metabolic activity and secretion of extracellular organic components. This leads to the metabolic accumulation of tryptophan, fulvic acids and humic acids in the pipeline, thus further promoting the regeneration of bacteria. Compared with the influence of biofilm on pipe wall caused by long-term use of water supply network, the contribution of cyanobacterial EPS to the water contamination of pipe network is increased several times. Therefore, even in the presence of residual chlorine, the secondary contamination caused by cyanobacterial EPS in the water supply system could not be neglected.

The role of nanoparticles (titanium dioxide, graphene oxide) on the inactivation of co-existing bacteria in the presence and absence of quartz sand.

The increased mass production and application of engineered nanomaterials (ENMs) have resulted in the release of nanoparticles (NPs) in the environment, raising uncertainties regarding their environmental impacts. This study examines the effect of graphene oxide (GO) and titanium dioxide (TiO2) NPs on the inactivation of the three model bacteria originated by mammalians including humans: Escherichia (E.) coli, Enterococcus (E.) faecalis, and Staphylococcus (S.) aureus. A series of dynamic batch experiments were conducted at constant room temperature (22°C) in order to examine the inactivation of co-existing bacteria by NPs, in the presence and absence of quartz sand. The inactivation experimental data were satisfactorily fitted with a pseudo-first order expression with a time dependent rate coefficient. The inactivation of E. coli and S. aureus was shown to increase in the co-presence of GO or TiO2 NPs and quartz sand comparing with the presence of GO or TiO2 NPs alone. For E. faecalis, no clear trend was observed. Moreover, quartz sand was shown to affect inactivation of bacteria by GO and TiO2 NPs. Among the bacteria examined, the highest inactivation rates were observed for S. aureus.

Removal Performance of Faecal Indicators by Natural and Silver-Modified Zeolites of Various Particle Sizes under Dynamic Batch Experiments: Preliminary Results

One of the oldest and most promising applications of natural zeolites (NZs) is in water and wastewater treatment processes. Modified zeolites (MZs), with improved ion exchange and adsorption capacities, have been extensively applied to the removal of pollutants from aqueous solutions. However, the application of MZs in pathogens or indicator organisms has not been extensively explored. This study examines the effect of both natural Greek zeolite (NZ), with a clinoptilolite content of up to 85% (OLYMPUS SA-INDUSTRIAL MINERALS), and modified Greek zeolite through incorporation with silver ions (Ag-MNZ), on the survival of two selected faecal indicator bacteria (Escherichia coli and Enterococcus faecalis). A series of dynamic batch experiments with a slow agitation of 12 rpm were conducted at a constant ambient temperature (22°C) in order to examine the inactivation of the above bacteria by NZ and Ag-MNZ. It was found that the Ag-MNZ resulted in a much higher reduction in the bacterial numbers when compared to the NZ and the control (absence of zeolites). Moreover, the reduction in bacterial numbers was affected by NZ particle size, with higher removal rates observed for coarse (1–3 mm) than for fine (0–1 mm) NZ. Finally, the E. faecalis was found to be more resistant than E. coli to Ag-MNZ.

Denitrification of leachate using composted domestic waste at different levels of stability: A batch test investigation

This study presents an investigation into the viability of pre-treated general wastes at different degrees of stability as carbon sources for in-situ bio-denitrification at landfills. Domestic waste was composted and stabilised for eight (8) and sixteen (16) weeks within two different mini-landfill cells located at the Bisasar Road landfill, Durban, South Africa. Eight substrate categories were developed using the composted domestic wastes and commercial garden refuse with stand-alone and combinatory approaches. Adopting small-scale dynamic batch tests, landfill leachate, treated with 500 mg/l nitrate concentration level, was used to comparatively assess the denitrification efficiency of the stand-alone and augmented substrates. Results demonstrate that substrates based on domestic waste were unable to independently sustain the denitrification process. Full denitrification was only achieved on the augmented substrates. Kinetic analyses show that a zero-order reaction better describes the denitrification rate independent of the nitrate concentration. These results thus elucidate the benefits inherent in adopting augmented substrates in treating landfill leachate.

Batch process monitoring based on global enhanced multiple neighborhoods preserving embedding

Batch process generally has varying dynamic characteristic that causes low fault detection rate and high false alarm rate, and it is necessary and urgent to monitor batch process. This paper proposes a global enhanced multiple neighborhoods preserving embedding based fault detection strategy for dynamic batch process. Firstly, the angle neighbor is defined and selected to compensate for the insufficient expression for the spatial similarity of samples only by using the distance neighbor, and the time neighbor is introduced to describe the time correlations between samples. These three types of neighbors can fully characterize the similarity of the samples in time and space. Secondly, considering the minimum reconstruction error and the order information of three types of neighbors, an enhanced objective function is constructed to prevent the loss of order information when neighborhood preserving embedding (NPE) calculates the reconstruction weights. Furthermore, the enhanced objective function and a global objective function are organically combined to extract both global and local features, to describe process dynamics and visualize process data in a low-dimensional space. Finally, a monitoring index based on support vector data description is constructed to eliminate adverse effects of non-Gaussian data for monitoring performance. The advantages of the proposed method over principal component analysis, neighborhood preserving embedding, dynamic principal component analysis and time NPE are demonstrated by a numerical example and the penicillin fermentation process simulation.

A general optimization protocol for molecular property prediction using a deep learning network.

The key to generating the best deep learning model for predicting molecular property is to test and apply various optimization methods. While individual optimization methods from different past works outside the pharmaceutical domain each succeeded in improving the model performance, better improvement may be achieved when specific combinations of these methods and practices are applied. In this work, three high-performance optimization methods in the literature that have been shown to dramatically improve model performance from other fields are used and discussed, eventually resulting in a general procedure for generating optimized CNN models on different properties of molecules. The three techniques are the dynamic batch size strategy for different enumeration ratios of the SMILES representation of compounds, Bayesian optimization for selecting the hyperparameters of a model and feature learning using chemical features obtained by a feedforward neural network, which are concatenated with the learned molecular feature vector. A total of seven different molecular properties (water solubility, lipophilicity, hydration energy, electronic properties, blood–brain barrier permeability and inhibition) are used. We demonstrate how each of the three techniques can affect the model and how the best model can generally benefit from using Bayesian optimization combined with dynamic batch size tuning.

Proactive Fragmentation Management Scheme Based on Crosstalk-Avoided Batch Processing for Spectrally-Spatially Elastic Optical Networks

Fragmentation with crosstalks is the major obstacle in spectrally-spatially elastic optical networks, which suppresses resource utilization while degrading the quality-of-transmission. To overcome this issue, this article proposes, for the first time, a proactive fragmentation management scheme based on batch processing while satisfying both inter-core and inter-mode crosstalks to enhance resource utilization. The proposed scheme adopts a batch processing method to create batches of lightpath requests received within a time threshold to utilize spectrum resources effectively. In batch processing, lightpath requests are prioritized based on the number of links in their routes and required slots. To maintain fairness in batch processing, when any request is rejected, the proposed scheme triggers a procedure that gives an equal opportunity to all arriving requests within the threshold, irrespective of numbers of hops and requested capacities, for allocation. We formulate the static batch processing of lightpath requests (SBPLR) as an integer linear programming (ILP) problem. We prove that SBPLR is an NP-Complete problem. We introduce a heuristic solution when ILP is intractable. To serve lightpath requests in each batch while avoiding inter-core and inter-mode crosstalks, we develop a core-mode-spectrum allocation algorithm. We present a dynamic batch processing based fragmentation management approach. Numerical results indicate that the proposed scheme outperforms the benchmark schemes.

Dynamic ticket pricing of airlines using variant batch size interpretable multi-variable long short-term memory

Research of airlines shows that seat inventory control and therefore, revenue management is based not on a systematic analysis but more on human judgement. Machine learning models have been developed and applied to support decisions for ticket pricing dynamically. However, conventional models and approaches yield low statistical evaluation scores. In this study, the features used in other studies were explored and the cost available seat kilometer (CASK) value and target revenue features were included for the first time to the best of our knowledge which are essential components of ticket price decision. Real data from a low-cost carrier airline in Turkey were collected and the observation data were splitted into two to study with the highest profit sale data. Then the outliers were filtered to let the models learn from and generate better price offerings businesswise. Observation datasets obtained in each step were recorded to be tested. 7 different model techniques were simulated and tested with 4 different datasets according to 6 different statistical evaluation criteria. A new approach to Interpretable Multi-Variable Long Short-Term Memory (IMV-LSTM) model was proposed by taking every flight and its sales as an independent series, that is to assign a dynamic batch size. Extensive experiments on real datasets reveal enhanced statistical evaluation scores by using the proposed approach and model. The proposed model can be used by the airlines to mitigate human judgement on ticket pricing, to manage their price offerings to reach their target revenues and to increase their profits. The model can be used by other business cases that have similar historical data as overlapping windows structure.

Messages Classification and Dynamic Batch Verification Scheme for VANETs

Under a high-density traffic environment in a vehicular ad hoc network (VANET), many safety messages will not be verified in time and then get dropped. To touch such an issue appropriately, we aim at improving the efficiency of message verification and the number of verified safety messages. In our proposed scheme, messages are properly classified and prioritized first so that more important basic safety messages can be verified earlier. To improve batch verification, a dynamic batch verification scheme with an adjustable batch size is designed accordingly. The batch size is halved when failures of batch verification occur consecutively over a certain threshold to improve the efficiency of batch verification. On the contrary, the batch size is doubled when successes of batch verification occur consecutively over a certain threshold to increase the messages in a batch to be verified. For a failure of batch verification, a divide-and-conquer approach is utilized to find the illegitimate message(s). Via analysis, the superiority of our design over some closely related schemes in terms of verification delay and number of verified basic safety messages etc. is successfully demonstrated. By simulation, performance of our design and some closely related schemes is further investigated in realistic scenarios.

A time-independent trajectory optimization approach for connected and autonomous vehicles under reservation-based intersection control

Rule-based and optimization-based autonomous intersection management (AIM) policies have been evaluated against traditional signal controls in terms of intersection safety, efficiency and emission. As one of AIM policies, reservation-based control has further taken advantage of the benefits of AIM, especially via optimization approaches. This paper presents a time-independent trajectory optimization approach for connected and autonomous vehicles under reservation-based intersection control. The existing approaches assign an arrival time and speed to vehicles ahead of entering the intersection. However, the vehicles may not follow the planned trajectory once the traffic condition varies sharply and thus the trajectory solution becomes infeasible with respect to the assigned arrival time and speed. The proposed approach aims to solve the fail-follow problem by separating the optimization between arrival time, speed, and trajectory planning by optimizing the trajectory without arrival time and speed predetermined. The approach finds the optimal solution in terms of the intersection efficiency meanwhile keeps the feasibility of trajectory planning by formulating the variation of acceleration rate and breaking a whole trajectory into an enlarged set of segments. Two different control strategies, BATCH and ZONE, are also proposed to test the performance of the optimization approach in comparison with another Dynamic Batch strategy. The results validate that the proposed approach can adapt to extremely high traffic demand scenario. Sensitivity analyses also evaluate the performance of the proposed approach under different problem settings in terms of intersection efficiency.

Dynamic batch size tuning based on stopping criterion for neural network training

In the neural network training, the selection of the minibatch size affects not only the computational cost but also the training performance, as it underlies the loss function. Generally, an approach based on increasing the batch size according to the linear and step functions during the training process is known to be effective in improving the generalization performance of a network. However, it requires specifying the way of increasing the batch size beforehand. In this study, we propose a more flexible method that implies temporarily varying a small batch size to destabilize the loss function when a change in the training loss satisfies the predefined stopping criterion. Repeating the destabilization step allows a parameter to avoid being trapped at the local minima and to converge at a robust minimum, thereby improving generalization performance. We experimentally demonstrate the superiority of the proposed method considering several benchmark datasets and neural network models.

A Fast Filtering Mechanism to Improve Efficiency of Large-Scale Video Analytics

Surveillance cameras are ubiquitous around us. Emerging full-feature object-detection models can analyze surveillance videos with high accuracy but consume much computation. Directly applying these models for practical scenarios with large-scale cameras is prohibitively expensive. This, however, is wasteful and unnecessary considering that user-defined anomalies occur rarely among these videos. Therefore, we propose FFS-VA, a multi-stage Fast Filtering Mechanism for Video Analytics, to make video analytics much cost-effective. FFS-VA filters out the frames without the user-defined events by two stream-specialized filters and a cheap full-function model, to reduce the number of frames reaching the full-feature model. FFS-VA presents a global feedback-queue approach to balance the processing speeds of different filters in intra-stream and inter-stream processes. FFS-VA designs a dynamic batch technique to achieve a trade-off between throughput and latency. FFS-VA can also efficiently scale to multiple GPUs. We evaluate FFS-VA against the state-of-the-art YOLOv3 under the same hardware and video workloads. The experimental results show that under a 12.88 percent target-object occurrence rate on two GPUs, FFS-VA can support up to 30 concurrent video streams (15× more than YOLOv3) in the online case, and obtain 10× speedup when offline analyzing a stream, with an accuracy loss of less than 2 percent.

Lead Time Independent Joint Inventory and Pricing Controls for Lost Sales System with Probabilistic Service Level Constraints

We consider the problem of joint inventory and pricing control in a finite-horizon setting with Poisson demand, positive lead time, and lost sales. We study the performance of a lead time independent heuristic control, which we call Dynamic Batch Pricing (DBP), and show that it is asymptotically optimal when the annual expected demand is large. Moreover, with a proper tuning of the control parameters, we also show that DBP has a robust performance with respect to the magnitude of lost sales cost, which is important in applications such as retails where the targeted service level is very high (equivalently, the lost sales cost is very large). Our result for the lost sales system complements the recent result of Chen et al. (2019), who show that a lead time independent heuristic can be near optimal in the backorder system with a large lead time. Together, both our result and Chen et al. (2019) reinforce the message that simple lead time independent heuristics can perform sufficiently well, at least in some practically relevant cases if not in all cases. This insight is important given that the task of solving the joint inventory and pricing problem (under either backorder or lost sales system) with positive lead time for the most general model remains very challenging.

In silico evolution of Aspergillus niger organic acid production suggests strategies for switching acid output

BackgroundThe fungus Aspergillus niger is an important industrial organism for citric acid fermentation; one of the most efficient biotechnological processes. Previously we introduced a dynamic model that captures this process in the industrially relevant batch fermentation setting, providing a more accurate predictive platform to guide targeted engineering. In this article we exploit this dynamic modelling framework, coupled with a robust genetic algorithm for the in silico evolution of A. niger organic acid production, to provide solutions to complex evolutionary goals involving a multiplicity of targets and beyond the reach of simple Boolean gene deletions. We base this work on the latest metabolic models of the parent citric acid producing strain ATCC1015 dedicated to organic acid production with the required exhaustive genomic coverage needed to perform exploratory in silico evolution.ResultsWith the use of our informed evolutionary framework, we demonstrate targeted changes that induce a complete switch of acid output from citric to numerous different commercially valuable target organic acids including succinic acid. We highlight the key changes in flux patterns that occur in each case, suggesting potentially valuable targets for engineering. We also show that optimum acid productivity is achieved through a balance of organic acid and biomass production, requiring finely tuned flux constraints that give a growth rate optimal for productivity.ConclusionsThis study shows how a genome-scale metabolic model can be integrated with dynamic modelling and metaheuristic algorithms to provide solutions to complex metabolic engineering goals of industrial importance. This framework for in silico guided engineering, based on the dynamic batch growth relevant to industrial processes, offers considerable potential for future endeavours focused on the engineering of organisms to produce valuable products.

Multiphase two-dimensional time-slice dynamic system for batch process monitoring

In this work, a multiphase 2D time-slice dynamic system approach has been proposed to characterize the batch-wise and variable-wise dynamics of batch processes simultaneously. First, in each phase, the 2D latent representation has been developed for time slice matrices so as to extract informative features from both batch-wise and variable-wise variations. Afterward, the first-order Markov chains are introduced for capturing the underlying 2D latent dynamics across time-slice evolutions. In this way, batch process dynamic behaviors in each working phase can be captured from both single batch run and batch-to-batch evolutions, while process noise can also be explained in the probabilistic modeling framework. Last, in order to carry out online monitoring with the proposed 2D dynamic system, an effective 2D online-monitoring mechanism is established. The feasibility of the dynamic batch process monitoring method is validated on the fed-batch penicillin fermentation benchmark.

Surface Modified of Cellulose Acetate Electrospun Nanofibers by Polyaniline/β-cyclodextrin Composite for Removal of Cationic Dye from Aqueous Medium

This paper concerns with the enhancement of a safe and clean procedure for remediation of water pollution. Cellulose acetate (CA) nanofibers were prepared via electrospinning technique prior to surface modification with polyaniline/β-cyclodextrin (PANI/β-CD) nanocomposite. Morphology and structure of the modified CA nanofibers were characterized by different analysis like X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), scanning electron microscopy (SEM) and Thermogravimetric analyses (TGA). We studied the influence of modification of CA with PANI/β-CD nanocomposite for removal of cationic dyes from colored aqueous solution in dynamic batch mode. Methylene blue (MB) dye was selected as a cationic dye model for the study. The effects of various parameters like pH, nanosorbent dose, dye concentration and reaction time on the adsorption capacity were studied. Some mathematical isotherm and kinetic models have been utilized to investigate the adsorption mechanism of MB into the CA-PANI/β-CD nanofibers. The reusability of the CA-PANI/β-CD nanofibers was also verified in sequent cycles and rapid process was detected. Different real water samples like tap water, industrial wastewater, municipal wastewater and sea water were applied to test the MB removal efficiency.