Optimal Bin Research Articles

Heuristics Integrated Deep Reinforcement Learning for Online 3D Bin Packing

Online 3D Bin Packing Problem (3D-BPP) has a wide range of industrial applications and there is an emerging research interest in learning optimal bin packing policy and deploying it for real logistics applications. From the heuristic methods to the deep reinforcement learning (DRL) methods, the previous works have proposed many solutions to solve the online 3D-BPP. However, none of them have studied what and how heuristics can be modelled into DRL to build a more effective and practical bin packing pipeline. In this work, we thoroughly investigate what heuristics can be used in online 3D-BPP and how to effectively integrate the heuristics with the DRL. First, we design 3 different heuristics based on the physical rules of the real world and the experiences of the human packers, including the Physics-Heuristics, the Packing-Heuristics and the Unpacking-Heuristics. Second, we model the 3 types of heuristics into the DRL framework and propose a novel heuristic DRL method to solve the online 3D-BPP. Extensive experimental results show that our method achieves state-of-the-art bin packing performance and the resulting real-world system is able to reliably finish the bin packing task in real logistics scenarios. Supplementary video is available at https://www.youtube.com/watch?v=x8GpmEELq18. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —The rapid growth of e-commerce has significantly increased the burden of human packers in logistic warehouses, where the workers need to pick the products from a conveyor and pack them into bins (i.e. the online 3D bin packing). Thus it is of great importance to develop intelligent robotic systems to replace human labor, which is a long-standing topic in the field of control and automation science. This paper makes a substantial contribution to the related field by studying the online 3D bin packing in terms of both the theory and practice. On the one hand, the simulated experiments suggest that the presented algorithm significantly improves the space utilization of bin packing. On the other hand, the robotic system developed based on the proposed method can favourably finish the bin packing task in real logistics scenarios, demonstrating the practical use of our approach. Consequently, the approach proposed in this paper is totally applicable in logistic warehouses and is promising to drastically improve the working efficiency of the product packing in real warehouses. In the future, we will extend the presented approach to pack irregular-shaped objects and then facilitate more logistics applications.

Optimal Online Bin Packing Algorithms for Some Cases with Two Item Sizes

Optimal Online Bin Packing Algorithms for Some Cases with Two Item Sizes

Improved Tomographic Binning of 3 × 2 pt Lens Samples: Neural Network Classifiers and Optimal Bin Assignments

Large imaging surveys, such as the Legacy Survey of Space and Time, rely on photometric redshifts and tomographic binning for 3 × 2 pt analyses that combine galaxy clustering and weak lensing. In this paper, we propose a method for optimizing the tomographic binning choice for the lens sample of galaxies. We divide the CosmoDC2 and Buzzard simulated galaxy catalogs into a training set and an application set, where the training set is nonrepresentative in a realistic way, and then estimate photometric redshifts for the application sets. The galaxies are sorted into redshift bins covering equal intervals of redshift or comoving distance, or with an equal number of galaxies in each bin, and we consider a generalized extension of these approaches. We find that bins of equal comoving distance produce the highest dark energy figure of merit of the initial binning choices, but that the choice of bin edges can be further optimized. We then train a neural network classifier to identify galaxies that are either highly likely to have accurate photometric redshift estimates or highly likely to be sorted into the correct redshift bin. The neural network classifier is used to remove poor redshift estimates from the sample, and the results are compared to the case when none of the sample is removed. We find that the neural network classifiers are able to improve the figure of merit by ∼13% and are able to recover ∼25% of the loss in the figure of merit that occurs when a nonrepresentative training sample is used.

Comparison of three different deconvolution methods for analyzing nanoindentation test data of hydrated cement paste

The micromechanical properties of hydrated cement paste were investigated by nanoindentation with a large data set (4000 indentation points) to reveal the influences of multiple factors on analysis results. Three different deconvolution analysis methods, i.e. the Gaussian mixture model (GMM) with the maximum likelihood evaluation (MLE) algorithm, the probability distribution function (PDF) and the cumulative distribution function (CDF) with least square estimate (LSE) algorithm, were employed to analyze the nanoindentation test data. It was found that the GMM with a diagonal-unshared matrix is the most appropriate for the deconvolution of nanoindentation data of hydrated cement paste for the MLE algorithm. The PDF and CDF methods are equally effective for the LSE algorithm, but the former is subjected to the influences of bin sizes and the optimal bin size is 1.5–2 GPa for elastic modulus and 0.075–0.1 GPa for hardness. The threshold number of indentation points necessary to obtain reliable micromechanical parameters and phase contents by all three deconvolution methods is approximately 800-1000, significantly higher than the values used in the literature. The phase content seems to show more variations than elastic modulus and hardness when the number of indentation point is insufficient. With sufficient number of indentation points, the three different deconvolution methods give consistent results regarding the properties and contents of six phases identified. The various phase contents calculated by deconvolution of nanoindentation test data are in reasonable agreement with that estimated by QXRD and SEM except for the pore phase.

Evaluation and Analysis of Absence of Homozygosity (AOH) Using Chromosome Analysis by Medium Coverage Whole Genome Sequencing (CMA-seq) in Prenatal Diagnosis.

Absence of homozygosity (AOH) is a genetic characteristic known to cause human diseases mainly through autosomal recessive or imprinting mechanisms. The importance and necessity of accurate AOH detection has become more clinically significant in recent years. However, it remains a challenging task for sequencing-based methods thus far. In this study, we developed and optimized a new bioinformatic algorithm based on the assessment of minimum sequencing coverage, optimal bin size, the Z-score threshold of four types of allele count and the frequency for accurate genotyping using 28 AOH negative samples, and redefined the AOH detection cutoff value. We showed the performance of chromosome analysis by five-fold coverage whole genome sequencing (CMA-seq) for AOH identification in 27 typical prenatal/postnatal AOH positive samples, which were previously confirmed by chromosomal microarray analysis with single nucleotide polymorphism array (CMA/SNP array). The blinded study indicated that for all three forms of AOH, including whole genomic AOH, single chromosomal AOH and segmental AOH, and all kinds of sample types, including chorionic villus sampling, amniotic fluid, cord blood, peripheral blood and abortive tissue, CMA-seq showed equivalent detection power to that of routine CMA/SNP arrays (750K). The subtle difference between the two methods is that CMA-seq is prone to detect small inconsecutive AOHs, while CMA/SNP array reports it as a whole. Based on our newly developed bioinformatic algorithm, it is feasible to detect clinically significant AOH using CMA-seq in prenatal diagnosis.

Method of cutting metal ingot with changing direction of deformation

The paper considers the priority problem of cutting an ingot with a changing direction of rolling which can be presented as a priority problem of packaging with incompatible categories and dynamically changing bin sizes. This problem is a NP-complete optimization problem which is often found in the production of metalwork. Priorities reflect the “importance” of manufacturing elements for production and incompatible categories show that elements of different thicknesses cannot be made from the same bin. In addition, we use guillotine restrictions to take into account the specifics of production. The problem of cutting the ingot is characterized by the fact that the bin sizes are not known in advance. Thus, the task is to find the optimal bin sizes, the sequence of actions and the placement of elements in these bins. We propose a sequential tree metaheuristic to solve this problem. It is a deterministic method. This method is based on sequential metaheuristic which was proposed by the authors earlier. The proposed method also has the ideas of dividing a set of elements into groups and subgroups by thickness and priorities, as well as their sequential packing. However, it differs from previous one in that it uses a tree to represent options for rolling operations. The best subtree from the point of view of any criterion, formed as a path from the root to one of the nodes, will be the solution. The ability to find several better solutions with the same value of the criterion is another advantage of the proposed method. Such a situation may arise, for example, for a square ingot. In this case, two solutions are possible with accuracy up to 90 degrees rotation of the workpieces. The priority heuristic is used as a heuristic to solve the packaging subproblem. The results of computational experiments are based on test cases which were compiled specially. Test instances differ in the number and characteristics of elements, categories, priorities. The results show that the metaheuristic proposed by the authors solves the problem effectively.

Determining minimum sample size for the conditioned Latin hypercube sampling algorithm

In digital soil mapping (DSM), a fundamental assumption is that the spatial variability of the target variable can be explained by the predictors or environmental covariates. Strategies to adequately sample the predictors have been well documented, with the conditioned Latin hypercube sampling (cLHS) algorithm receiving the most attention in the DSM community. Despite advances in sampling design, a critical gap remains in determining the number of samples required for DSM projects. We propose a simple workflow and function coded in R language to determine the minimum sample size for the cLHS algorithm based on histograms of the predictor variables using the Freedman-Diaconis rule for determining optimal bin width. Data preprocessing was included to correct for multimodal and non-normally distributed data, as these can affect sample size determination from the histogram. Based on a user-selected quantile range (QR) for the sample plan, the densities of the histogram bins at the upper and lower bounds of the QR were used as a scaling factor to determine minimum sample size. This technique was applied to a field-scale set of environmental covariates for a well-sampled agricultural study site near Guelph, Ontario, Canada, and tested across a range of QRs. The results showed increasing minimum sample size with an increase in the QR selected. Minimum sample size increased from 44 to 83 when the QR increased from 50% to 95% and then increased exponentially to 194 for the 99% QR. This technique provides an estimate of minimum sample size that can be used as an input to the cLHS algorithm.

Multiple Histograms-Based Reversible Data Hiding Using Fast Performance Optimization and Adaptive Pixel Distribution

Abstract In prediction error-based reversible data hiding, multiple histograms modification (MHM) is well known for high image quality and thus has received wide attention in recent years. However, the computational cost for performance optimization in MHM is too high, which is particularly critical for real-time applications. This manuscript aims to reduce the computational complexity of MHM by presenting two techniques, including fast performance optimization and adaptive pixel distribution. Fast performance optimization provides a two-stage process for optimal bin selection by exploiting the concept of per-bit distortion of data embedding within a prediction error histogram (PEH). In fast performance optimization, the distribution characteristics of the per-bit distortion are investigated to significantly narrow down the solution space of optimal bin selection. The second technique is adaptive pixel distribution, which tries to nonuniformly allocate pixels into multiple PEHs to further reduce the time complexity. Extensive experiments show that the computational complexity of MHM is significantly reduced while well preserving the image quality.

Recognition of unknown wafer defect via optimal bin embedding technique

Recognition of unknown wafer defect via optimal bin embedding technique

Optimization and inference of bin widths for histogramming inelastic neutron scattering spectra.

A data-driven bin-width optimization for the histograms of measured data sets based on inhomogeneous Poisson processes was developed in a neurophysiology study [Shimazaki & Shinomoto (2007). Neural Comput. 19, 1503-1527], and a subsequent study [Muto, Sakamoto, Matsuura, Arima & Okada (2019). J. Phys. Soc. Jpn, 88, 044002] proposed its application to inelastic neutron scattering (INS) data. In the present study, the results of the method on experimental INS time-of-flight data collected under different measurement conditions from a copper single crystal are validated. The extrapolation of the statistics on a given data set to other data sets with different total counts precisely infers the optimal bin widths on the latter. The histograms with the optimized bin widths statistically verify two fine-spectral-feature examples in the energy and momentum transfer cross sections: (i) the existence of phonon band gaps; and (ii) the number of plural phonon branches located close to each other. This indicates that the applied method helps in the efficient and rigorous observation of spectral structures important in physics and materials science like novel forms of magnetic excitation and phonon states correlated to thermal conductivities.

A New Method (MINDED-BA) for Automatic Detection of Burned Areas Using Remote Sensing

This work presents a change detection method (MINDED-BA) for determining burned extents from multispectral remote sensing imagery. It consists of a development of a previous model (MINDED), originally created to estimate flood extents, combining a multi-index image-differencing approach and the analysis of magnitudes of the image-differencing statistics. The method was implemented, using Landsat and Sentinel-2 data, to estimate yearly burn extents within a study area located in northwest central Portugal, from 2000–2019. The modelling workflow includes several innovations, such as preprocessing steps to address some of the most important sources of error mentioned in the literature, and an optimal bin number selection procedure, the latter being the basis for the threshold selection for the classification of burn-related changes. The results of the model have been compared to an official yearly-burn-extent database and allow verifying the significant improvements introduced by both the pre-processing procedures and the multi-index approach. The high overall accuracies of the model (ca. 97%) and its levels of automatization (through open-source software) indicate potential for being a reliable method for systematic unsupervised classification of burned areas.

The LSST-DESC 3x2pt Tomography Optimization Challenge

This paper presents the results of the Rubin Observatory Dark Energy Science Collaboration (DESC) 3x2pt tomography challenge, which served as a first step toward optimizing the tomographic binning strategy for the main DESC analysis. The task of choosing an optimal tomographic binning scheme for a photometric survey is made particularly delicate in the context of a metacalibrated lensing catalogue, as only the photometry from the bands included in the metacalibration process (usually riz and potentially g) can be used in sample definition. The goal of the challenge was to collect and compare bin assignment strategies under various metrics of a standard 3x2pt cosmology analysis in a highly idealized setting to establish a baseline for realistically complex follow-up studies; in this preliminary study, we used two sets of cosmological simulations of galaxy redshifts and photometry under a simple noise model neglecting photometric outliers and variation in observing conditions, and contributed algorithms were provided with a representative and complete training set. We review and evaluate the entries to the challenge, finding that even from this limited photometry information, multiple algorithms can separate tomographic bins reasonably well, reaching figures-of-merit scores close to the attainable maximum. We further find that adding the g band to riz photometry improves metric performance by ~15% and that the optimal bin assignment strategy depends strongly on the science case: which figure-of-merit is to be optimized, and which observables (clustering, lensing, or both) are included.

Analyzing the gamma-ray activity of neutrino emitter candidates: comparing TXS 0506+056 with other blazars

Abstract On 22nd of September 2017 the IceCube collaboration sent an alert for an EHE (Extreme High Energy) event, corresponding to the reconstruction of a muonic neutrino (IC-170922A) with energy ∼290 TeV. A multi-wavelength follow-up campaign associated this neutrino event with a gamma-ray flaring state of the BL Lac TXS 0506+056 located at z=0.3365. From the same position of the sky a muonic neutrino excess is observed in a time window of 110 days around 13th of December 2014. These observations together suggest TXS 0506+056 as a possible neutrino emitter. We report here a long term gamma-ray monitoring of this source and we compare it with other blazars spatially correlated with astrophysical muonic neutrino events observed by IceCube. We characterize the most significant gamma-ray flares of the blazars in the sample and introduce the blazar duty cycle as an important parameter to be considered when assessing a possible neutrino counterpart. For the selected blazars we show the expected neutrino flux variability with different time binning assuming the gamma-rays observed by Fermi-LAT as a product of a leptohadronic emission scenario. The neutrino expectations from the blazar sample are then compared with the IceCube discovery flux searching for the optimal time bin in a multi-messenger context. This analysis indicates that the detection of a single flare represents a challenge for a single cubic kilometer detector, underlining the importance of increasing the size of current neutrino telescopes and obtaining a good monitoring of the entire sky through a global neutrino network.

On the Use of Equivalence Classes for Optimal and Suboptimal Bin Packing and Bin Covering

Bin packing and bin covering are important optimization problems in many industrial fields, such as packaging, recycling, and food processing. The problem concerns a set of items, each with its own value, that are to be sorted into bins in such a way that the total value of each bin, as measured by the sum of its item values, is not above (for packing) or below (for covering) a given target value. The optimization problem concerns minimizing, for bin packing, or maximizing, for bin covering, the number of bins. This is a combinatorial NP-hard problem, for which true optimal solutions can only be calculated in specific cases, such as when restricted to a small number of items. To get around this problem, many suboptimal approaches exist. This article describes the formulations of the bin packing and covering problems that allow finding the true optimum, for instance, counting hundreds of items using general-purpose MILP-solvers. Also presented are suboptimal solutions that come within less than 10% of the optimum while taking significantly less time to calculate, even ten to 100 times faster, depending on the required accuracy.

MDM-PCCT: Multiple Dynamic Modulations for High-Performance Spectral PCCT Imaging

Photon counting computed tomography (PCCT) has the ability to identify individual photons, resulting in quantitative material identification. Meanwhile, several technical challenges still exist in current PCCT imaging systems, including increased noise and suboptimal bin selection. These nonideal effects can substantially degrade the reconstruction performance and material estimation accuracy. To address these issues, in this work, we present a novel system for high-performance spectral PCCT imaging, which is a combination of multiple dynamic modulations, interpolation-based measurements processing strategy and advanced reconstruction method. For simplicity, this new PCCT imaging system is referred to as "MDM-PCCT". Specifically, the multiple dynamic modulations consist of dynamic kVp modulation, dynamic spectrum modulation and dynamic energy threshold modulation. In the dynamic kVp modulation, three kVp values, i.e., 80, 110 and 140, are included, and the tube voltage waveform follows a sinusoidal curve which is more practical than the rectangular curve in the fast kV switching mode. In the dynamic spectrum modulation, the X-ray spectra are processed by selective spatial-spectral filters to balance the X-ray fluxes and increase the spectral separation. In the dynamic energy threshold modulation, the energy threshold is adaptively changed to determine the optimal bin selection. Furthermore, we propose an energy threshold determination method and interpolation-based measurements processing strategy to address the issue of non-uniform and sparse-view PCCT measurements, respectively. In addition, by considering the intrinsic characteristics of the MDM-PCCT images, we utilize an enhanced total variation regularized model for images reconstruction. Finally, numerical and preclinical studies demonstrate that the presented MDM-PCCT imaging system is capable of yielding uniform and high-fidelity PCCT measurements with noise consistency, and the presented reconstruction method further improves the image quality and material decomposition accuracy.

A Study of TOC Supply Chain Replenishment System in LED Chip Manufacturing Plant

The LED chip manufacturing (LED-CM) is an important process in the LED supply chain. To meet the severely competitive pressure these years, the issue of inventory risk for the variety of products is an imperative task for the LED-CM plant. The TOC supply chain replenishment system (TOC-SCRS) is one of the potential solutions to effectively manage inventory in the industry. However, the special features of the unstable production output and a product composed of the chips of different Bins exist in the LED-CM plant. Therefore, the application of TOC-SCRS to the LED-CM plant will encounter the following issues: (1) How to determine the optimal inventory buffer of the chips in the different Bins because a product can be composed of different combination of the chips in the feasible Bins? This paper proposed optimal Bin allocation models to resolve this problem. The optimal Bins inventory combination of a case is further performed via LINGO system and utilized to demonstrate the feasibility of these models. Employing the proposed models will facilitate LED-CM plants to improve their throughput and competitiveness.

Inferring Neuronal Couplings From Spiking Data Using a Systematic Procedure With a Statistical Criterion.

Recent remarkable advances in experimental techniques have provided a background for inferring neuronal couplings from point process data that include a great number of neurons. Here, we propose a systematic procedure for pre- and postprocessing generic point process data in an objective manner to handle data in the framework of a binary simple statistical model, the Ising or generalized McCulloch-Pitts model. The procedure has two steps: (1) determining time bin size for transforming the point process data into discrete-time binary data and (2) screening relevant couplings from the estimated couplings. For the first step, we decide the optimal time bin size by introducing the null hypothesis that all neurons would fire independently, then choosing a time bin size so that the null hypothesis is rejected with the strict criteria. The likelihood associated with the null hypothesis is analytically evaluated and used for the rejection process. For the second postprocessing step, after a certain estimator of coupling is obtained based on the preprocessed data set (any estimator can be used with the proposed procedure), the estimate is compared with many other estimates derived from data sets obtained by randomizing the original data set in the time direction. We accept the original estimate as relevant only if its absolute value is sufficiently larger than those of randomized data sets. These manipulations suppress false positive couplings induced by statistical noise. We apply this inference procedure to spiking data from synthetic and in vitro neuronal networks. The results show that the proposed procedure identifies the presence or absence of synaptic couplings fairly well, including their signs, for the synthetic and experimental data. In particular, the results support that we can infer the physical connections of underlying systems in favorable situations, even when using a simple statistical model.

Selecting optimal bin size to account for growth variability in Electronic LEngth Frequency ANalysis (ELEFAN)

Electronic LEngth Frequency ANalysis (ELEFAN) is a widely used method to fit von-Bertalanffy Growth Function (VBGF) using length frequency data for data-poor fisheries. However accuracy of this method depends on diverse assumptions, among which the variability of individual growth may be particularly influential. In addition, the selection of class interval of length frequency data, defined as bin size, is also considerable source of uncertainty. However, rare guidance exists on how to account for the growth variability and to select bin size. In this study, we considered the effects of growth variability and bin size together by testing whether appropriate bin size can reduce estimating bias due to growth variability. We used a Monte Carlo simulation framework to evaluate the performances of ELEFAN under scenarios of different bin size and levels of individual growth variability for 10 fish species. The results indicated that growth variability might result in large variations and biased estimation of growth parameters if the bin size was chosen arbitrarily. On the contrary, there was an optimal bin size for each species, which led to unbiased estimates even in a substantially high level of growth variability. Summarizing the optimal bin sizes of the 10 species, we proposed a rule of thumb to select optimal bin size (OBS) for the future applications of ELEFAN according to the maximum body length (OBS=0.23×Lmax0.6) and age (OBS=1.86×(Lmax/A)0.45) of fish species. The improvement of accuracy of VBGF parameter estimation will help to understand the growth characteristics and reduce the uncertainty for data-limited stock assessment methods.

Enhanced MDT-Based Performance Estimation for AI Driven Optimization in Future Cellular Networks

Minimization of drive test (MDT) allows coverage estimation at a base station by leveraging measurement reports gathered at the user equipment (UE) without the need for drive tests. Therefore, MDT is a key enabling feature for data and artificial intelligence driven autonomous operation and optimization in future cellular networks. However, to date, the utility of MDT feature remains thwarted by issues such as sparsity of user reports and user positioning inaccuracy. We characterize three key types of errors in MDT-based coverage estimation that stem from inaccurate user positioning, scarcity of user reports and quantization. For the first time, the presented analysis shows existence of joint interplay between these errors on coverage estimation that result from inter-dependency between positioning error and bin width. The analysis also shows that there exists an optimal bin width for given user positioning inaccuracy and user density that minimizes the overall error in MDT-based estimated coverage. Utility of our framework is presented by addressing two applications from network optimization perspective: determining optimal bin width to maximize accuracy of MDT-based coverage estimation and its calibration to further improve its accuracy.

GenomomFF: Cost-Effective Method to Measure Fetal Fraction by Adaptive Multiple Regression Techniques With Optimally Selected Autosomal Chromosome Regions

The accurate measure of fetal fraction is important to ensure the results of noninvasive prenatal testing. However, measuring fetal fraction could require a substantial amount of data and additional costs. Therefore, this study proposes an alternative method of measuring fetal fraction with a limited sample size and low sequencing reads. Adaptive machine-learning algorithms customized to each laboratory’s environment, were used to measure fetal fraction. Pregnant women with female fetuses were tested to exclude the bias caused by training data of women carrying male fetuses. The accuracy of fetal DNA fraction prediction was enhanced by increasing the training sample size. When trained with 1,000 samples (males) and tested with 45 samples (females), the optimal bin sizes using the read count and size features were 300 kb and 800 kb, respectively. Comparing the new 300-kb bin to the 50-kb bin used by SeqFF with 4,000–5,000 training samples, the correlation was approximately 3-5% higher with the 300-kb bin. Therefore, we propose an effective and tailored method to measure fetal fraction in individual laboratories with limited sample collecting conditions and relatively low-coverage sequencing data.