Using AI and ML to predict shipment times of therapeutics, diagnostics and vaccines in e-pharmacy supply chains during COVID-19 pandemic

PurposeThis paper aims to address the pressing problem of prediction concerning shipment times of therapeutics, diagnostics and vaccines during the ongoing COVID-19 pandemic using a novel artificial intelligence (AI) and machine learning (ML) approach.Design/methodology/approachThe present study used organic real-world therapeutic supplies data of over 3 million shipments collected during the COVID-19 pandemic through a large real-world e-pharmacy. The researchers built various ML multiclass classification models, namely, random forest (RF), extra trees (XRT), decision tree (DT), multilayer perceptron (MLP), XGBoost (XGB), CatBoost (CB), linear stochastic gradient descent (SGD) and the linear Naïve Bayes (NB) and trained them on striped datasets of (source, destination, shipper) triplets. The study stacked the base models and built stacked meta-models. Subsequently, the researchers built a model zoo with a combination of the base models and stacked meta-models trained on these striped datasets. The study used 10-fold cross-validation (CV) for performance evaluation.FindingsThe findings reveal that the turn-around-time provided by therapeutic supply logistics providers is only 62.91% accurate when compared to reality. In contrast, the solution provided in this study is up to 93.5% accurate compared to reality, resulting in up to 48.62% improvement, with a clear trend of more historic data and better performance growing each week.Research limitations/implicationsThe implication of the study has shown the efficacy of ML model zoo with a combination of base models and stacked meta-models trained on striped datasets of (source, destination and shipper) triplets for predicting the shipment times of therapeutics, diagnostics and vaccines in the e-pharmacy supply chain.Originality/valueThe novelty of the study is on the real-world e-pharmacy supply chain under post-COVID-19 lockdown conditions and has come up with a novel ML ensemble stacking based model zoo to make predictions on the shipment times of therapeutics. Through this work, it is assumed that there will be greater adoption of AI and ML techniques in shipment time prediction of therapeutics in the logistics industry in the pandemic situations.

Hydrochemical Characterization of Nabeghlavi Underground Mineral Waters

Georgia is rich in groundwater deposits, which renew over time and are characterized by the best indicators of water quality and a stable regime. Groundwater is one of the main natural productive forces of Georgia, which plays an important role in the economic development and export industry. During 2015–2020, chemical composition of the Nabeghlavi mineral waters has been studied by the means of the modern unified methods. Almost all data from previous chemical analyzes have also been retrieved and systematized/collated. Using mathematical statistical analysis, the maximum, minimum, and mean arithmetic values of the major and specific components of water and the empirical deviation from the arithmetic mean have been calculated. Though comparison of the recent and historic data on chemical composition, based on the results of statistical analysis of the major ions, the natural fluctuation limits of the waters and stability of waters from all exploitation drill holes have been shown. According to the chemical composition and total mineralization there are three groups of waters in fresh and low mineralized category. It is shown that chemical composition of water from all exploitation drill holes meets requirements for the natural mineral water category, both the normative document of Georgia and directive of the European Commission.

Trend Analysis of Temperature and Rainfall of Rajasthan, India

Increasing temperature and declining and erratic rainfall is one of the greatest global challenges. This study presents the trend analysis of temperature and rainfall in five divisional headquarters of Rajasthan, namely, Bikaner, Jaipur, Jodhpur, Kota, and Udaipur. The historic data of minimum and maximum temperature and rainfall for a period of 49 years from 1971 to 2019 were collected from Climate Research and Services, India Meteorological Department, Pune. Detection of trends and change in magnitude was done using the Mann–Kendall (MK) test and Sen’s slope, respectively. The results of the study indicated a significant increase in both minimum and maximum temperature over time for all the five stations. However, rainfall showed a nonsignificant increasing trend for Kota and Udaipur district, whereas Bikaner, Jaipur, and Jodhpur detected a negative trend.

An Overview of Hydrogeological Works at Nabeghlavi Mineral Water Deposit and Replenishment of the Natural Resources of the Deposit by Atmospheric Precipitation

Mineral waters are very important resources of the country. Their moderate consumption and protection is possible through sustainable and efficient management. The paper presents an overview of hydrogeological works carried out on the Nabeghlavi carbonic mineral water deposit. Information on calculation of the natural resource replenishment coefficient with atmospheric precipitation based on the hydrogeological and meteorological data and characterization of geological-technical condition of exploitation drill holes is described. Coefficient of replenishment of the Nabeghlavi mineral water resources with infiltrated atmospheric precipitation has been determined by Darcy’s method. According to the reference historic data, back in 1986, the magnitude of the coefficient was 25,307 m3/ day, whereas in 2000-2020 it has equated 28,125 m3/day. Conventional average of these values made 26,715 m3/day, which is 90 times higher than the current approved operating reserve (296 m3/day) of the deposit.

Reviewing the UK’s Action Levels for the Management of Dredged Material

Action Levels (ALs) are thresholds which are used to determine whether dredged material is suitable for disposal at sea by providing a proxy risk assessment for potential impacts to biological features such as fish and benthos. This project tested proposed scenarios for changes to the UK Action Levels to determine the likely implications for navigational dredge licensing in England and Wales. Approximately 3000 sample data records from 2009 to 2018 were collated with varying numbers of concentrations for contaminant parameters including trace metals, organotins, polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), organochlorine pesticides (OCPs) and polybrominated diphenyl ethers (PBDEs). Initially, these data were assessed using current ALs to determine the percentages of the samples with levels below AL1 (generally acceptable for disposal), between AL1 and AL2 and those showing levels above AL2 (generally unacceptable for disposal). These results were then used to compare with the results of the proposed new AL scenarios for each contaminant type derived from literature reviews and historic data. The results indicate that there are changes to the ALs which could be made such as updating the current ALs with the revised ALs, as well as the introduction of ALs where there are currently none set. The benefits of changing the ALs include reducing contaminant disposal to the marine environment and increased transparency in decision making. Any proposed scenarios will need to be phased in carefully in full liaison with stakeholders.

An update on the physiologic changes during pregnancy and their impact on drug pharmacokinetics and pharmacogenomics

For many years, the medical community has relied in clinical practice on historic data about the physiological changes that occur during pregnancy. However, some newer studies have disputed a number of assumptions in these data for not being evidence-based or derived from large prospective cohort-studies. Accurate knowledge of these physiological changes is important for three reasons: Firstly, it facilitates correct diagnosis of diseases during pregnancy; secondly, it enables us to answer questions about the effects of medication during pregnancy and the ways in which pregnancy alters pharmacokinetic and drug-effects; and thirdly, it allows for proper modeling of physiologically-based pharmacokinetic models, which are increasingly used to predict gestation-specific changes and drug–drug interactions, as well as develop new knowledge on the mode-of-action of drugs, the mechanisms underlying their interactions, and any adverse effects following drug exposure. This paper reviews new evidence regarding the physiologic changes during pregnancy in relation to existing knowledge.

Accessible Modeling of the German Energy Transition: An Open, Compact, and Validated Model

We present an easily accessible model for dispatch and expansion planning of the German multi-modal energy system from today until 2050. The model can be used with low efforts while comparing favorably with historic data and other studies of future developments. More specifically, the model is based on a linear programming partial equilibrium framework and uses a compact set of technologies to ease the comprehension for new modelers. It contains all equations and parameters needed, with the data sources and model assumptions documented in detail. All code and data are openly accessible and usable. The model can reproduce today’s energy mix and its CO2 emissions with deviations below 10%. The generated energy transition path, for an 80% CO2 reduction scenario until 2050, is consistent with leading studies on this topic. Our work thus summarizes the key insights of previous works and can serve as a validated and ready-to-use platform for other modelers to examine additional hypotheses.

Combining Morphological and Histogram based Text Line Segmentation in the OCR Context

Text line segmentation is one of the pre-stages of modern optical character recognition systems. The algorithmic approach proposed by this paper has been designed for this exact purpose. Its main characteristic is the combination of two different techniques, morphological image operations and horizontal histogram projections. The method was developed to be applied on a historic data collection that commonly features quality issues, such as degraded paper, blurred text, or presence of noise. For that reason, the segmenter in question could be of particular interest for cultural institutions, that want access to robust line bounding boxes for a given historic document. Because of the promising segmentation results that are joined by low computational cost, the algorithm was incorporated into the OCR pipeline of the National Library of Luxembourg, in the context of the initiative of reprocessing their historic newspaper collection. The general contribution of this paper is to outline the approach and to evaluate the gains in terms of accuracy and speed, comparing it to the segmentation algorithm bundled with the used open source OCR software.

Representing and Validating Cultural Heritage Knowledge Graphs in CIDOC-CRM Ontology

In order to unify access to multiple heterogeneous sources of cultural heritage data, many datasets were mapped to the CIDOC-CRM ontology. CIDOC-CRM provides a formal structure and definitions for most cultural heritage concepts and their relationships. The COURAGE project includes historic data concerning people, organizations, cultural heritage collections, and collection items covering the period between 1950 and 1990. Therefore, CIDOC-CRM seemed the optimal choice for describing COURAGE entities, improving knowledge sharing, and facilitating the COURAGE dataset unification with other datasets. This paper introduces the results of translating the COURAGE dataset to CIDOC-CRM semantically. This mapping was implemented automatically according to predefined mapping rules. Several SPARQL queries were applied to validate the migration process manually. In addition, multiple SHACL shapes were conducted to validate the data and mapping models.

Azooxanthellate Scleractinia (Cnidaria, Anthozoa) from South Africa

Globally, South Africa ranks in the top five countries regarding marine species richness per unit area. Given the high diversity, it is not surprising that many invertebrate taxa in the region are poorly characterised. The South African azooxanthellate Scleractinia (Anthozoa) is one such taxonomic group, and was last reviewed by Boshoff in 1980. Although more recent regional publications have reported on some species, there has not been a faunistic review that accounts for the country’s species diversity since then. Moreover, numerous unidentified specimens representing more than three decades of sampling effort have accumulated. In this study the authors update the state of knowledge of South African azooxanthellate coral species. Specimens, particularly those within the extensive collections of the Iziko South African and Smithsonian museums, were morphologically examined and identified. Other data considered included historic data represented as imagery data, associated species data from recent research surveys, and the scientific literature. To date, the study has increased the total number of known species from 77 to 108 across eleven families, 28 new South African records, and three are new species with one new genus.