This chapter presents the concept of a Smart Rice Mill as an intelligent, connected, automated, and traceable rice-processing ecosystem. It integrates IoT sensors, computer vision, deep learning, Edge AI, cloud analytics, predictive maintenance, intelligent control, and blockchain to improve rice-processing operations. The chapter discusses automated grain inspection, variety classification, defect detection, broken-rice estimation, milling-quality prediction, machine monitoring, process optimization, and digital recording of batch history. It also examines implementation challenges involving legacy machinery, hardware and sensor reliability, cybersecurity, staff training, integration, and economic feasibility. The proposed future direction is a closed-loop Smart Rice Mill capable of sensing paddy and machine conditions, predicting quality, adjusting processing parameters, verifying output, and maintaining complete traceability.
Khoya (khoa or mawa) is a traditional dairy product, prepared by heating and concentrating milk, which is widely used in preparation of indigenous milk sweets. But, challenges such as process variability, quality deterioration, microbial contamination, adulteration, limited shelf life and inefficient supply chain management hinder its production and distribution. New solutions to these challenges are available across the khoya value chain due to recent advancements in artificial intelligence (AI) and Industry 4.0 technologies. This review highlights the applications of AI in khoya processing, packaging, transportation, distribution and quality management. The role of machine learning, deep learning, computer vision, Internet of Things (IoT), digital twins, smart sensors, and blockchain in process optimization, automated quality inspection, adulteration detection, shelf-life prediction, intelligent packaging, cold-chain monitoring, logistics optimization and demand forecasting is explored. We also review AI-enabled analytical tools for rapid and non-destructive quality assessment, such as hyperspectral imaging, electronic nose, and electronic tongue. The review also discusses the contribution of AI to improving food safety, traceability, sustainability and operational efficiency, as well as to reducing post-harvest losses and environmental impacts. Finally, the paper discusses the existing challenges, future research directions, and prospects of AI-enabled smart dairy manufacturing. The review finds that AI can play a significant role in improving the quality, safety, efficiency, and sustainability of the khoya industry and helping its transition to intelligent and data-driven dairy processing.
Given the high volatility and susceptibility to extreme events in the cryptocurrency market, forecasting tail risk is of paramount importance. Value-at-Risk (VaR), a quantile-based risk measure, is widely used for assessing tail risk and is central to monitoring financial market stability. In data-rich environments, functional data from various domains are employed to forecast conditional quantiles. However, the infinite-dimensional nature of functional data introduces uncertainty. This paper addresses this uncertainty problem by proposing a novel data-driven conditional quantile model averaging (MA) approach. With a set of candidate models varying by the number of components, MA assigns weights to each model determined by a K-fold cross-validation criterion. We prove the asymptotic optimality of the selected weights in terms of minimizing the excess final prediction error when all candidate models are misspecified. Additionally, when the true regression relationship belongs to the set of candidate models, we provide consistency results for the averaged estimators. Numerical studies indicate that, in most cases, the proposed method outperforms other model selection and averaging methods, particularly for extreme quantiles in cryptocurrency markets.
Nima Sedaghat, Brianna Smart, J. Bryce Kalmbach, Erin L. Howard · 5 authors
We report a study exploring how the use of deep neural networks with astronomical Big Data may help us find and uncover new insights into underlying phenomena: through our experiments towards unsupervised knowledge extraction from astronomical Big Data we serendipitously found that deep convolutional autoencoders tend to reject telluric lines in stellar spectra. With further experiments we found that only when the spectra are in the barycentric frame does the network automatically identify the statistical independence between two components, stellar vs telluric, and rejects the latter. We exploit this finding and turn it into a proof-of-concept method for removal of the telluric lines from stellar spectra in a fully unsupervised fashion: we increase the inter-observation entropy of telluric absorption lines by imposing a random, virtual radial velocity to the observed spectrum. This technique results in a non-standard form of ``whitening'' in the atmospheric components of the spectrum, decorrelating them across multiple observations. We process more than 250,000 spectra from the High Accuracy Radial velocity Planetary Search (HARPS) and with qualitative and quantitative evaluations against a database of known telluric lines, show that most of the telluric lines are successfully rejected. Our approach, `Stellar Karaoke', has zero need for prior knowledge about parameters such as observation time, location, or the distribution of atmospheric molecules and processes each spectrum in milliseconds. We also train and test on Sloan Digital Sky Survey (SDSS) and see a significant performance drop due to the low resolution. We discuss directions for developing tools on top of the introduced method in the future.
This study evaluated hospital demographics, staffing, pharmacy variables, health care outcomes measures (severity of illness-adjusted mortality rates, drug costs, total cost of care, and length of stay) and medication errors. A database was constructed from the 1992 American Hospital Association's Abridged Guide to the Health Care Field, the 1992 National Clinical Pharmacy Services database, and 1992 mortality data from the Health Care Financing Administration. Simple statistical tests and a severity of illness-adjusted multiple regression analysis were employed. The study population consisted of 1116 hospitals that reported information on medication errors and 913 hospitals that reported information on medication errors that adversely affected patient care outcomes. We evaluated factors associated with the 430,586 medication errors and 17,338 medication errors that adversely affected patient care outcomes. Medication errors occurred in 5.07% of the patients admitted each year to these hospitals. Each hospital experienced a medication error every 22.7 hours (every 19.73 admissions). Medication errors that adversely affected patient care outcomes occurred in 0.25% of all patients admitted to these hospitals/year. Each hospital experienced a medication error that adversely affected patient care outcomes every 19.23 days (or every 401 admissions). The following factors were associated with increased medication errors/occupied bed/year: lack of pharmacy teaching affiliation (slope = 0.8875, p=0.0416), centralized pharmacists (slope = 1.0942, p=0.0001), number of registered nurses/occupied bed (slope = 1.624, p=0.032), number of registered pharmacists/occupied bed (slope = 25.0573, p=0.0001), hospital mortality rate (slope = 2.8017, p=0.0192), and total cost of care/occupied bed/year (slope = 0.01432, p=0.0091). Factors associated with decreased medication errors were location in the Mid-Atlantic census region (slope = -1.5182, p=0.03), affiliation with a pharmacy teaching program (slope = -1.0252, p=0.0349), decentralized pharmacists (slope = -0.9843, p=0.0037), and number of medical residents/occupied bed (slope = -1.478, p=0.0014). There was a 45% decrease in medication errors (1.81-fold decrease) in hospitals that had decentralized pharmacists, compared with hospitals that had centralized pharmacists. In addition, there was a 94% decrease in medication errors that adversely affected patient care outcomes (16.88-fold decrease) in hospitals that had decentralized pharmacists compared with hospitals that had only centralized pharmacists. Based on previous field studies and our findings in 1116 hospitals, it appears that one of the most effective ways to prevent or reduce medication errors is to decentralize pharmacists to patient care areas. The results of this study should help hospitals reduce the number of medication errors that occur each year.