Consideration Of Uncertainties Research Articles

Multiobjective optimization for sizing and placing electric vehicle charging stations considering comprehensive uncertainties

The rapid growth of electric vehicles (EVs) demands a robust and efficient charging infrastructure. To address this, we propose a particle swarm optimization algorithm designed for optimal placement and sizing of EV charging stations. This study hypothesizes that comprehensive consideration of uncertainties in vehicle types, user behaviors, road dynamics, and environmental impacts will enhance infrastructure effectiveness. Our method integrates data from road networks, driver patterns, station owners, and EV manufacturers to meet diverse charging requirements. Results indicate that 14 fast charging stations are needed along the studied freeway, with a total installation cost of $289,820 and annual operational costs of $4,223,050, leading to annual CO2 emissions of 1,843,572.57 kg. This strategic approach balances technical, environmental, and economic criteria, providing an essential tool for policymakers and urban planners in establishing sustainable EV charging networks.

Probability analysis of shallow landslides in varying vegetation zones with random soil grain-size distribution

The physically-based landslide susceptibility models are widely used to guide disaster prevention and mitigation in mountainous areas due to their significant predictive capability. However, this method faces limitations in regions with complex topography and vegetation types, primarily due to a lack of consideration for the spatial uncertainty of planted soil caused by variations in soil particle size composition. Therefore, a new model is established to predict shallow landslide occurrence considering the impact of the uncertainty of soil particle size composition on soil shear strength parameters. This model optimizes the assignment strategy for soil physical strength parameters with the support of the random soil grain-size field theory. Subsequently, it organically integrates the impact of plants on slope stability, involving root reinforcing, moisture regulation (preferential flow and root water uptake), and the canopy's interception and weight loading effects, based on the infinite slope model. The model is validated in a region with significant vegetation zonality in Sichuan Province, China. The results show: (i) the testing indicator AUC values range from 0.862 to 0.873, indicating that the model can effectively predict the spatial occurrence probability of shallow landslides, (ii) the proposed LSM-VEG-GSD model exceeds by 17.50% the traditional pseudo-static model according to the AUC score, and (iii) regardless of water height ratio interval, the probability of slope failure in different vegetation zones increases with slope angle, following an S-shaped curve regression pattern. Overall, the findings of this study contribute to predicting the stability of shallow landslides in terrain transition zones with high potential landslide concealment and uncertainty under the influence of vegetation.

Delineations for Police Patrolling on Street Network Segments with p-Median Location Models

Police patrolling intends to enhance traffic safety by mitigating the risks associated with vehicle crashes and accidents. From a view of operations, patrolling requires an effective distribution of resources and often involves area delineations for this distribution purpose. Given constraints such as budget and human resources for traffic safety, delineating geographic areas optimally for police patrol areas is an important agenda item. This paper considers two p-median location models using segments on a street network as observational units on which traffic issues such as vehicle crashes occur. It also uses two weight sets to construct an enhanced delineation of police patrol areas in the City of Plano, Texas. The first model for the standard p-median formulation gives attention to the cumulative number of motor vehicle crashes from 2011 to 2021 on the major transportation networks in Plano. The second model, an extension of this first p-median one, uses balancing constraints to achieve balanced spatial coverage across patrol areas. These two models are also solved with network kernel density count estimates (NKDCE) instead of crash counts. These smoothed densities on a network enable consideration of uncertainty affiliated with this aggregation. The analysis results of this paper suggest that the p-median models provide effective specifications, including their capability to define patrol areas that encompass the entire study region while minimizing distance costs. The inclusion of balancing constraints ensures a more equitable distribution of workloads among patrol areas, improving overall efficiency. Additionally, the model with NKDCE results in an improved workload balance among delineated areas for police patrolling activities, thus supporting more informed spatial decision-making processes for public safety.

Evaluating manufacturing plants readiness for distributed ledger technology: a network framework for ambiguity levels

The utilisation of blockchain technology has gained significant traction within contemporary supply chains owing to its ability to enhance transparency, security, and traceability. Manufacturing plants, as pivotal components of the supply chain, stand to benefit from improved tracking and transparency of goods movement, real-time visibility, quality control processes, and adherence to industry standards through blockchain implementation. Nonetheless, without a comprehensive assessment of manufacturing plants’ readiness to embrace blockchain technology, the anticipated benefits may give way to unforeseen challenges. In this study, a novel network framework is offered to evaluate manufacturing plants’ readiness for adopting distributed ledger technology, specifically blockchain, under varying levels of ambiguity, including high (fuzzy) and low (scenario) ambiguity. This framework is distinguished by its ability to address uncertainty in evaluations, incorporating both scenario-based and fuzzy programming approaches. Furthermore, the framework treats evaluation criteria as interconnected entities, fostering a network perspective rather than a black-box approach. The proposed framework is then validated through a case study involving five manufacturing plants and twenty-four evaluation criteria. Our findings underscore the pivotal role of uncertainty considerations in ranking manufacturing plants, with the fifth plant emerging as the frontrunner across both fuzzy and scenario-based assessments in most instances.

An Efficient LC–HRMS-Based Approach to Evaluate Pesticide Contamination in Water Bodies with Measurement Uncertainty Considerations

Over recent decades, the global occurrence of pesticide residues in aquatic environments has been a pivotal issue; however, their trace-level concentrations necessitate the establishment of ultra-sensitive and reliable analytical approaches. To this end, the present study describes the optimization and validation of an LC-HRMS-based method for the accurate determination of 18 pesticides in river and sea water, accompanied by a measurement uncertainty estimation. This method was applied to analyze 17 real samples from agriculture and aquaculture-impacted areas in Greece and Albania. Different solid-phase extraction (SPE) protocols were tested. For the analysis, cutting-edge Orbitrap MS technology and MS/MS fragmentation, along with the use of matrix-matched calibration curves, provided unprecedented accuracy (<5 ppm) and sensitivity for the confirmation of positive detections. Regarding method performance, exceptional linearity was obtained; the limits of quantification ranged from 1.7 ng L−1 to 90 ng L−1, recoveries varied from 61% to 96% in river water, while slightly higher recoveries (60–111%) were observed in seawater. In all cases, repeatability and intra-laboratory reproducibility were below 15%. The measurement expanded uncertainty (U′, k = 2) was estimated considering precision and bias. MU% values were lower than 50% in all cases, as recommended in SANTE guidelines and applied to the quantified results. The matrix effect study exhibited negative values (<20%) for all compounds. Application to real samples showed a low pesticide contamination load that should not be underestimated.

Benchmark dose modeling for epidemiological dose-response assessment using case-control studies.

Following a previous article that focused on integrating epidemiological data from prospective cohort studies into toxicological risk assessment, this paper shifts the focus to case-control studies. Specifically, it utilizes the odds ratio (OR) as the main epidemiological measure, aligning it with the benchmark dose (BMD) methodology as the standard dose-response modeling approach to determine chemical toxicity values for regulatory risk assessment. A standardized BMD analysis framework has been established for toxicological data, including input data requirements, dose-response models, definitions of benchmark response, and consideration of model uncertainty. This framework has been enhanced by recent methods capable of handling both cohort and case-control studies using summary data that have been adjusted for confounders. The present study aims to investigate and compare the "effective count" based BMD modeling approach, merged with an algorithm used for converting odds ratio to relative risk in cohort studies with partial data information (i.e., the Wang algorithm), with the adjusted OR-based BMD analysis approach. The goal is to develop an adequate BMD modeling framework that can be generalized for analyzing published case-control study data. As in the previous study, these methods were applied to a database examining the association between bladder and lung cancer and inorganic arsenic exposure. The results indicate that estimated BMDs and BMDLs are relatively consistent across both methods. However, modeling adjusted OR values as continuous data for BMD estimation aligns better with established practices in toxicological BMD analysis, making it a more generalizable approach.

Panel-based assessment of ecosystem condition as a platform for adaptive and knowledge driven management.

Ecosystems are subjected to increasing exposure to multiple anthropogenic drivers. This has led to the development of national and international accounting systems describing the condition of ecosystems, often based on few, highly aggregated indicators. Such accounting systems would benefit from a stronger theoretical and empirical underpinning of ecosystem dynamics. Operational tools for ecosystem management require understanding of natural ecosystem dynamics, consideration of uncertainty at all levels, means for quantifying driver-response relationships behind observed and anticipated future trajectories of change, and an efficient and transparent synthesis to inform knowledge-driven decision processes. There is hence a gap between highly aggregated indicator-based accounting tools and the need for explicit understanding and assessment of the links between multiple drivers and ecosystem condition as a foundation for informed and adaptive ecosystem management. We describe here an approach termed PAEC (Panel-based Assessment of Ecosystem Condition) for combining quantitative and qualitative elements of evidence and uncertainties into an integrated assessment of ecosystem condition at spatial scales relevant to management and monitoring. The PAEC protocol is founded on explicit predictions, termed phenomena, of how components of ecosystem structure and functions are changing as a result of acting drivers. The protocol tests these predictions with observations and combines these tests to assess the change in the condition of the ecosystem as a whole. PAEC includes explicit, quantitative or qualitative, assessments of uncertainty at different levels and integrates these in the final assessment. As proofs-of-concept we summarize the application of the PAEC protocol to a marine and a terrestrial ecosystem in Norway.

Feasibility of coaxial deep borehole heat exchangers in southern California

This paper investigates the feasibility of coaxial deep borehole heat exchanger (CDBHE) applications to the University of California San Diego (UCSD) campus. By collecting different geophysical source data for various formations and well logs around the UCSD campus, a multilayered thermophysical model for the ground on the site is established. Water circulation within a closed coaxial loop system considers the geothermal energy extraction under uncertainty consideration of the unknown deeper layers heat flow gradient as coupled with the variation of pipe insulation properties, flow rates, outer pipe diameter, grout, and depths between 1 and 4 km. A finite-element framework models the Navier–Stokes fluid flow and heat transfer in the CDBHE system, validated with a field test on CDBHE from the literature. Results show that a 4-km CDBHE could produce a thermal power of 600 kW under the optimum geological conditions at the UCSD site: the water flow rate of 2.78 L/s and a ground thermal gradient of 60 ℃/km. Thermal power shares from different layers indicate that deeper formation layers contribute more to the thermal power than the shallower layers because increasing the CDBHE length from 1 to 4 km can lead to a maximum of 900% increase in thermal power and a 50% expansion in thermal plume for a CDBHE with an insulated inner pipe between the upper and lower bound heat flow bounds. An inner pipe with an insulated depth of 2 km produces only 1–6% less power than a fully insulated inner pipe for the 4-km CDBHE, and thus, a partially insulated vacuum-insulated tube (VIT)-plastic inner pipe is suggested as the best practice. Furthermore, the CDBHE thermal power increases by 5% when the grout thermal conductivity increases from 1 to 3.65 W/(K∙m), close to the formation thermal conductivity, and then maintains almost the same, and the 4-km CDBHE with flow rates of 2.78–6.94 L/s at the UCSD site can directly supply a low-temperature heating radiator system for room heating. This study suggests practical ranges for geothermal energy extraction for southern California. A CDBHE with a well-insulated inner pipe of 0.05 W/(m∙K), the thermal power of lower and upper-bound heat flow cases can vary by 60% from the mean. Finally, water as the working fluid is more efficient than CO2, doubling CDBHE's thermal power. The effects of the investigated factors provide guidelines for future geothermal resource exploitation in southern California.

Adaptive multi-model predictive control with optimal model bank formation: Consideration of local models uncertainty and stability

The Multiple Model Control (MMC) structure comprises three main components: the model bank, controller bank, and supervisor algorithm. Precise design of these components is crucial for achieving high control performance within the MMC framework, albeit this effort is not without its challenges. These challenges involve optimizing the model and controller banks ensuring system stability when dealing with uncertainties in the local models and enabling smooth switching between model-controller pairs. This paper addresses these challenges by presenting a comprehensive approach. Firstly, the optimal model bank is designed using an automatic clustering approach. Then, the design of the adaptive multi-model predictive controller bank and a supervisor algorithm capable of performing soft switching are discussed. The proposed control system exhibits the capability to ensure closed-loop system stability within each individual subspace, as well as during the transitioning between distinct subspaces. This stability is preserved even in the face of inherent uncertainties associated with the local models comprising the model bank. To evaluate and validate the performance of the proposed control system, it is applied to a satellite attitude control system. The results confirm the effectiveness and performance of the control system. The proposed control system holds promise for controlling highly nonlinear, complex, or switched systems, ensuring closed-loop system stability, and achieving high control performance.

Uncertainty-Aware Time Series Anomaly Detection

Traditional anomaly detection methods in time series data often struggle with inherent uncertainties like noise and missing values. Indeed, current approaches mostly focus on quantifying epistemic uncertainty and ignore data-dependent uncertainty. However, consideration of noise in data is important as it may have the potential to lead to more robust detection of anomalies and a better capability of distinguishing between real anomalies and anomalous patterns provoked by noise. In this paper, we propose LSTMAE-UQ (Long Short-Term Memory Autoencoder with Aleatoric and Epistemic Uncertainty Quantification), a novel approach that incorporates both aleatoric (data noise) and epistemic (model uncertainty) uncertainties for more robust anomaly detection. The model combines the strengths of LSTM networks for capturing complex time series relationships and autoencoders for unsupervised anomaly detection and quantifies uncertainties based on the Bayesian posterior approximation method Monte Carlo (MC) Dropout, enabling a deeper understanding of noise recognition. Our experimental results across different real-world datasets show that consideration of uncertainty effectively increases the robustness to noise and point outliers, making predictions more reliable for longer periodic sequential data.

Net-zero carbon emission oriented Bi-level optimal capacity planning of integrated energy system considering carbon capture and hydrogen facilities

The integrated energy system (IES) is considered an efficient energy provision paradigm to improve the utilization efficiency of renewable generation (RG), reduce energy cost and improve energy supply reliability. However, the multi-dimensional uncertainties and the hydrogen utilization have posed enormous challenges to the optimal planning of the IES. This paper proposes a net-zero carbon emission oriented bi-level optimal planning model for the IES with full consideration of the multiple operational uncertainties in the IES (RG, multiple demands). The upper level is an optimal capacity configuration model, and the economic costs, energy selling profits, the penalty for power abandonment and energy supply shortage are settled as the objectives. The lower level is an optimal energy dispatch model with the objective function of minimizing the total operating cost of the IES on typical days. An integrated energy system framework consisting of multiple energy flows is formulated in the presence of carbon capture system (CCS) and hydrogen-to-power (H2P) facilities. To fully characterize the uncertainties of RG and multiple demands, a data-driven scenario generation method is adopted to expand the operational scenarios, in which probability distribution assumptions and prior knowledge are not necessary. Finally, the effectiveness of the proposed solution is extensively demonstrated through numerical simulation, and a sensitivity analysis is carried out to assess the impact of changes in electricity and gas prices on the IES planning results.

Method for prediction magnetic silencing of uncertain energy-saturated extended technical objects in prolate spheroidal coordinate system

Aim. Development of method for prediction by energy-saturated extended technical objects magnetic silencing based on magnetostatics geometric inverse problems solution and magnetic field spatial spheroidal harmonics calculated in prolate spheroidal coordinate system taking into account of technical objects magnetic characteristics uncertainties. Methodology. Spatial prolate spheroidal harmonics of extended technical objects magnetic field model calculated as magnetostatics geometric inverse problems solution in the form of nonlinear minimax optimization problem based on near field measurements for prediction far extended technical objects magnetic field magnitude. Nonlinear objective function calculated as the weighted sum of squared residuals between the measured and predicted magnetic field COMSOL Multiphysics software package used. Nonlinear minimax optimization problems solutions calculated based on particle swarm nonlinear optimization algorithms. Results. Results of prediction extended technical objects far magnetic field magnitude based on extended technical objects spatial prolate spheroidal harmonics of the magnetic field model in the prolate spheroidal coordinate system using near field measurements with consideration of extended technical objects magnetic characteristics uncertainty. Originality. The method for prediction by extended technical objects magnetic cleanliness based on spatial prolate spheroidal harmonics of the magnetic field model in the prolate spheroidal coordinate system with consideration of magnetic characteristics uncertainty is developed. Practical value. The important practical problem of prediction extended technical objects magnetic silencing based on the spatial prolate spheroidal harmonics of the magnetic field model in the prolate spheroidal coordinate system with consideration of extended technical objects magnetic characteristics uncertainty solved. References 48, figures 2.

A distributionally robust optimization approach of multi-park integrated energy systems considering shared energy storage and Uncertainty of Demand Response

To enhance the economic efficiency and renewable energy integration capacity of multi-park integrated energy systems (MPIES) and address the issue of insufficient consideration of demand response uncertainty in existing studies, this paper proposes a distributionally robust optimization approach for multi-park integrated energy systems, considering shared energy storage and the uncertainty of demand response. First, models of the shared energy system and demand response are established. Based on these models, a deterministic optimization scheduling model for MPIES is developed, aiming to minimize system costs while considering constraints such as grid power balance. To address uncertainty in demand response, this paper employs Interval Type-2 Fuzzy Theory to construct uncertainty sets and considers system reliability constraints. The original optimization problem is then transformed into an equivalent robust counterpart model, and the optimal distributionally robust solution is obtained through parameter domain decomposition methods. Finally, the proposed method is validated using the IEEE 33-bus system. The results show that considering shared energy storage and demand response individually can reduce total system costs by 4.86 % and 26.46 %, respectively. After accounting for the uncertainty in demand response, the total system cost increases only slightly by 4.36 %, but this improves the system's robustness.

Feasibility of Photonuclear Transmutation of Long-Lived Fission Products Extracted from Used Nuclear Fuel

To achieve the goal of net-zero carbon emission in energy production, nuclear power capacity and waste generation are expected to expand significantly in the next few decades. In the condition of continuous fuel recycling, long-lived fission products (LLFPs) are dominant contributors to the disposal impacts of nuclear waste. In this study, six LLFPs, including 99Tc, 129I, 135Cs, 126Sn, 93Zr, and 79Se, were identified as the primary contributors to more than 99% of long-term radiotoxicity of disposed nuclear waste across a wide range of fuel cycle scenarios. To reduce the amounts of LLFPs sent to geological repositories, the nuclear transmutation of LLFPs is being pursued. Specifically, this work systematically assessed the feasibility of transmuting LLFPs via photonuclear reactions. Photon transmutation is physically viable for the identified primary LLFPs except for 99Tc. For the five transmutable LLFPs, the achievable photon transmutation performance without isotopic separation was evaluated based on scoping calculations and consideration of nuclear data uncertainties. Using an extremely intense laser Compton photon source of 1019 γ /s, the effective transmutation half-life can be reduced to a few years. However, the absolute transmutation rates of LLFPs remain below 1 kg/yr. The energy required to power the photon source for transmuting all LLFPs produced in a nuclear reactor exceeds the net energy output of the reactor. Several potential strategies for improving photon transmutation performance were analyzed. None can substantially enhance the performance to make it practical for industrial applications.

Efficient Approximation of Varying Fiber Orientation States in Injection Molded Parts Under Consideration of Multiple Manufacturing Uncertainties

AbstractThe production of high-quality fiber reinforced polymer parts is an important aspect in several industrial areas. However, due to unavoidable uncertainties in material and manufacturing processes, the part quality scatters. One important aspect here is the fiber orientation, being crucial for the thermo-mechanical properties of the part and being influenced by the uncertain material state and process conditions. Process simulations are an important tool for predicting the fiber orientation, but state-of-the-art simulations are normally deterministic and represent only one specific case. Performing enough deterministic simulations to model manufacturing uncertainties requires high numerical effort. Therefore, this work presents methods to quickly and efficiently approximate the fiber orientation under varying material and process parameters, requiring only a few simulations as input. Different schemes for approximation are evaluated and compared with each other and with 3D process simulations.

Mathematical modelling for pandemic preparedness in Canada: Learning from COVID-19.

The COVID-19 pandemic underlined the need for pandemic planning but also brought into focus the use of mathematical modelling to support public health decisions. The types of models needed (compartment, agent-based, importation) are described. Best practices regarding biological realism (including the need for multidisciplinary expert advisors to modellers), model complexity, consideration of uncertainty and communications to decision-makers and the public are outlined. A narrative review was developed from the experiences of COVID-19 by members of the Public Health Agency of Canada External Modelling Network for Infectious Diseases (PHAC EMN-ID), a national community of practice on mathematical modelling of infectious diseases for public health. Modelling can best support pandemic preparedness in two ways: 1) by modelling to support decisions on resource needs for likely future pandemics by estimating numbers of infections, hospitalized cases and cases needing intensive care, associated with epidemics of "hypothetical-yet-plausible" pandemic pathogens in Canada; and 2) by having ready-to-go modelling methods that can be readily adapted to the features of an emerging pandemic pathogen and used for long-range forecasting of the epidemic in Canada, as well as to explore scenarios to support public health decisions on the use of interventions. There is a need for modelling expertise within public health organizations in Canada, linked to modellers in academia in a community of practice, within which relationships built outside of times of crisis can be applied to enhance modelling during public health emergencies. Key challenges to modelling for pandemic preparedness include the availability of linked public health, hospital and genomic data in Canada.

Gravimetric and SIM-Headspace GC-MS for Residual Organic Solvents Detection in Halal and Wholesomeness Food Analysis

The demand for halal foods and beverages is increasing globally. While most halal analysis focuses on porcine, this study focuses on assessing residual organic solvents to ensure their halal compliance and wholesomeness, following several Malaysian standards and guidelines. A significant challenge in this study was the volatility of the residual solvents during the preparation of standards and quality control. To address this issue, a gravimetric technique was employed and effectively minimized the difference between theoretical (1,000 ppm) and actual (710 – 892 ppm) concentrations of the residual organic standard stock solution, except for acetone (588 ppm). The aim of this study was to establish a validated, reliable, and accurate method using SIM-headspace GC-MS to identify and quantify residual organic solvents for halal and wholesomeness analysis. Confirmation of each residual organic solvent was achieved by comparing the obtained spectra with the NIST 11 spectral database, containing 70,832 compounds, with similarity ranging from 80.9% to 96.6%, except for acetonitrile at 52.2%. The validation parameters were carried out according to ISO 17025:2017, the Center for Drug Evaluation and Research, and the European Guidelines. The parameters included recovery ranging from 95.65% to 95.68%, precision from 10.08% to 19.65% RSD, linearity between 0.996 to 0.999, limit of detection from 0.01 to 0.08 ppm, and limit of quantification from 0.02 to 0.24 ppm. Uncertainty considerations were limited to recovery, precision, and linearity, as other uncertainties were negligible based on the bottom-up approach using in-house validation data. This combination of gravimetric and SIM-headspace GC-MS techniques has provided valuable insights for discussions and collaborations among halal authorities worldwide to establish a consensus analytical methodology for halal and wholesomeness assessment.

A probabilistic approach for the levelized cost of energy of floating offshore wind farms

This paper aims to analyze the levelized cost of energy (LCOE) of floating offshore wind farms from a probabilistic point of view. Understanding and addressing the uncertainty associated with the main parameters that influence the wind energy generated by a wind farm during its lifetime is crucial for the economic evaluation of offshore wind energy in the broader energy landscape. The methodology for probabilistic assessment of LCOE is introduced, and the uncertainty in input parameters are discussed. In a base case study, an assumed Floating Offshore Wind Farm (FOWT) consisting of 250 5-MW wind turbines is considered. The use of bias and randomness in key random variables is discussed and studied in detail. Results indicate that LCOE estimates of 15 EURc/kWh for offshore wind turbines are achievable with reasonable confidence, while estimates of 5 EURc/kWh require careful consideration of uncertainty in the wind farm’s parameters. The feasibility analysis showed that techno-economic parameters are more influenced by wind characteristics and efficient use of wind turbines than by the cost of the wind farm. This paper provides general guidance on how to carry out early-stage techno-economic analysis of FOWFs.

Psychosis Prognosis Predictor: A continuous and uncertainty-aware prediction of treatment outcome in first-episode psychosis.

Machine learning models have shown promising potential in individual-level outcome prediction for patients with psychosis, but also have several limitations. To address some of these limitations, we present a model that predicts multiple outcomes, based on longitudinal patient data, while integrating prediction uncertainty to facilitate more reliable clinical decision-making. We devised a recurrent neural network architecture incorporating long short-term memory (LSTM) units to facilitate outcome prediction by leveraging multimodal baseline variables and clinical data collected at multiple time points. To account for model uncertainty, we employed a novel fuzzy logic approach to integrate the level of uncertainty into individual predictions. We predicted antipsychotic treatment outcomes in 446 first-episode psychosis patients in the OPTiMiSE study, for six different clinical scenarios. The treatment outcome measures assessed at both week 4 and week 10 encompassed symptomatic remission, clinical global remission, and functional remission. Using only baseline predictors to predict different outcomes at week 4, leave-one-site-out validation AUC ranged from 0.62 to 0.66; performance improved when clinical data from week 1 was added (AUC = 0.66-0.71). For outcome at week 10, using only baseline variables, the models achieved AUC = 0.56-0.64; using data from more time points (weeks 1, 4, and 6) improved the performance to AUC = 0.72-0.74. After incorporating prediction uncertainties and stratifying the model decisions based on model confidence, we could achieve accuracies above 0.8 for ~50% of patients in five out of the six clinical scenarios. We constructed prediction models utilizing a recurrent neural network architecture tailored to clinical scenarios derived from a time series dataset. One crucial aspect we incorporated was the consideration of uncertainty in individual predictions, which enhances the reliability of decision-making based on the model's output. We provided evidence showcasing the significance of leveraging time series data for achieving more accurate treatment outcome prediction in the field of psychiatry.

Integrating heterogenous robustness levels to the multi-objective target-oriented robust optimization of a microalgal biorefinery

Integrated biorefineries offer an efficient way to produce biofuels by enhancing profit through the generation of marketable by-products and by reducing environmental impact by reutilizing waste streams originating from different processes within the system. Employing a variety of conversion technologies, integrated biorefineries yield a wide range of products and byproducts, encompassing biofuels, biochemicals, and bioenergy, all derived from biomass feedstocks. Several mathematical methods have been developed in order to optimize the design of biorefineries with consideration for uncertainty, including the multi-objective target-oriented robust optimization (MOTORO) approach that enables target-seeking behavior while maximizing the tolerable deviation of parameters in the system. However, the phenomenon of heterogenous robustness levels that tradeoff risks with each other has not been investigated in both MOTORO and in wider robust optimization literature. The present study addresses this gap by applying multiple, potentially non-equal degrees of robustness to each customer's demand, allowing stakeholders to place unequal risks on each customer. As a result, limited resources may be more efficiently utilized in line with existing knowledge or preference regarding uncertainties. The proposed improvement was applied to an algal biorefinery system with two customers, where robustness successfully exhibited tradeoffs with each other while retaining inverse relationships with system underperformance and variation as evaluated from a Monte Carlo simulation.