Harshavardhan Kamarthi, Shangqing Xu, Xinjie Tong
et al.
Hierarchical time-series forecasting is essential for demand prediction across various industries. While machine learning models have obtained significant accuracy and scalability on such forecasting tasks, the interpretability of their predictions, informed by application, is still largely unexplored. To bridge this gap, we introduce a novel interpretability method for large hierarchical probabilistic time-series forecasting, adapting generic interpretability techniques while addressing challenges associated with hierarchical structures and uncertainty. Our approach offers valuable interpretative insights in response to real-world industrial supply chain scenarios, including 1) the significance of various time-series within the hierarchy and external variables at specific time points, 2) the impact of different variables on forecast uncertainty, and 3) explanations for forecast changes in response to modifications in the training dataset. To evaluate the explainability method, we generate semi-synthetic datasets based on real-world scenarios of explaining hierarchical demands for over ten thousand products at a large chemical company. The experiments showed that our explainability method successfully explained state-of-the-art industrial forecasting methods with significantly higher explainability accuracy. Furthermore, we provide multiple real-world case studies that show the efficacy of our approach in identifying important patterns and explanations that help stakeholders better understand the forecasts. Additionally, our method facilitates the identification of key drivers behind forecasted demand, enabling more informed decision-making and strategic planning. Our approach helps build trust and confidence among users, ultimately leading to better adoption and utilization of hierarchical forecasting models in practice.
CV XYZ is a company engaged in the distribution of soybean commodities, located in Tangerang Regency. The company distributes its products to five retail outlets across the Jabodetabek area using two operational vehicles. Currently, CV XYZ faced issues with distribution scheduling due to the lack of a fixed policy, leading to occasional increases in delivery frequencies caused by insufficient stock and an inefficient existing scheduling system. As a result, the fulfillment rate of retail demands only reached 93% of the company’s target of 99%, creating a 6% gap that the company aims to address. To resolve this issue, a new distribution scheduling plan was developed using the Distribution Requirement Planning (DRP) method. Distribution Requirement Planning (DRP) method consists of four stages: first Netting for the process of calculating the amount of net requirements, second Lotting for the process of calculating the ideal order quantity, third Offsetting to determine the order plan, and finally the fourth explosion for the process of calculating gross requirements in distribution. The results of the new plan show an increase in the fulfillment rate of retail demands to 99.8%. Additionally, total distribution costs were successfully minimized, decreasing from Rp 593,980,120 to Rp 551,934,498, resulting in a saving of Rp 42,045,622 or approximately 7% of the previous total distribution costs.
Lahoucine Ech-Chatir, Salah Er-Raki, Julio Cesar Rodriguez
et al.
World food systems are subject to many challenges related to land degradation, rapid population growth, climate change, and limited resources. Crop growth models are being recognized as efficient tools for agricultural research to investigate trends in crop yield production and address these challenges under various pedoclimatic, genotypic, and management conditions. Crop growth models have come a long way in terms of development and use in recent decades but are still bound to be improved, especially for various perennial crops and the incorporation of beneficial soil microorganisms. Based on research papers published since 1965 across all continents, this review gives a brief history of crop models, explores 44 selected process-based crop growth models, their origin, usefulness, and applicability, and discusses some of their characteristics and their application in water management in arid and semi-arid areas. For the first time, this review highlights the modeling approaches in simulating the effects of beneficial soil microorganisms on crop growth, including plant growth-promoting rhizobacteria and mycorrhizal fungi, and discusses the advances in modeling perennial crops by exploring 35 studies found for fruit trees, perennial legumes, and vegetables, as well as 45 studies on perennial forage and bioenergy grasses. In addition, the review discusses crop modeling applications in the context of precision agriculture when combined with machine learning and remote sensing. The review concludes by emphasizing key limitations and challenges facing the use of crop growth models. Accordingly, this review can be a valuable resource for researchers, providing insights into existing crop models with a view to what needs to be improved.
Jackline W. Muturi, Christopher E. Ndehedehe, Mark J. Kennard
Accurate monitoring of irrigation water use (IWU) is essential for improving irrigation efficiency and countering global water stress. A key technique for monitoring IWU is satellite remote sensing, which observes the Earth ex-situ. This review synthesizes recent studies (2005–2024) to clarify the progress made and challenges in remote sensing of IWU, with a focus on global developments and specific emphasis on Australia, where vast geographic extent, regulatory complexity, and intensive irrigation pose unique challenges. Our objectives were to examine the relationships among methods, sensor types and measured variables, evaluate validation practices and identify study gaps in Australia. We found that studies have predominantly used optical sensors (82 %) to assess IWU while most other sensor types remain underutilized, particularly at regional scales, such as in Australia. Of the studies reviewed, 61 % validated their IWU estimates against field data, 24 % against reference data, 9 % against both reference and field data while 6 % did not perform validation. We highlight the need for strategic integration of remote sensing methods and sensor types to estimate IWU, supported by rigorous validation processes. Australia’s diverse agro-climatic landscape provides a valuable ground for evaluating and comparing different sensor types when used within various methods to estimate IWU. Furthermore, recent technological advances in optical, radar and microwave sensors, and future satellite missions present new opportunities in IWU monitoring. Overall, a coordinated approach within which policy makers are actively involved is essential for strengthening practical applicability of IWU estimates.
Iva Kertusha, Gebremariem Assress, Onur Duman
et al.
Web application testing is an essential practice to ensure the reliability, security, and performance of web systems in an increasingly digital world. This paper presents a systematic literature survey focusing on web testing methodologies, tools, and trends from 2014 to 2025. By analyzing 259 research papers, the survey identifies key trends, demographics, contributions, tools, challenges, and innovations in this domain. In addition, the survey analyzes the experimental setups adopted by the studies, including the number of participants involved and the outcomes of the experiments. Our results show that web testing research has been highly active, with ICST as the leading venue. Most studies focus on novel techniques, emphasizing automation in black-box testing. Selenium is the most widely used tool, while industrial adoption and human studies remain comparatively limited. The findings provide a detailed overview of trends, advancements, and challenges in web testing research, the evolution of automated testing methods, the role of artificial intelligence in test case generation, and gaps in current research. Special attention was given to the level of collaboration and engagement with the industry. A positive trend in using industrial systems is observed, though many tools lack open-source availability
Dimitrios Emmanoulopoulos, Ollie Olby, Justin Lyon
et al.
Large language models (LLMs) are increasingly deployed in agentic frameworks, in which prompts trigger complex tool-based analysis in pursuit of a goal. While these frameworks have shown promise across multiple domains including in finance, they typically lack a principled model-building step, relying instead on sentiment- or trend-based analysis. We address this gap by developing an agentic system that uses LLMs to iteratively discover stochastic differential equations for financial time series. These models generate risk metrics which inform daily trading decisions. We evaluate our system in both traditional backtests and using a market simulator, which introduces synthetic but causally plausible price paths and news events. We find that model-informed trading strategies outperform standard LLM-based agents, improving Sharpe ratios across multiple equities. Our results show that combining LLMs with agentic model discovery enhances market risk estimation and enables more profitable trading decisions.
Syed Tauhid Ullah Shah, Mohammad Hussein, Ann Barcomb
et al.
Artificial Intelligence (AI) tools for automating design artifact generation are increasingly used in Requirements Engineering (RE) to transform textual requirements into structured diagrams and models. While these AI tools, particularly those based on Natural Language Processing (NLP), promise to improve efficiency, their adoption remains limited in regulated industries where transparency and traceability are essential. In this paper, we investigate the explainability gap in AI-driven design artifact generation through semi-structured interviews with ten practitioners from safety-critical industries. We examine how current AI-based tools are integrated into workflows and the challenges arising from their lack of explainability. We also explore mitigation strategies, their impact on project outcomes, and features needed to improve usability. Our findings reveal that non-explainable AI outputs necessitate extensive manual validation, reduce stakeholder trust, struggle to handle domain-specific terminology, disrupt team collaboration, and introduce regulatory compliance risks, often negating the anticipated efficiency benefits. To address these issues, we identify key improvements, including source tracing, providing clear justifications for tool-generated decisions, supporting domain-specific adaptation, and enabling compliance validation. This study outlines a practical roadmap for improving the transparency, reliability, and applicability of AI tools in requirements engineering workflows, particularly in regulated and safety-critical environments where explainability is crucial for adoption and certification.
In this comprehensive work, David Eltis offers a two-thousand-year perspective on the trafficking of people, and boldly intervenes in the expansive discussions about slavery in the last half-century. Using new and underexplored data made available by slavevoyages.org, Eltis offers compelling explanations of why the slave trades began and why they ended, and in the process debunks long-held assumptions, including how bilateral rather than triangular voyages were the norm, and how the Portuguese rather than the British were the leading slave traders. Eltis argues that two-thirds of all enslaved people ended up in the Iberian Americas, where exports were most valuable throughout the slave trade era, and not in the Caribbean or the US. Tracing the mass involvement of people in the slave trade business from all parts of the Atlantic World, Eltis also examines the agency of Africans and their experiences in the aftermath of liberation.
Photovoltaic-powered drip irrigation is a vital approach to address the irrigation requirements in regions with limited water resources and energy deficiencies, thereby ensuring the provision of sustenance and horticultural produce for local inhabitants. However, the susceptibility of the drip irrigation system to clogging as well as the fluctuations in photovoltaic output can significantly impact irrigation quality. Moreover, conventional storage methods commonly employed in photovoltaic-powered drip irrigation systems, such as elevated water tanks and batteries, exhibit notable technological, economic, and environmental limitations. The present study introduces a novel photovoltaic drip irrigation technology (CAES-PVDI) that utilizes solar energy as the exclusive source of power, enabling stable and cost-effective high-quality drip irrigation. This technology actively regulates solar energy through compressed air energy storage, employing a cyclic pulse discharge method to ensure uniformity in irrigation outflow and significantly enhance the anti-clogging performance of the drip irrigation system. The proposed technology was implemented in a solar greenhouse for drip irrigation, and subsequent tests were conducted to assess its hydraulic performance and anti-clogging properties The results demonstrated that the system achieved a discharge uniformity of no less than 91.76 %. Furthermore, there was no blocked emitter in CAES-PVDI system, and the sedimentation inside the capillary tube decreased by 78.95 %-93.36 % compared to traditional drip irrigation system. In comparison to existing photovoltaic-powered drip irrigation technology, the CAES-PVDI system exhibited exceptional technical indicators and offered significant economic and environmental benefits, thereby presenting a novel approach to promote environmentally friendly and efficient operation of drip irrigation systems.
Seok Jin Kwon, Jung Won Seo, Seong Kwang Hong
et al.
The surface of railway wheels running on rails is subject to damage due to rail and frictional wear, damage from wheel tread and flange wear caused by curved track operations, and damage from flats and concave wear due to braking friction heat from brake shoes. Although the surface of wheels is regularly reprofiled through periodic grinding cycles, damage occurring to the wheel surface during operation can lead to deteriorated ride quality and potential failure due to crack propagation. In domestic railway components technical standards, wheel integrity is mandated to be demonstrated through non-destructive testing. To prevent and detect failures caused by damage occurring on railway wheels, it is necessary to develop methods that could detect and evaluate surface damage. The present study investigated a method for detecting and evaluating surface damage on railway wheels using electromagnetic imaging. Results demonstrated that defects with a length of 10 mm, a width of 0.8 to 1.0 mm, and a depth of 0.2 to 1.0 mm could be adequately detected using electromagnetic scan images.
Usled potrebe za smanjenjem zagađenja životne sredine, usporavanja globalnog zagrevanja i sve manje količine raspoloživih fosilnih goriva, poslednjih godina svedočimo intenzivnom razvoju proizvodnje električne energije iz obnovljivih izvora, kao i razvoju tržišta električne energije. U okviru ovog rada biće izvršena uporedna analiza učesnika na tržištu električne energije u Crnoj Gori i Republici Srbiji, kao i njihovog uticaja na distributivni elektroenergetski sistem. Posebna pažnja biće posvećena kupcima-proizvođačima. Biće prikazani modeli obračuna električne energije kupaca-proizvođača i mogućnosti za njihove dodatne uštede kako energetske, tako i finansijske. Na kraju rada biće date preporuke za buduću lakšu integraciju kupaca-proizvođača i proizvođača električne energije iz obnovljivih izvora energije u distributivni elektroenergetski sistem kroz modele agregiranja, skladištenja i upravljanja proizvodnjom i potrošnjom
Energy industries. Energy policy. Fuel trade, Economics as a science
Ilya Hajiaghayi, MohammadTaghi Hajiaghayi, Gary Peng
et al.
We study bilateral trade with a broker, where a buyer and seller interact exclusively through the broker. The broker strategically maximizes her payoff through arbitrage by trading with the buyer and seller at different prices. We study whether the presence of the broker interferes with the mechanism's gains-from-trade (GFT) achieving a constant-factor approximation to the first-best gains-from-trade (FB). We first show that the GFT achieves a $1 / 36$-approximation to the FB even if the broker runs an optimal posted-pricing mechanism under symmetric agents with monotone-hazard-rate distributions. Beyond posted-pricing mechanisms, even if the broker uses an arbitrary incentive-compatible (IC) and individually-rational (IR) mechanism that maximizes her expected profit, we prove that it induces a $1 / 2$-approximation to the first-best GFT when the buyer and seller's distributions are uniform distributions with arbitrary support. This bound is shown to be tight. We complement such results by proving that if the broker uses an arbitrary profit-maximizing IC and IR mechanism, there exists a family of problem instances under which the approximation factor to the first-best GFT becomes arbitrarily bad. We show that this phenomenon persists even if we restrict one of the buyer's or seller's distributions to have a singleton support, or even in the symmetric setting where the buyer and seller have identical distributions.
François Bachoc, Nicolò Cesa-Bianchi, Tommaso Cesari
et al.
In online bilateral trade, a platform posts prices to incoming pairs of buyers and sellers that have private valuations for a certain good. If the price is lower than the buyers' valuation and higher than the sellers' valuation, then a trade takes place. Previous work focused on the platform perspective, with the goal of setting prices maximizing the gain from trade (the sum of sellers' and buyers' utilities). Gain from trade is, however, potentially unfair to traders, as they may receive highly uneven shares of the total utility. In this work we enforce fairness by rewarding the platform with the fair gain from trade, defined as the minimum between sellers' and buyers' utilities. After showing that any no-regret learning algorithm designed to maximize the sum of the utilities may fail badly with fair gain from trade, we present our main contribution: a complete characterization of the regret regimes for fair gain from trade when, after each interaction, the platform only learns whether each trader accepted the current price. Specifically, we prove the following regret bounds: $Θ(\ln T)$ in the deterministic setting, $Ω(T)$ in the stochastic setting, and $\tildeΘ(T^{2/3})$ in the stochastic setting when sellers' and buyers' valuations are independent of each other. We conclude by providing tight regret bounds when, after each interaction, the platform is allowed to observe the true traders' valuations.
Hyeongyeong Jeong, Hyunbin Park, Jaemin Shin
et al.
This study reports an autonomous fine dust source tracking system of a water spray robot for high-rise building demolition. The core function of this system is performing a self-controlled fine dust tracking of the endpoint of the excavator, which is the fine dust generation point. The water spray robot has a lift with a parallelogram-shaped linkage to lift the water spray drum to 10 m from the ground. The sensor network system is connected to the robot and the excavator to calculate the relative position of the water spray drum and excavator endpoint using forward kinematics. RTK-GPS is attached to the robot and the excavator to calculate the relative distance. By sensor network, forward kinematics, and RTK-GPS, the water spray robot can autonomously track fine dust generation point and spray water to the endpoint of the excavator. The experiment was conducted to confirm the accuracy of kinematics calculation and tracking performance of the robot. The first experiment showed that the calculation result of forward kinematics was accurate enough to fulfill tracking operations. The second experiment showed that the tracking accuracy was precise enough, meaning that the robot could autonomously track fine dust generation point.
Elvis Dze Achuo, Honore Oumbe Tekam, Nembo Leslie Ndam
et al.
As regards the question whether natural resource affluence is a benediction or curse to sustainable development, the jury's verdict is still awaited. While we impatiently await the jury's verdict, this study provides empirical evidence that Mother Nature is responsible for Africa's predicaments with regard to economic development and environmental sustainability. Specifically, the system GMM estimates from 37 African economies reveal that: (i) natural resource affluence inhibits economic development, (ii) resource rents exacerbates carbon emissions thereby impeding environmental sustainability (iii) natural resource rents interacts with governance to produce negative synergy effects on economic growth and environmental pollution, (iv) resource rents interacts with ICT to produce respective positive net effects and negative synergy effects on economic growth and pollution emissions, (v) while non-renewable energy consumption inhibits economic growth and exacerbates pollution emissions, renewable energy consumption promotes environmental protection, (vi) we provide evidence of the U-shaped and inverted N-shaped EKC for natural resources, while also validating the inverted U-shaped EKC hypothesis relating to the nexus between per capita GDP and pollution emissions. Contingent on these findings, African countries can break the chains of the resource curse by designing sound and complementary policies upon attainment of the established thresholds by the policy modulating variables. Equally, various governments should strengthen governance quality and encourage digitalisation of the resource sector. Furthermore, African governments should propel the energy transition process by increasing investments in alternative clean energy sources in order to catalyse the attainment of the continent's Agenda 2063.
Oveis Hassanijalilian, C. Igathinathane, Stephanie Day
et al.
Iron deficiency chlorosis (IDC), a symptom of reduction in chlorophyll and stunted growth, causes a great yield loss in soybean every year in the Midwest, USA and the most efficient method to manage IDC is to plant tolerant cultivars. The assessment of cultivars' tolerance is traditionally performed by visually rating the IDC symptoms based on leaves discoloration twice during the growing season. However, the visual rating method is time-consuming, subjective, not suitable at large scales, labor-intensive, and unaffordable for frequent observation. Therefore, in this study, we used an unmanned aerial vehicle (UAV) as a tool to monitor the soybean cultivars more frequently and more efficiently through image processing approach of the whole field. Images were taken with a DJI Phantom 4 and orthomosaicked in Agisoft Photoscan. A 40-cultivar soybean experimental plots (3000 m2; Image 1) at 5 locations in North Dakota, USA (Amenia, Colfax, Leonard (2), and Hunter) for 2 years (2016 and 2017) were used in the study. The orthomosaicked images were processed in MATLAB to calculate the dark green color index (DGCI), which is a good indicator of chlorophyll in soybean leaves. The grayscale DGCI images were then processed in ArcGIS to extract the average DGCI and canopy size (CS) for each plot for each flight. The area under the curve (AUC) was calculated for DGCI, CS, and CS × DGCI product (CDP) to aggregate the values of all flights within each year. The correlation of AUC of CDP and the yield was more consistent among both years and was the better predictor of yield (R2=0.74 and R2=0.79). The latest growth stage (more representative of yield) values of both years were combined to build models for yield prediction and the CDP produced the lowest error (11.72%). Future studies should look into IDC progress measurement involving more cultivars, geographical locations, frequent imaging, as well as methods applied to regular soybean production sites to evaluate various image-based parameters and their interaction for yield predictions.
Substrate-grown crops represent approximately 10% of the cropping area of intensive greenhouse horticulture in southern Spain. The vast majority are free draining, in that they do not collect and recirculate drainage. The substantial nutrient loss in drainage contributes to contamination of water bodies. This study examined the effectiveness of dynamic management approaches to appreciably reduce the large nutrient loss associated with free-draining substrate-grown crops. For three tomato crops, grown in substrate, this study (i) compared management of N, P and K based on the ratio of the concentration in drainage to that in the nutrient solution, to conventional management, (ii) derived uptake concentration values for N, P and K throughout the crop cycle, and (iii) evaluated uptake concentration as a nutrition management tool. Ratio-based management reduced the amounts of N, P and K in drainage, in relation to conventional management by 58–61%, 65–80% and 55–77% respectively. The amounts of applied N, P and K were reduced by 22–28%, 37–43%, and 28–34% compared to conventional management. N, P and K concentrations in the applied nutrient solution slightly below the uptake concentration were associated with much lower concentrations in the drainage solution. In contrast, when the applied nutrient concentration exceeded the uptake concentration this was associated with much higher drainage nutrient concentrations. In conclusion, the two nutritional management strategies examined, ratio-based management, and use of uptake concentrations were both associated with reduced nutrient application and a considerable reduction of nutrient loss in drainage. Fruit production was maintained with the improved nutrient management practices. These strategies offer approaches that can considerably reduce the substantial nutrient loss in drainage associated with free-draining substrate cropping in greenhouse production in southern Europe.