Pareto Surface Research Articles

Safety-driven design of carbon capture utilization and storage (CCUS) supply chains: A multi-objective optimization approach

Carbon capture, utilization, and storage supply chains (CCUS) play a pivotal role in achieving sustainability targets but necessitate meticulous risk identification and mitigation measures. Traditional safety assessments often occur post-design, constraining proactive risk management efforts. Hence, there is a pressing need to optimize safety performance during the design stages. To address this challenge, a framework for evaluating and optimizing CCUS supply chain safety performance using inherent safety index system (ISI) is introduced. Recognizing the trade-offs between total cost, environmental impact reduction, and risk mitigation, our approach considers multi-objective optimization to concurrently address these sustainability objectives and generate a Pareto set of solutions. Utilizing the augmented ε-constraint method, we applied this framework to optimize CCUS networks and develop sustainable designs across three key objectives. The method was applied to a CCUS system that includes various CO2 utilization pathways to minimize the total annual cost, CO2 emissions, and safety risks. The resulting Pareto surface illustrates unique network configurations, each representing a distinct trade-off scenario. Through a case study, we optimized a CCUS network to achieve economic, environmental, and safety objectives. The most economically viable design, with a total annual cost of $97 million and a 40 % net carbon reduction, prioritizes CO2 utilization for value-added products, while limiting CO2 sequestration. Conversely, safety-focused designs shift utilization towards safer routes, including CO2 sequestration and algae production. The proposed framework offers a systematic approach to developing sustainable CCUS supply chain designs, balancing economic viability, environmental sustainability, and safety.

Multi-choice conic goal programming model-based network data envelopment analysis

In multi-stage processes, classical Data Envelopment Analysis (DEA) acts like a black box and measures the efficiency of decision-making units (DMUs) without considering the internal structure of the system. Unlike classical DEA, recent studies have shown that the overall system efficiency scores are more meaningful if researched using the Network DEA (NDEA) methodology. NDEA performs simultaneous efficiency evaluations of sub-processes and the entire system. Recently, the composition method integrated with multi-objective programming (MOP) has been preferred by many authors to alleviate the drawbacks of earlier methods such as decomposition, slack-based measure (SBM) and the system-centric approach. This study proposes a novel approach incorporating Multi-Choice Conic Goal Programming into the NDEA (MCCGP-NDEA). It provides a more accurate representation of the Pareto front by revealing potential Pareto optimal solutions which are overlooked by the composition methods. Due to its ability to modify stage weights based on the decision makers’ (DMs) preferences, it is likely to gather more samples from the Pareto surface. Computational results on available benchmark problems confirm that the proposed MCCGP-NDEA is a viable alternative to existing methods.

Cardinality-constrained plan-quality and delivery-time optimization method for proton therapy.

While minimizing plan delivery time is beneficial for proton therapy in terms of motion management, patient comfort, and treatment throughput, it often poses a tradeoff with optimizing plan quality. A key component of plan delivery time is the energy switching time, which is approximately proportional to the number of energy layers, that is, the cardinality. This work aims to develop a novel optimization method that can efficiently compute the pareto surface between plan quality and energy layer cardinality, for the planner to navigate through this quality-and-efficiency tradeoff and select the appropriate plan of a balanced tradeoff. A new IMPT method CARD is proposed that (1) explicitly incorporates the minimization of energy layer cardinality as an optimization objective, and (2) automatically generates a set of plans sequentially with a descending order in number of energy layers. The energy layer cardinality is penalized through the l1,0-norm regularization with an upper bound, and the upper bound is monotonically decreased to compute a series of treatment plans with gradually decreased energy layer cardinality on the quality-and-efficiency pareto surface. For any given treatment plan, the plan optimality is enforced using dose-volume planning objectives and the plan deliverability is imposed through minimum-monitor-unit (MMU) constraints, with optimization solution algorithm based on iterative convex relaxation. The new method CARD was validated in comparison with the benchmark plan of all energy layers (P0), and a state-of-the-art method called MMSEL, using prostate, head-and-neck (HN), lung, pancreas, liver and brain cases. While labor-intensive and time-consuming manual parameter tuning was needed for MMSEL to generate plans of predefined energy layer cardinality, CARD automatically and efficiently computed all plans with sequentially decreasing predefined energy layer cardinality all at once. With the acceptable plan quality (i.e., no more than 110% of total optimization objective value from P0), CARD achieved the reduction of number of energy layers to 52% (from 77 to 40), 48% (from 135 to 65), 59% (from 85 to 50), 67% (from 52 to 35), 80% (from 50 to 40), and 30% (from 66 to 20), for prostate, HN, lung, pancreas, liver, and brain cases, respectively, compared to P0, with overall better plan quality than MMSEL. Moreover, due to the nonconvexity of the MMU constraint, CARD provided the similar or even smaller optimization objective than P0, at the same time with fewer number of energy layers, that is, 55 versus 77, 85 versus 135, 45 versus 52, and 25 versus 66 for prostate, HN, pancreas, and brain cases, respectively. We have developed a novel optimization algorithm CARD that can efficiently and automatically compute a series of treatment plans of any given energy layer sequentially, which allows the planner to navigate through the plan-quality and energy-layer-cardinality tradeoff and select the appropriate plan of a balanced tradeoff.

Control co-design optimization of floating offshore wind turbines with tuned liquid multi-column dampers

Abstract. The technical progress in the development and industrialization of floating offshore wind turbines (FOWTs) over the past decade has been significant. Yet, the higher levelized cost of energy (LCOE) of FOWTs compared to onshore wind turbines is still limiting the market share. One of the reasons for this is the larger motions and loads caused by the rough environmental excitations. Many prototype projects tend to employ more conservative substructure designs to meet the requirements for motion dynamics and structural safety. Another challenge lies in the multidisciplinary nature of a FOWT system, which consists of several strongly coupled subsystems. If these subsystems cannot work in synergy, the overall system performance may not be optimized. Previous research has shown that a well-designed blade pitch controller is able to reduce the motions and structural loads of FOWTs. Nevertheless, due to the negative aerodynamic damping effect, improvement in the performance by tuning the controller is limited. One of the solutions is adding tuned liquid multi-column dampers (TLMCDs), meaning that there is a structural solution to mitigate this limiting factor for the controller performance. It has been found that the additional damping, provided by TLMCDs, is able to improve the platform pitch stability, which allows a larger blade pitch controller bandwidth and thus a better dynamic response. However, if a TLMCD is not designed with the whole FOWT system dynamics taken into account, it may even deteriorate the overall performance. Essentially, an integrated optimization of these subsystems is needed. For this paper, we develop a control co-design optimization framework for FOWTs installed with TLMCDs. Using the multi-objective optimizer non-dominated sorting genetic algorithm II (NSGA-II), the objective is to optimize the platform, the blade pitch controller, and the TLMCD simultaneously. Five free variables characterizing these subsystems are selected, and the objective function includes the FOWT's volume of displaced water (displacement) and several motion and load indicators. Instead of searching for a unique optimal design, an optimal Pareto surface of the defined objectives is determined. It has been found that the optimization is able to improve the dynamic performance of the FOWT, which is quantified by motions and loads, when the displacement remains similar. On the other hand, if motions and loads are constant, the displacement of the FOWT can be reduced, which is an important indication of lower manufacturing, transportation, and installation costs. In conclusion, this work demonstrates the potential of advanced technologies such as TLMCDs to advance FOWTs for commercial competitiveness.

Simultaneous reduction of number of spots and energy layers in intensity modulated proton therapy for rapid spot scanning delivery.

Reducing proton treatment time improves patient comfort and decreases the risk of error from intrafractional motion, but must be balanced against clinical goals and treatment planquality. To improve the delivery efficiency of spot scanning proton therapy by simultaneously reducing the number of spots and energy layers using the reweighted regularizationmethod. We formulated the proton treatment planning problem as a convex optimization problem with a cost function consisting of a dosimetric plan quality term plus a weighted regularization term. We iteratively solved this problem and adaptively updated the regularization weights to promote the sparsity of both the spots and energy layers. The proposed algorithm was tested on four head-and-neck cancer patients, and its performance, in terms of reducing the number of spots and energy layers, was compared with existing standard and group regularization methods. We also compared the effectiveness of the three methods ( , group , and reweighted ) at improving plan delivery efficiency without compromising dosimetric plan quality by constructing each of their Pareto surfaces charting the trade-off between plan delivery and planquality. The reweighted regularization method reduced the number of spots and energy layers by an average over all patients of and , respectively, with an insignificant cost to dosimetric plan quality. From the Pareto surfaces, it is clear that reweighted provided a better trade-off between plan delivery efficiency and dosimetric plan quality than standard or group regularization, requiring the lowest cost to quality to achieve any given level of deliveryefficiency. Reweighted regularization is a powerful method for simultaneously promoting the sparsity of spots and energy layers at a small cost to dosimetric plan quality. This sparsity reduces the time required for spot scanning and energy layer switching, thereby improving the delivery efficiency of protonplans.

Response surface methodology and artificial intelligence modeling for in vitro regeneration of Brazilian micro sword (Lilaeopsis brasiliensis)

In this study, response surface methodology (RSM) was used to optimize in vitro regeneration of the Brazilian micro sword (Lilaeopsis brasiliensis) aquatic plant, followed by data prediction and validation using machine learning algorithms. The basal salt, sucrose and Benzyaminopurine (BAP) concentrations were derived from Box-Behnken design of RSM. The response surface regression analysis revealed that 1.0 g/L MS + 0.1 mg/L BAP + 25 g/L sucrose was optimized for maximum regeneration (100%), shoot counts (63.2), and fresh weight (1.382 g). The RSM-based predicted scores were fairly similar to the actual scores, which were 100% regeneration, 63.39 shoot counts, and 1.44 g fresh weight. Pareto charts analysis illustrated the significance of MS for regeneration and fresh weight but remained insignificant. Conversely, MS × BAP was found to be the most crucial factor for the shoot counts, with MS coming in second and having a major influence. The analysis of the normal plot ascertained the negative impact of elevated MS concentration on shoot counts and enhanced shoot counts from the combination of MS × BAP. Results were further optimized by constructing contour and surface plots. The response optimizer tool demonstrated that maximum shoot counts of 63.26 and 1.454 g fresh weight can be taken from the combination of 1.0 g/L MS + 0.114 mg/L BAP + 23.94 g/L. Using three distinct performance criterias, the results of machine learning models showed that the multilayer perceptron (MLP) model performed better than the random forest (RF) model. Our findings suggest that the results may be utilized to optimize various input variables using RSM and verified via ML models.Key messageOptimization of in vitro whole plant regeneration of Brazilian sword wood using response surface methodologyData analysis through ANOVA, response surface regression anlaysis and machine learningGraphical presentation of data via Pareto charts, normal plots, contour plots and surface plots for optimizationBetter performance of ANN-based MLP model compared to decision tree based RF modelGraphical abstract

Synergizing LED Technology and Hydropriming for Intelligent Modeling and Mathematical Expressions to Optimize Chickpea Germination and Growth Indices

The influence of hydropriming and Light Emitting Diodes (LED) on germination and growth indices, followed by optimizing and validation via artificial intelligence-based models was carried out in this research. White LEDs (W-LEDs) were more effective by yielding the most effective growth indices, such as mean germination time (MGT) (1.11 day), coefficient of variation of germination time (CVt) (20.72%), mean germination rate (MR) (0.81 day−1), uncertainty (U) (0.40 bit), and synchronization (Z values) (0.79); the optimum MGT (1.09 day), CVt (15.97%), MR (0.77 day−1), U (0.32 bit), and Z (0.55) values were found after 2 h of hydropriming, which was responsible for all efficient growth indicators. W-LEDs with 1 h hydropriming proved to be the ideal LED and hydropriming combination. Results on growth indices for in vitro seedlings were completely different from those on germination indices, and the most desirable germination indices were linked to red LEDs (R-LEDs). Whereas 4 h hydropriming was most effective for the post-germination process. Pareto charts, normal plots, contour plots, and surface plots were created to optimize the input variables. Finally, the data were predicted using Arificial Neural Network (ANN) inspired multilayer perceptron (MLP) and machine learning-based random forest (RF) algorithms. For both models, plant height was correlated with maximum R2 values. Whereas, all output variables had relatively low mean absolute error (MAE), mean square error (MSE), and mean absolute percentage error (MAPE) scores, indicating that both models performed well. The results of this investigation disclosed a link between certain LEDs and hydropriming treatment for in vitro germination indices and plant growth.Graphical Graphical presentation of actual and predicted values for germination indices in chickpea

Interpretable self-organizing map assisted interactive multi-criteria decision-making following Pareto-Race

The problem-solving task of the multi-criteria decision-making (MCDM) approach involves decision makers’ (DMs’) interaction by incorporating their preferences to arrive at one or more preferred near Pareto-optimal solution(s). The Pareto-Race is an interactive MCDM approach that relies on finding preferred solutions as the DM navigates on the Pareto surface by selecting the preferred reference direction and uniform/varying speed using the concept of Achievement Scalarization Function (ASF) or its augmented version (AASF). Selecting a reference direction or speed is possible, algorithmically, but the effectiveness of the Pareto-Race concept relies on suitable visualization methods that present the series of past solutions and convey the current objective trade-off information to assist the DM’s in decision-making task. Traditional multi-dimensional Pareto-optimal front visualization methods like parallel coordinate plots, radial visualization, and heat maps are deemed insufficient for this purpose. Therefore, the paper proposes integrating the concept of the Pareto-Race MCDM approach with a recently developed interpretable self-organizing map (iSOM) based visualization method. The iSOM maps high-dimensional data into lower-dimensional space, facilitating visual convenience for the DM. The proposed method requires a finite-size representation of non-dominated solutions near the complete Pareto Front to generate iSOM plots of objectives. We also propose visualizing the metrics such as closeness to constraint boundaries, trade-off value, and robustness using iSOM to check the quality of the preferred solutions, which is rarely considered in existing MCDM approaches. iSOM plots of objective functions, closeness to constraint, trade-off, and robustness metrics assist the DM in effectively choosing new reference direction and step size in each iteration to arrive at the most preferred solution. The proposed iSOM-enabled Pareto-Race approach is demonstrated on benchmark analytical and real-world engineering examples with 3 to 10 objectives.

Multi-Criteria Optimization and RapidPlan for Improved Organs at Risk Sparing in Treatment Planning of Malignant Pleural Mesothelioma

Treatment planning for malignant pleural mesothelioma is complex and time-consuming owing to the large volume of the target as well as its overlap and proximity to critical organs. Knowledge-based planning (KBP) model using RapidPlan (RP) was previously developed and clinically tested at our institution for treatment planning of these cases that had undergone pleurectomy-decortication, and therefore had two intact-lungs. The aim of this work is to investigate if multi-criteria optimization (MCO) can further improve the RP model with respect to organ at risk (OAR) sparing without compromising target coverage. The RP model was trained with clinically accepted plans of 57 patients that used Volumetric Modulated Arc Therapy (VMAT) with 2 partial arcs and 6 MV photons. The dose volume histogram (DVH) estimation model was trained to estimate doses to the heart, ipsilateral lung, total lung, contralateral lung, stomach, esophagus, kidneys and liver. Clinical treatment plans for 12 patients were re-planned using RP as well as a combination of RP and MCO. Application of MCO was after RP and used tradeoff exploration to navigate to the improved dose distribution on a Pareto surface. Selection of a plan that further improved OAR sparing while maintaining coverage constraints of PTV D95 and V95 ≥ 94% was made. Dosimetric parameters for clinical plans (CP), plans using RP as well as plans generated with RP and MCO were all compared for 12 new validation cases. Wilcoxon sign-rank test was used for statistical significance testing. The mean heart dose was reduced from 20.9 Gy ± 3.2 Gy for CP to 17.1 Gy ± 4.2 Gy with RP and further to 13.8 Gy ± 3.3 Gy with RP and MCO. The heart V30 Gy was reduced from 23.7% ± 8% to 18.2% ± 9.4% with RP and further to 13.6% ± 5.7% with RP and MCO. The corresponding results for contralateral lung V5 Gy were 68.6% ± 14.1% with CP, 50% ± 26.3% with RP and 45.3% ± 22.8% with RP and MCO, while the mean esophagus dose was reduced from 26.5 Gy ± 3.1 Gy to 21.5 Gy ± 4.7 Gy with RP and further to 19.1 Gy ± 4.2 Gy with RP and MCO. All these dosimetric improvements were statistically significant (p<0.001). However, improvements with RP and MCO for the total lung mean and V20 Gy, liver mean, stomach mean and kidney V18 Gy were marginal over the RP. Sparing of ipsilateral lung V20 Gy was maintained at ≥ 50 cc on average for all plans. PTV D95 and V95 were both normalized at 94%. Combination of RP and MCO significantly improved sparing of the OARs, especially the heart, contralateral lung and esophagus without compromising coverage or doses to other structures. Since doses to the heart, and contralateral lung are known to correlate with incidence of radiation pneumonitis, it would be prudent to consider planning with both RP and MCO to help determine the optimal treatment plan for the individual patient's anatomy.

A heterogeneous ensemble architecture coupling model selection sorting and residual error iterative correction for crude oil price forecasting

Although crude oil price forecasts can provide policymakers, investors, consumers, and producers with scientific and reliable guidance, the nonlinearity and complexity of crude oil markets still pose serious challenges to this work. To this end, we propose a crude oil price ensemble prediction model that combines model selection sorting, multi-objective heterogeneous ensemble learning, and residual error iterative correction. First, a selection strategy based on the weighted information criterion is used to screen benchmark models in the model pool. Subsequently, candidate models are heterogeneously integrated using the multi-objective multi-verse optimization algorithm. VIseKriterijumska Optimizacija I Kompromisno Resenje, Technique for Order Preference by Similarity to Ideal Solution, and Grey Relational Analysis are utilized to select the optimal non-dominated solution of the Pareto surface. Finally, correction technology is designed to adaptively correct the residual error. Optimum error shrinkage and the number of corrections ensure maximum predictable component development. West Texas Intermediate and Europe Brent spot price data are utilized for experimental verification. Results show that the model can accurately describe crude oil price fluctuations, and has excellent robustness and stability. The forecasting method can be used as a reliable and promising energy economic decision algorithm.

Multi Objective Nonlinear Model Predictive Control of Diabetes Models Considering the Effects of Insulin and Exercise

Rigorous multiobjective nonlinear model predictive control on the diabetes model incorporating single and multiple control strategies. The amount of glucose is minimized with the Bergman model considering the effects of insulin and exercise. The optimization language pyomo is used in conjunction with the state-of-the-art global optimization solvers IPOPT and Baron. Pareto surfaces are generated. When some optimal control profiles were found to exhibit sharp spikes, an activation factor involving the hyperbolic tangent function was used. It is observed that a greater amount of glucose minimization is achieved when more control procedures were incorporated. This demonstrates that it is more beneficial to use multiple control strategies

A CFD multi-objective optimization framework to design a wall-type heat recovery and ventilation unit with phase change material

This work applies multi-objective optimization based on a genetic algorithm to design a wall-type heat recovery and ventilation (HRV) unit equipped with a phase change material (PCM) to show the benefits of using this thermal energy storage system in buildings. The initial design – taken from a reference work – consists of an airflow driven by a fan through a copper tube bundle filled with PCM. Starting from this, a thermo-fluid dynamic 3D model is developed to assess the natural convection negligibility inside the PCM. Then, a 2D model is solved with a commercial finite element code and validated. Once the validation is assessed, this model is coupled with MATLAB® to perform the multi-objective optimization procedure. Optimal system configurations (tube diameter, thickness, pitches) and operational variables (fan velocity, cycle time, PCM melt temperature) are achieved on a 3D Pareto surface that collects the non-dominated solutions, i.e., minimizing costs and pressure drop (air side), and maximizing the average storage rate. The 3D Pareto approach enables to find solutions that outperform the initial configuration, i.e., the average storage rate is increased by 28.4%, simultaneously improving the other two objective functions. Finally, two sub-optimal solutions from the 2D fronts are numerically compared to show the difference in the air stream distribution across the new tubes’ arrangement and the PCM melting evolution. The HRV efficiency is enhanced from 37.5% to 44.4%, confirming that the proposed approach provides an economically-feasible solution, while reducing energy consumption and maintaining good indoor air quality.

Urban growth boundaries optimization under low-carbon development: Combining multi-objective programming and patch cellular automata models

Urban Growth Boundaries (UGBs) are a tool to control urban sprawl. However, the way to optimize future urban land uses and fix their boundaries is not clear. This paper presents a new framework to delimit UGBs while accounting for ecological, economic, and carbon storage benefits. Aggregate land-use constraints are included in a multi-objective optimization algorithm to capture non-inferior solutions on the Pareto Surface (PS) under different objective scenarios. A patch-level cellular automata simulation model is then used to spatially allocate these land uses, followed by a new two-step adjustment method to delineate the UGBs. This modeling is applied to Wuhan, China. The results show that: (1) One district (Caidian) will have a strong economic growth under low-carbon development. (2) The maximization of carbon storage reduces losses in ecological benefits, suggesting that carbon storage be considered in urban growth planning. (3) The combined model framework and two-step boundary adjustment method can help urban planners define different UGB scenarios and make science-based policy decisions.

Multi-objective Bayesian optimization for the design of nacre-inspired composites: optimizing and understanding biomimetics through AI.

The hierarchical structures found in biological materials lead to an outstanding balance of multiple material properties, and numerous research studies have been initiated to emulate the key concepts for the designing of engineering materials, the so-called bioinspired composites. However, the optimization of bioinspired composites has long been difficult as it usually falls into the category of 'black-box problem', the objective functions not being available in a functional form. Also, bioinspired composites possess multiple material properties that are in a trade-off relationship, making it impossible to reach a unique optimal design solution. As a breakthrough, we propose a data-driven material design framework which can generate bioinspired composite designs with an optimal balance of material properties. In this study, a nacre-inspired composite is chosen as the subject of study and the optimization framework is applied to determine the designs that have an optimal balance of strength, toughness, and specific volume. Gaussian process regression was adopted for the modeling of a complex input-output relationship, and the model was trained with the data generated from the crack phase-field simulation. Then, multi-objective Bayesian optimization was carried out to determine pareto-optimal composite designs. As a result, the proposed data-driven algorithm could generate a 3D pareto surface of optimal composite design solutions, from which a user can choose a design that suits his/her requirement. To validate the result, several pareto-optimal designs are built using a PolyJet 3D printer, and their tensile test results show that each of the characteristic designs is well optimized for its specific target objective.

Multi-objective optimization of a lithium cell design operating in a cyclic process

Multi-objective optimization frameworks of lithium-ion whole-cell design operating in a cyclic process are presented in this paper. The objective functions involved are maximization of capacity, maximization of specific energy in discharge and maximization of cycle energy efficiency. The considered design variables are electrode thicknesses, solid volume fraction and porosity of electrode compartments, and cell mass per unit cross-sectional area. In addition, the constant current value of charge and discharge is optimized. A phenomenological mathematical model developed in GAMS (General Algebraic Modeling System) environment is applied to achieve the stated goals and the multi-objective problems are solved using an epsilon constraint method-based, which is effective for finding solutions in non-convex feasible regions. Solutions for two different cell chemistries are analyzed: LMO and LCO. Pareto curves and surfaces are obtained and sets of solutions with the highest contradictions and with the most balanced agreement between the objectives involved are identified. When comparing optimal LMO designs with optimal LCO designs, the former result on average 48 % less capacity, 19 % less specific energy and 31 % more energy efficiency. Results show that material type-dependent variables such as active solids have a strong influence on the specific energy and cycle energy efficiency. While the electrode thicknesses show a general trend regardless of the type of chemistry, presenting a strong influence on cell capacity values.

Multi-objective optimization of equation of state molecular parameters: SAFT-VR Mie models for water

The determination of a suitable set of molecular interaction parameters for use with an equation of state (EoS) can be viewed as a multi-objective optimization (MOO) problem, where each objective quantifies the quality of the description for a particular type of thermodynamic property. We outline a methodology for the determination of a set of Pareto-optimal interaction parameters. The Pareto front is generated efficiently using a sandwich algorithm where one solves a sequence of weighted-sum scalarized single objective optimization problems. The algorithm presented can be used for any number of objective functions, allowing for the consideration of multiple thermodynamic property types as competing objectives in the MOO. The methodology is applied to the determination of suitable parameter sets for models of water within the SAFT-VR Mie framework. Three competing property targets are considered as objective functions: saturated-liquid density, vapour pressure and isobaric heat capacity. Two different types of molecular models are considered: spherical models of water, and non-spherical model of water. We analyse the two- and three-dimensional Pareto surfaces and parameter sets obtained for different property combinations in the MOO. The proposed methodology can be used to provide a rigorous comparison between different model types. Numerous Pareto-optimal parameter sets for SAFT-VR Mie water models are documented, and we recommend two new models (one spherical model and one non-spherical model) with an appropriate compromise between the competing objectives.

Multi-objective optimisation of polymerase chain reaction continuous flow systems

A surrogate-enabled multi-objective optimisation methodology for a continuous flow Polymerase Chain Reaction (CFPCR) systems is presented, which enables the effect of the applied PCR protocol and the channel width in the extension zone on four practical objectives of interest, to be explored. High fidelity, conjugate heat transfer (CHT) simulations are combined with Machine Learning to create accurate surrogate models of DNA amplification efficiency, total residence time, total substrate volume and pressure drop throughout the design space for a practical CFPCR device with sigmoid-shape microfluidic channels. A series of single objective optimisations are carried out which demonstrate that DNA concentration, pressure drop, total residence time and total substrate volume within a single unitcell can be improved by up to sim5.7%, sim80.5%, sim17.8% and sim43.2% respectively, for the practical cases considered. The methodology is then extended to a multi-objective problem, where a scientifically-rigorous procedure is needed to allow designers to strike appropriate compromises between the competing objectives. A series of multi-objective optimisation results are presented in the form of a Pareto surface, which show for example how manufacturing and operating cost reductions from device miniaturisation and reduced power consumption can be achieved with minimal impact on DNA amplification efficiency. DNA amplification has been found to be strongly related to the residence time in the extension zone, but not related to the residence times in denaturation and annealing zones.

Multi-objective optimization tool for PATs operation in water pressurized systems

ABSTRACT The use of pump working as turbine (PAT) instead of the traditional pressure regulation systems could allow for a recovery of the excess hydraulic energy to reduce the energy footprint of the water supply industry and at the same time control the water losses by an effective reduction in pressure induced by the turbine head drop. This research aims to explore the option of applying multiple recovery systems in a water network with an integrated multi-objective optimization using genetic algorithms. The objective of the optimization is to ensure a better use and effectiveness in the implementation of these solutions. A methodology to approach this multi-objective solution and the interface between components of the optimization is developed and presented. The evolutionary capacities of the optimization is analysed and the effects of the general convergence of the Pareto surface front with the adaptation of the final solutions to the available PATs.

Probabilistic Pareto plan generation for semiautomated multicriteria radiation therapy treatment planning

Objective. We propose a semiautomatic pipeline for radiation therapy treatment planning, combining ideas from machine learning–automated planning and multicriteria optimization (MCO). Approach. Using knowledge extracted from historically delivered plans, prediction models for spatial dose and dose statistics are trained and furthermore systematically modified to simulate changes in tradeoff priorities, creating a set of differently biased predictions. Based on the predictions, an MCO problem is subsequently constructed using previously developed dose mimicking functions, designed in such a way that its Pareto surface spans the range of clinically acceptable yet realistically achievable plans as exactly as possible. The result is an algorithm outputting a set of Pareto optimal plans, either fluence-based or machine parameter–based, which the user can navigate between in real time to make adjustments before a final deliverable plan is created. Main results. Numerical experiments performed on a dataset of prostate cancer patients show that one may often navigate to a better plan than one produced by a single-plan-output algorithm. Significance. We demonstrate the potential of merging MCO and a data-driven workflow to automate labor-intensive parts of the treatment planning process while maintaining a certain extent of manual control for the user.

Multi-Objective Optimization Under Uncertainty of Part Quality in Fused Filament Fabrication

Abstract This work presents a data-driven methodology for multi-objective optimization under uncertainty of process parameters in the fused filament fabrication (FFF) process. The proposed approach optimizes the process parameters with the objectives of minimizing the geometric inaccuracy and maximizing the filament bond quality of the manufactured part. First, experiments are conducted to collect data pertaining to the part quality. Then, Bayesian neural network (BNN) models are constructed to predict the geometric inaccuracy and bond quality as functions of the process parameters. The BNN model captures the model uncertainty caused by the lack of knowledge about model parameters (neuron weights) and the input variability due to the intrinsic randomness in the input parameters. Using the stochastic predictions from these models, different robustness-based design optimization formulations are investigated, wherein process parameters such as nozzle temperature, nozzle speed, and layer thickness are optimized under uncertainty for different multi-objective scenarios. Epistemic uncertainty in the prediction model and the aleatory uncertainty in the input is considered in the optimization. Finally, Pareto surfaces are constructed to estimate the tradeoffs between the objectives. Both the BNN models and the effectiveness of the proposed optimization methodology are validated using the actual manufacturing of the parts.