Calculation Of Posterior Probabilities Research Articles

Off-grid DOA estimation for metasurface antenna systems using sparse Bayesian learning

Direction-of-arrival (DOA) estimation is a classic research topic with numerous applications in the field of signal processing. However, multiple radio frequency channels are required to receive array signals in order to achieve excellent performance, which is complicated in hardware to be applied to portable or cost-effective equipment. In this paper, we investigate the DOA estimation for transmissive metasurface antenna systems, where the signals are modulated by metasurface elements before reaching the receiving antennas. Then, we proposed an off-grid sparse Bayesian learning algorithm for metasurface aided systems, where the DOA estimation process consists of three stages, i.e., off-grid operation, calculation of posterior probability, and iterative update for hyperparameters using expectation maximization algorithm. Simulation results show that the proposed method can obtain superior estimation performance with a low number of snapshots and low signal-to-noise ratio. Meanwhile, it can achieve satisfactory performance even when the angle interval of the incoming signal is small.

Quantitative assessment of CO2 leakage risk in geologic carbon storage management

AbstractLarge‐scale geological storage of carbon dioxide (CO2) is indispensable for mitigating climate change but faces significant challenges, especially in the accurate quantitative assessment of leakage risks to ensure long‐term security. Given these circumstances, this paper proposes an innovative approach for quantitatively assessing CO2 leakage risk to address the previous limitations of limited accuracy and insufficient data. We construct a fault tree and transform it into a Bayesian network–directed acyclic graph, and then use judgment sets along with fuzzy set theory to obtain prior probabilities of root nodes. The feature, event, and process method was utilized to identify key components and subsequently determine the conditional probability table (CPT) of the leaf node. The subjective experience assessments from experts are defuzzified to obtain the CPTs of intermediate nodes. The obtained basic probability parameters are input into the directed acyclic graph to complete the model construction. After calculating the leakage probability using this model, it is combined with the severity of impacts to conduct a comprehensive risk assessment. Furthermore, critical CO2 risk sources can be determined through posterior probability calculations when intermediate nodes are designated as deterministic risk events. The gradual implementation process of the proposed model is demonstrated via a typical case study. The results indicate an overall CO2 leakage probability of 29%, with probabilities of leakage along faults/fractures, caprock, and well identified as 32%, 28%, and 19%, respectively. The project is categorized as a medium‐low risk level. When leakage is confirmed, tectonic movement, thickness, and delamination at interface connections/the presence of cracks are the critical risk sources, and measures to mitigate key risks are outlined. The identified key risk factors conform to empirical evidence and previous research, validating the accuracy of the model. This study is instrumental in CO2 geological storage risk assessment and scalable development program design. © 2024 Society of Chemical Industry and John Wiley & Sons, Ltd.

Updated Results from a Phase I Dose Escalation Study of the Rapidly-Switchable Universal CAR-T Therapy UniCAR-T-CD123 in Relapsed/Refractory AML

Introduction CAR-T efficacy in AML has been limited by a lack of ideal target antigens as well as the aggressive and resistant nature of disease. CD123 expression on leukemic blasts has been observed in about 80% of AML patients. However, the antigen is also expressed on non-malignant hematopoietic cells including neutrophils and endothelial cells. As consequence, targeting of CD123 with conventional CAR-T has been shown to result in persistent cytopenias, capillary leakage syndrome, CRS, and ICANS. AVC-101 (UniCAR02-T-CD123) is an adapter CAR-T, consisting of a universal CAR-T cell (UniCAR-T) and a CD123 targeting module (TM). The TM binds to the CAR-T cell via a peptide tag and induces T cell activity against CD123 bearing cells. Due to the short half-life of the TM, interruption of its continuous IV administration will rapidly switch T cell activity off, and mitigate toxicities as shown previously. Since the presentation of initial results, the study was amended to include a stricter and prolonged schedule of TM consolidation cycles. Given the excellent safety enabled by the switchable mechanism, additional dose levels (DL) of TM were added. Methods Key inclusion criteria included rrAML patients with ECOG ≤ 1, CD123 expression in ≥ 20% of blasts, and no suitable approved treatment option available or MRD positive AML patients. Initial dosing was applied using an adaptive Bayesian optimal interval design for drug-combination trials, escalating either cell dose or TM concentration based on posterior probability calculation of AEs. Dose escalation for the two additional dose levels was guided by a classical 3+3 design. Patients received standard Flu/Cy lymphodepletion (LD) for 3 days and TM was administered as continuous infusion, whereas UniCAR-T cells were given as one-time infusion on cycle 1 day 1. After completion of the initial 20-day induction cycle, responding patients were allowed to receive up to 3 consolidation cycles of 12 days continuous TM infusion with 7-14 days treatment free intervals in between cycles. Results As of this time 19 patients have received the induction cycle, 8 patients have also received at least one consolidation cycle (1-3, total 13 cycles). Patients were heavily pretreated with a median of 4 (2-7) prior treatment lines. Twelve of these patients had received a prior allogeneic (Allo) stem cell transplantation (SCT). Most of the patients (60%) were in adverse risk category according to ELN classification. In the initial escalation phase, 19 patients were treated in 16 dose cohorts. We observed a single DLT at DL 8 (reversible drop in fibrinogen levels, reported earlier) but no DLTs occurred in 11 patients treated in the yet highest dose cohort of 4 mg/day TM and 500 million UniCAR-T. Treatment was generally well tolerated irrespective of age and prior treatments. CRS was observed in 12 patients, mostly grade 1 or 2. Three grade 3 CRS and 1 grade 2 ICANS were reported, all resolving within 24 hrs from interruption of TM administration. Treatment-related AEs were most frequent in cycle 1, none were reported in cycles 3 and 4. No treatment-induced lasting myelosuppression was observed, and no patient required SCT support for WBC reconstitution. ORR in the first 19 patients was 53% (8/15) for the rrAML population and 75% (3/4) for the MRD population. Since the last presentation, extended administration of TM (20-day induction followed by three 12-day consolidation cycles 1-2 weeks apart) initiated by protocol amendment resulted in durable, ongoing responses approaching 5 months, longer than any prior. A 64-year-old MRD+ patient (#19) with 4 prior lines of therapy experienced a 3-log reduction in MRD levels (NPM1 mutation by qPCR) in both bone marrow and blood (Figure 1) and became transplant eligible. Re-administration of TM resulted in a significant re-expansion of UniCAR-T (Figure 2) without further cell dosing or LD. Another patient (#20) with overt hematological relapse after AlloSCT achieved a CR, ongoing at 5 months. Full safety and efficacy results of the additional 2 dose cohorts with prolonged TM consolidation will be presented at the meeting. Conclusions AVC-101 continues to be safe and tolerable. Repeated dosing of TM in consolidation cycles resulted in robust re-expansion of UniCAR-T with deeper remissions of extended durability. The switchable mechanism provides proven rapid reversal of safety events, enabling higher dose levels.

Polar Sea Ice Detection Using a Rotating Fan Beam Scatterometer

Scatterometers are dedicated to monitoring sea surface wind vectors, but they also provide valuable data for polar applications. As a new type of scatterometer, the rotating fan beam scatterometer delivers a higher diversity of incidence angles and more azimuth sampling. The paper takes the first rotating fan beam scatterometer, the China France Oceanography Satellite scatterometer (CSCAT), as an example to explore the effectiveness of this new type of scatterometer in polar sea ice detection. In this paper, a Bayesian method with consideration of geometric characteristics of CSCAT is developed for sea ice detection. The implementation of this method includes the definition of CSCAT backscatter space, an estimation of the sea ice Physical Model Function (GMF), a calculation of the sea ice backscatter distance to the sea ice GMF, a probability distribution function (PDF) estimation of the square distance to the GMF (sea ice GMF and wind GMF), and a calculation of the sea ice Bayesian posterior probability. This algorithm was used to generate a daily CSCAT polar sea ice mask during the CSCAT mission period (2019–2022) (by setting a 55% threshold on the Bayesian posterior probability). The sea ice masks were validated against passive microwaves by quantitatively comparing the sea ice edges and extents. The validation suggests that the CSCAT sea ice edge and extent show good agreement with the sea ice concentration distribution (i.e., sea ice concentration ≥ 15%) of the Advanced Microwave Scanning Radiometer 2 (AMSR2). The average Euclidean distance of the sea ice edges was basically less than 12.5 km, and the deviation of the sea ice extents was less than 0.3 × 106 km2.

Attribute fusion transfer for zero-shot fault diagnosis

In most industrial processes, it is impossible to collect samples of all fault modes, which greatly limits fault diagnosis applications in industrial fields. The fault diagnosis for the inaccessible fault modes having no labeled samples, termed zero-shot fault diagnosis, is quite challenging. This paper proposed an Attribute Fusion Transfer (AFT) for zero-shot fault diagnosis. First, we proposed a multi-Gaussian assumption for the posterior probability calculation in fault attribute transfer process and theoretically and experimentally proved its feasibility and effectiveness. Then, considering the ability difference of various attribute transfer methods to predict the same attribute, we proposed an attribute fusion network, which fuses the Bayesian fault attribute transfer and the attribute transfer under the multi-Gaussian assumption to attain better attribute transfer ability. Finally, AFT transfers fault attribute knowledge from accessible fault modes to inaccessible fault modes, thereby getting the ability to identify inaccessible fault modes. A series of ablation experiments and comparative experiments on the Tennessee–Eastman process were conducted to demonstrate the rationality and effectiveness of AFT fusion structure and the superiority of AFT. Furthermore, we conducted attribute-level and data-level interpretability analyses to reveal why our proposed AFT can obtain excellent zero-shot fault diagnosis ability. Codes are available at https://github.com/foryichuanqi/ADVEI-Paper-2023.11-Zero-shot-fault-diagnosis-by-attribute-fusion-transferGithub.

A Bayesian Approach to Network Monitoring for Progressive Failure Localization

Boolean Network Tomography (BNT) aims at identifying failures of internal network components by means of end-to-end monitoring paths. However, when the number of failures is not known a priori, failure identification may require a huge number of monitoring paths. We address this problem by designing a Bayesian approach that progressively selects the next path to probe on the basis of its expected information utility, conditioned on prior observations. As the complexity of the computation of posterior probabilities of node failures is exponential in the number of failed paths, we propose a polynomial-time greedy strategy which approximates these values. To consider aging of information in dynamic failure scenarios where node states can change during a monitoring period, we propose a monitoring technique based on a sliding observation window of adaptive length. By means of numerical experiments conducted on real network topologies we demonstrate the practical applicability of our approach, and the superiority of our algorithms with respect to state of the art solutions based on classic BNT as well as sequential group testing.

GNSS integer ambiguity posterior probability calculation with controllable accuracy

GNSS integer ambiguity posterior probability calculation with controllable accuracy

SEA v2.0: an R software package for mixed major genes plus polygenes inheritance analysis of quantitative traits

<p id="C3">The phenotypic values for quantitative trait from bi-parental segregation populations can be used to identify its mixed major genes plus polygenes inheritance model, which provides important information for the genetic basis of quantitative traits and crop breeding. To comprehensively summarize the research results of methodological advances, add the new functions of the software and correct its shortcomings in previous versions, an R software package SEA v2.0 with interactive graphical user interface is developed under R studio-1.4.1103 platform and R environment. In this software, there were 14 types of bi-parental segregation populations, and each type included four modules: data input, data analysis, posterior probability calculation, and distribution curve drawing. To save running time, doParallel was used to conduct parallel computing, data.table was used to quickly read and write datasets, and MASS was used to estimate the parameters in component distributions. KScorrect, kolmim, and shiny packages were used to simplify the programs. As long as users uploaded the data file with *.csv format and set the related parameters, the results could be quickly displayed. The software was validated by real data analysis of soybean podding habit and Monte Carlo simulation studies, and can be downloaded from https://cran.r-project.org/web/packages/SEA/index.html.

The Ci3+3 design for dual-agent combination dose-finding clinical trials

ABSTRACT We propose a rule-based statistical design for combination dose-finding trials with two agents. The Ci3 + 3 design is an extension of the i3 + 3 design with simple up-and-down decision rules comparing the observed toxicity rates and equivalence intervals that define the maximum tolerated dose combination. Ci3 + 3 consists of two stages to allow fast and efficient exploration of the dose-combination space. Statistical inference is restricted to a beta-binomial model for dose evaluation, and the entire design is built upon a set of fixed rules. We show via simulation studies that the Ci3 + 3 design exhibits similar and comparable operating characteristics to more complex designs utilizing model-based inferences. Implementation of Ci3 + 3 for practical trials is simple for the first stage, where the up-and-down decisions may be carried out using a decision table based on the preselected escalation path and i3 + 3. The second stage is not simpler than model-based designs, however, since it also requires computation of posterior probabilities based on a Bayesian model. We believe that the Ci3 + 3 design may provide an alternative choice to help simplify the design and conduct of combination dose-finding trials in practice.

Model Selection Based Algorithm in Neonatal Chest EIT.

This paper presents a new method for selecting a patient specific forward model to compensate for anatomical variations in electrical impedance tomography (EIT) monitoring of neonates. The method uses a combination of shape sensors and absolute reconstruction. It takes advantage of a probabilistic approach which automatically selects the best estimated forward model fit from pre-stored library models. Absolute/static image reconstruction is performed as the core of the posterior probability calculations. The validity and reliability of the algorithm in detecting a suitable model in the presence of measurement noise is studied with simulated and measured data from 11 patients. The paper also demonstrates the potential improvements on the clinical parameters extracted from EIT images by considering a unique case study with a neonate patient undergoing computed tomography imaging as clinical indication prior to EIT monitoring. Two well-known image reconstruction techniques, namely GREIT and tSVD, are implemented to create the final tidal images. The impacts of appropriate model selection on the clinical extracted parameters such as center of ventilation and silent spaces are investigated. The results show significant improvements to the final reconstructed images and more importantly to the clinical EIT parameters extracted from the images that are crucial for decision-making and further interventions.

Shaping Distribution Identification of Phase Rotated Probabilistically Shaped Signals With Radius-Based Expectation Maximization

A radius-based expectation maximization (R-EM) algorithm is proposed for shaping distribution identification (SDI) with phase rotated probabilistically shaped (PS) signals in a flexible transceiver. Since the proposed SDI algorithm is independent of carrier phase, shaping distribution (SD)-aware algorithms for frequency offset estimation and carrier phase recovery can be implemented after SDI to fully leverage the shaping gain. To remove the impact of the carrier phase on the calculation of posterior probability, the channel is modeled as a function of the radius rather than the complex symbol as in the conventional EM SDI. An approximation method is employed to analytically derive the parameter updating equation. A low-complexity initialization step is also proposed for the R-EM algorithm. The performance of the proposed algorithm is investigated numerically via Monte-Carlo simulation of a polarization division multiplexing PS-64QAM system. The optimization of the R-EM algorithm and the impacts of the shaping granularity, the block length and the signal-to-noise ratio (SNR) estimation accuracy on the success rates are analyzed. In the case of ideal SNR estimation, high success rates (≥99.6%) are achieved at operation points (i.e., SNR threshold at the normalized generalized mutual information of 0.9) with an entropy granularity of 0.21 bit/symbol. For a coarser entropy granularity (0.42 bit/symbol), even under a SNR estimation error up to ±1 dB, the proposed algorithm provides robust performance (≥99.6% success rate) at operation points. The results indicate that the R-EM algorithm is a promising SDI method for elastic optical networks with PS signals.

Speech enhancement through improvised conditional generative adversarial networks

Speech enhancement works towards improvising the quality of speech through various post processing algorithms. Intelligibility enhancement along with overall perceptual quality score improvement is the main objective of many speech signal processing techniques. Generative Adversarial Networks (GAN) aims to generate a new set of data with the help of training set statistics and is seen to be impressive for enhancing the speech signals in the recent years. Though GAN's does not involve prior and posterior probability calculations, they are hard to train in general. The problem aggravates with low-data regime and hence there is a need for effective GAN mechanism. In this research work, we propose to use an improvised conditional generative adversarial network where the generator will enhance the input data that is noisy while the discriminator on the other hand embedded with improvised techniques will try to differentiate between the generator output and the database clean content with the help of GAN conditions discussed. The results of the proposed method are assessed in terms of PEAQ score and equal error rate. Experimental results from the Aurora-2 signal set proves us that the improved cGAN is very effective as compared to traditional GAN networks. We have also tested the algorithm with the subjective preference and 82.14% of the subjects were found to prefer the proposed cGAN to that obtained with other conventional methods.

A parametric empirical Bayes (PEB) approach for estimating maize progress percentage at field scale

The Crop Progress Percentage (CPP) in a given phenology stage reflects growth status in the crop life cycle. Generally, routine field measurements of this variable are missing, hence various alternative approaches have been proposed for its estimation. Hidden Markov Models (HMMs) which follow the Bayesian structure are helpful tools for this aim. In the current study, an approach based on the parametric empirical Bayes (PEB) method is used for more accurate estimation of the maize CPP at field scale. The CPP information recorded in three experiment sites i.e. Karaj, Darab and Zarghan were used to validate the performance of the PEB method and to test the robustness. The PEB method includes a non-homogeneous HMM along with an empirical method based on fitting a gamma probability density function (PDF) on prior probabilities. Temporal sequence of phonological stages is regarded as the hidden layer and temporal sequence of NDVI and AGDD indices as the observable layer. The procedure of CPPs estimation is based on calculation of prior and posterior probabilities and an inverse normalization. The overall RMSE of the non-homogeneous HMM before applying the empirical method was 15.1, 11.5 and 7.8% for Karaj, Darab and Zarghan, respectively. However, it was found that the applied HMM fails to estimate the CPPs of final phenological stages accurately. The averaging process of the prior probabilities does not include the errors produced by factors such as climate variability or farming practices. To overcome these problems, we used an empirical Bayes method to estimate the hyperparameters of a gamma density function which was applied as a prior density. This simple approach to maintain the inter-annual variability due to non physiological factors, made the prior probabilities more flexible. The applied empirical Bayes approach (PEB) had significantly smaller RMSE (8, 7.5 and 6.4%, respectively); especially in final phenological stages, and led to more accurate prediction of the phenological dates. The findings derived by PEB are more consistent with those obtained by HMM when the inter-annual variability, mainly date of sowing, is minimum (Specifically, for Zarghan station). The proposed modified approach can be recommended for use at the field scale and serve as a promising tool especially in the regions which suffer the inter-annual variability.

Optimised multi‐hypothesis tracking algorithm based on the two‐dimensional constraints and manoeuvre detection

With the increase of target mobility and environment complexity, highly reliable single-period statistical judgments become more and more difficult. The multiple hypothesis tracking (MHT) algorithm is a method based on delay logic, and can effectively solve the problem of data association in the course of tracking. However, the number of hypotheses generated by the MHT algorithm is exponentially related to the false alarm rate and the number of targets. Therefore, the hypotheses reduction techniques are required for the implementation of the algorithm. The N-scan-back method and K-best method are often used. On the basis of understanding the K-Best method and the N-Scan-back method, this study proposes the two-dimensional constraints by the K-Best method and the N-Scan-back method, and besides adds the target manoeuvre detection method to jointly manage the hypotheses based on the mean of the filter residual. According to the manoeuvre detection results, the optimised MHT algorithm can adjust the likelihood probability calculation model of the plot-track association and the posterior probability calculation model of the hypothetical branch track, and shorten the decision depth N in the N-Scan-back method. Through simulation, it is proved that the optimised algorithm can reduce the calculation amount and improve the tracking accuracy.

SiNPle: Fast and Sensitive Variant Calling for Deep Sequencing Data.

Current high-throughput sequencing technologies can generate sequence data and provide information on the genetic composition of samples at very high coverage. Deep sequencing approaches enable the detection of rare variants in heterogeneous samples, such as viral quasi-species, but also have the undesired effect of amplifying sequencing errors and artefacts. Distinguishing real variants from such noise is not straightforward. Variant callers that can handle pooled samples can be in trouble at extremely high read depths, while at lower depths sensitivity is often sacrificed to specificity. In this paper, we propose SiNPle (Simplified Inference of Novel Polymorphisms from Large coveragE), a fast and effective software for variant calling. SiNPle is based on a simplified Bayesian approach to compute the posterior probability that a variant is not generated by sequencing errors or PCR artefacts. The Bayesian model takes into consideration individual base qualities as well as their distribution, the baseline error rates during both the sequencing and the PCR stage, the prior distribution of variant frequencies and their strandedness. Our approach leads to an approximate but extremely fast computation of posterior probabilities even for very high coverage data, since the expression for the posterior distribution is a simple analytical formula in terms of summary statistics for the variants appearing at each site in the genome. These statistics can be used to filter out putative SNPs and indels according to the required level of sensitivity. We tested SiNPle on several simulated and real-life viral datasets to show that it is faster and more sensitive than existing methods. The source code for SiNPle is freely available to download and compile, or as a Conda/Bioconda package.

DBM Optimization with Additional Category Information

For unconstrained training of restricted Boltzmann machine (RBM), it is easy to appear that feature homogenization leads to poor generalization ability. This paper introduces category conditions into DBM training, proposes label conditional RBM which is used in the construction of DBM. According to unsupervised training characteristics of RBM, this paper adds the category information as the model hidden unit training condition to the implicit unit posterior activation probability calculation. This paper applies the model as the underlying structure of the deep Boltzmann machine (DBM) to the deep network construction. Through handwritten digit recognition set test, compared with the shallow model, the new model after adding the category condition has a great improvement in the model training speed and feature extraction effectiveness, and can effectively enhance the feature learning ability of the deep model.

A unified neural circuit of causal inference and multisensory integration

Causal inference and multisensory integration are two fundamental processes of perception. It is generally believed that there should be one unified neural circuit in the brain to realize these two processes in an optimal way. However, there is no solution yet due to the complicated neural implementation for posterior probability computation. In this study, we propose a unified neural network by solving the complicated posterior probability computation. A unified theoretical framework is presented from the viewpoint of expectation. In addition, a biologically realistic neural circuit is proposed with the combination of importance sampling and probabilistic population coding. Theoretical analyses and simulation results manifest that our proposed neural circuit can implement both causal inference and multisensory integration. Taken together, our framework provides a new perspective of how different perceptual tasks can be performed by the same neural circuit.

Noise Robust Speaker Recognition Based on Adaptive Frame Weighting in GMM for i-Vector Extraction

Even though speaker recognition has gained significant progress in recent years, its performance is known to be deteriorated severely with the existence of strong background noises. Inspired by a recently proposed clean-frame selection approach, this work investigates a relatively elegant weighting method when computing the Baum-Welch statistics of Gaussian mixture models (GMMs) in i-vector extraction. By introducing weighting parameters to the frames of enrollment/testing utterances, the optimization problem is redefined and solved. New updating rules are derived by incorporating weights to the computation of posterior probabilities, mean vectors, and covariance matrices of the GMM. The experiments conducted on the Speakers in the Wild (SITW) database show that the proposed algorithm has significantly improved the performance of i-vector-based speaker recognition systems in noisy environments. Compared with the GMM i-vector baseline, the equal error rate is reduced from 5.75 to 4.72 and the minimum value of cost function (C det min ) is reduced from 0.4825 to 0.4505. Slight but significant superiority is also observed over the method with an additional feature enhancement frontend by using deep neural networks.

Approximate Message Passing Detector for Index Modulation With Multiple Active Resources

In this correspondence paper, we study the application of approximate message passing (AMP) algorithm in the signal detection of index modulation (IM) with multiple active resources (antennas or subcarriers). First, for uplink generalized spatial modulation (GSM), AMP detectors in both spatially uncorrelated and correlated channels are proposed. Specifically, an effective method to utilize the activation pattern constraint of GSM signal vector per user in the posterior probability computation of scalar symbol node is derived. Furthermore, a simple while effective damping strategy is introduced in AMP detector to guarantee convergence in correlated channels. Second, for orthogonal frequency division multiplexing with IM in rapidly time-varying channels, the damped AMP detector is further studied. In addition, the convergence of proposed detectors is analyzed by state evolution. Finally, the effectiveness of proposed AMP detectors is verified by simulations.

Foundry Data Analytics to Identify Critical Parameters Affecting Quality of Investment Castings

Investment castings are used in industrial sectors including automobile, aerospace, chemical, biomedical, and other critical applications; they are required to be of significant quality (free of defects and possess the desired range of mechanical properties). In practice, this is a big challenge, since there are large number of parameters related to process and alloy composition are involved in process. Also, their values change for every casting, and their effect on quality is not very well understood. It is, however, difficult to identify the most critical parameters and their specific values influencing the quality of investment castings. This is achieved in the present work by employing foundry data analytics based on Bayesian inference to compute the values of posterior probability for each input parameter. Computation of posterior probability for each parameter in turn involves computation of local probability (LP), prior odd, conditional probability (CP), joint probability (JP), prior odd, likelihood ratio (LR) as well as posterior odd. Computed value of posterior probability helps (parameters are considered to be critical if the value of posterior probability is high) in identifying critical parameter and their specific range of values affecting quality of investment castings. This is demonstrated on real-life data collected from an industrial foundry. Controlling the identified parameters within the specific range of values resulted in improved quality. Unlike computer simulation, artificial neural networks (ANNs), and statistical methods explored by earlier researchers, the proposed approach is easy to implement in industry for controlling and optimizing the parameters to achieve castings that are defect free as well as in desired range of mechanical properties. The current work also shows the way forward for building similar systems for other casting and manufacturing processes.