To meet order fulfillment targets, manufacturers seek to optimize production schedules. Machine learning can support this objective by predicting throughput times on production lines given order specifications. However, this is challenging when manufacturers produce customized products because customization often leads to changes in the probability distribution of operational data—so‐called distributional shifts. Distributional shifts can harm the performance of predictive models when deployed to future customer orders with new specifications. The literature provides limited advice on how such distributional shifts can be addressed in operations management. Here, we propose a data‐driven approach based on adversarial learning, which allows us to account for distributional shifts in manufacturing settings with high degrees of product customization. We empirically validate our proposed approach using real‐world data from a job shop production that supplies large metal components to an oil platform construction yard. Across an extensive series of numerical experiments, we find that our adversarial learning approach outperforms common baselines. Overall, this paper shows how production managers can improve their decision making under distributional shifts.

Hibiscus is a widely used plant, which has been proven to have numerous positive effects on human health, such as lowering blood pressure, maintaining optimal blood cholesterol levels, liver protection, prevention of oxidative stress, etc. In this study, the content of polyphenols, flavonoids and antioxidant capacity of aqueous, ethanolic and hydroethanolic (50/50 v/v) hibiscus extracts, prepared by maceration and ultrasonic extraction, was analyzed. Analysis of antioxidant activity was performed in vitro, using FRAP and DPPH methods. The results showed that the mixture of water and ethanol had a significantly higher effect of extraction of bioactive components from hibiscus than the remaining two solvents. The lowest content of polyphenols and flavonoids, and thus the weakest antioxidant activity was recorded in extracts prepared in absolute ethanol. By comparing the efficiency of the techniques used, maceration proved to be slightly more efficient in the case of aqueous and hydroethanol extracts, while higher polyphenol content and higher antioxidant activity were observed in ethanolic extracts prepared by ultrasonic extraction.

We consider two division problems on narrow strips of square and hexagonal lattices. In both cases we compute the bivariate enumerating sequences and the corresponding generating functions, which allowed us to determine the asymptotic behavior of the total number of such subdivisions and the expected number of parts. For the square lattice we extend results of two recent references by establishing polynomiality of enumerating sequences forming columns and diagonals of the triangular enumerating sequence. In the hexagonal case, we find a number of new combinatorial interpretations of the Fibonacci numbers and find combinatorial proofs of some Fibonacci related identities. We also show how both cases could be treated via the transfer matrix method and discuss some directions for future research.

Significance Mars has a liquid iron alloy core at its center. Using seismic data gathered by the InSight mission, we have made the first observations of seismic waves traveling through Mars’ core. We use the travel times of core-transiting seismic waves, relative to ones which remain in the mantle, to constrain properties of the core and construct the first models of the elastic properties of the entire planet. Our results are consistent with a core rich in sulfur, with smaller fractions of oxygen, carbon and hydrogen.

Apple accessions, currently maintained within the two main ex situ collections in Bosnia and Herzegovina (B&H), have previously been genotyped using microsatellite markers. The obtained molecular data provided insight into mislabeled accessions and redundancies, as well as the overall genetic structure of the germplasm. The available dataset enabled the creation of a core collection consisting of 52 accessions. The reliability and usefulness of microsatellites has made this low-density marker system a norm in studies on apple germplasm. However, the increased access to medium- and high-density SNP arrays, developed specifically for apples, has opened new avenues of research into apple genetic resources. In this study, 45 apple genotypes consisting of 33 diploid core collection accessions from B&H and 12 international reference cultivars were genotyped using an Axiom® Apple 480 K SNP array in order to examine their genetic relationships, population structure and diversity, as well as to compare the obtained results with those calculated on previously reported SSR profiles. The SNPs displayed a better ability to differentiate apple accessions based on their origin, as well as to cluster them according to their pedigree. Calculating identity by descent revealed 16 pairings with first-degree relationships and uncovered the introgression of ‘Delicious’ and ‘Golden Delicious’ into the core collection.

Accurate altimetry is essential for location-based services in commercial and industrial applications. However, current altimetry methods only provide low-accuracy measurements, particularly in multistorey buildings with irregular structures, such as hollow areas found in various industrial and commercial sites. This paper innovatively proposes a tightly coupled indoor altimetry system that utilizes floor identification to improve height measurement accuracy. The system includes two optimized algorithms that improve floor identification accuracy through activity detection and address the problem of difficult convergence of z-axis coordinates due to indoor coplanarity by applying constraints to iterative least squares (ILS). Two experiments were conducted in a teaching building and a laboratory, including an irregular environment with a hollow area. The results show that our proposed method for identifying floors based on activity detection outperforms other methods. In dynamic experiments, our method effectively eliminates repeated transformations during the up- and downstairs process, and in static experiments, it minimizes the impact of barometric drift. Furthermore, our proposed altimetry method based on constrained ILS achieves significantly improved positioning accuracy compared to ILS, 1D-CNN, and WC. Specifically, in the teaching building, our method achieves improvements of 0.84 m, 0.288 m, and 0.248 m, respectively, while in the laboratory, the improvements are 2.607 m, 0.696 m, and 0.625 m.

The objective of this paper is to introduce some new logarithm operational laws for intuitionistic fuzzy sets. Some structure properties have been developed and based on these, various aggregation operators, namely confidence logarithmic intuitionistic fuzzy Einstein weighted geometric (CLIFEWG) operator, confidence logarithmic intuitionistic fuzzy Einstein ordered weighted geometric (CLIFEOWG) operator, confidence logarithmic intuitionistic fuzzy Einstein hybrid geometric (CLIFEHG) operator, confidence logarithmic intuitionistic fuzzy Einstein weighted averaging (CLIFEWA) operator, confidence logarithmic intuitionistic fuzzy Einstein ordered weighted averaging (CLIFEOWA) operator, confidence logarithmic intuitionistic fuzzy Einstein hybrid averaging (CLIFEHA) operator have been presented. To show the validity and the superiority of the proposed operators, we compared these methods with the existing methods and concluded from the comparison and sensitivity analysis our proposed techniques are more effective.

PurposeThe study aims to examine whether there is a difference in entrepreneurial intentions (EI) and intrapreneurial intentions (II) between current labor force (CLF) and future labor force (FLF). Moreover, the role of entrepreneurial orientation (EO) as a common determinant has been presented.Design/methodology/approachA cross-sectional survey was used to collect the data from 472 members of the CLF and 310 members of the FLF in Bosnia and Herzegovina. To test the hypotheses, Welch's t-test and multiple regression were used.FindingsThe findings show that there is a difference in both EI and II between CLF and FLF and that individual EO, to a certain degree, is a determinant of them.Originality/valueThe study extends the literature by offering a comparative approach to two populations within the same country. Furthermore, the study deviates from a more traditional approach that EI and II are divergent constructs with different determinants.

BACKGROUND: Atherosclerotic cardiovascular diseases (ASCVD) are still the leading cause of morbidity and mortality in Bosnia and Herzegovina. Elevated LDL-cholesterol is established as a strong marker of cardiovascular risk. Some researchers believe that measuring triglyceride levels gives a good assessment of the residual risk for ASCVD besides the measurement of LDL-cholesterol. OBJECTIVE: The aim of this study was to evaluate the overall prevalence of major risk factors for ASCVD, lipid profile and 10-year fatal cardiovascular risk using the HeartSCORE scoring system. Further, we want to evaluate the prevalence and relationship between elevated triglyceride levels and high 10-year fatal cardiovascular risk calculated as a HeartSCORE. METHODS: This is a cross-sectional study conducted on 832 volunteers aged between 40 and 65 years without a diagnosis of diabetes and without known preexisting cardiovascular disease, as a part of the preventive program conducted at the Family Medicine office. Data were collected for ASCVD risk factors and lipid panel (total cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides). 10-year fatal cardiovascular risk was calculated using the HeartSCORE scoring system for countries with high CV risk. RESULTS: Among 832 participants included, 565 (67.9%) were female, and 267 (32.1%) were male. We found high prevalence of hypertension (27.7%), obesity (32.2%), and smoking (36.2%). All lipid parameters, except HDL-C, were not optimal. Only 17.4% of participants had normal estimated HeartSCORE risk, while more than one-third (33.9%) had high or very high estimated HeartSCORE risk. Although we found a higher percentage of participants with elevated triglycerides in groups with higher HeartSCORE, there was a very weak positive correlation between values of triglycerides and the 10-year risk of a fatal cardiovascular event ( r = 0.249, p = 0.000). CONCLUSION: The high prevalence of major known risk factors and high estimated HeartSCORE risk indicate a high overall risk for ASCVD in the sample. The proportion of participants with elevated triglycerides was increased in patients with high HeartSCORE risk what implicates importance of triglyceride measurement.

Near-infrared (NIR) fluorescence imaging using exogenous fluorescent agents provides whole-field images in real-time to assist the surgeon in the excision of a tumor. Although the method has high sensitivity, the specificity can sometimes be lower than expected. Raman spectroscopy can detect tumors with high specificity. Therefore, a combination of both techniques can be advantageous. A complication that must be addressed is that the NIR spectral region is favored by both techniques for (in vivo) tissue analysis. When fluorescence and Raman emissions spectrally overlap, it becomes challenging or impossible to detect the Raman signal. In this paper, by avoiding this overlap, we describe a Raman spectroscopy setup capable of recording high-quality Raman spectra from tissue containing NIR exogenous fluorescent agents. We identify an optimal wavelength interval (900-915 nm) for Raman excitation, which avoids both excitation of fluorescent dyes and Raman signal self-absorption by the tissue. In this way, Raman spectroscopy can be combined with the currently most-used NIR fluorescent dyes. This combined novel setup could pave the way for clinical trials benefiting from both fluorescence imaging and Raman spectroscopy to avoid positive margins in cancer surgery.