Robust Optimization Process Research Articles

Exploring the dosimetric impact of systematic and random setup uncertainties in robust optimization of head and neck IMPT plans

PurposeThis study aims to compare the dosimetric impact of incorporating systematic and random setup uncertainties in the robust optimization of head and neck cancer (HNC) Intensity Modulated Proton Therapy (IMPT) plans. MethodsBilateral HNC patients (n = 10) previously treated with conventional photon therapy at our institution were included. Both systematic and random setup uncertainties were incorporated into the robust optimization process of IMPT planning. Dosimetric comparisons were made between plans optimized with systematic (IMPT-S) versus random (IMPT-R) setup uncertainties, assessing both the clinical target volume (CTVs) and organs at risk (OARs) across various dosimetric metrics. Both plans applied a fixed range uncertainty of ± 3 % and a maximum setup uncertainty of ± 3 mm. ResultsBoth IMPT-S and IMPT-R plans achieved similar target coverage, meeting robustness criteria for CTVs. On average, the D95% voxel-wise min to the high-risk CTV (CTV_HR) was slightly higher in IMPT-S plans by 1.78 ± 0.72 % compared to IMPT-R plans. However, IMPT-R plans provided better OAR sparing, which was evident in both nominal and voxel-wise maximum values. While random setup errors in robust optimization improved OAR sparing, the clinical impact may be minimal where OAR doses are already well below tolerance levels. ConclusionBoth IMPT-S and IMPT-R techniques met the robustness criteria for CTVs in HNC IMPT planning. Incorporating random setup uncertainties in robust optimization improves OAR sparing compared to systematic setup uncertainties. Further research is needed to explore the broader applicability of random setup errors and to integrate random uncertainties in robustness evaluations for a more comprehensive assessment of treatment plans.

Direct MultiSearch optimization of TPMS scaffolds for bone tissue engineering

BackgroundScaffolds designed for tissue engineering must consider multiple parameters, namely the permeability of the design and the wall shear stress experienced by the cells on the scaffold surface. However, these parameters are not independent from each other, with changes that improve wall shear stress, negatively impacting permeability and vice versa. This study introduces a novel multi-objective optimization framework using Direct MultiSearch (DMS) to design triply periodic minimal surface (TPMS) scaffolds for bone tissue engineering. MethodThe optimization algorithm focused on maximizing the permeability of the scaffolds and obtaining a desired value of average wall shear stress (which ranges between the values that promote osteogenic differentiation of 0.1 mPa and 10 mPa). Multiple fluid inlet velocities and target wall shear stress were analyzed. The DMS method successfully generated Pareto fronts for each configuration, enabling the selection of optimized scaffolds based on specific structural requirements. ResultsThe findings reveal that increasing the target wall shear stress results in a greater number of non-dominated points on the Pareto front, highlighting a more robust optimization process. Additionally, it was also demonstrated that the tested Schwartz diamond scaffolds had a better permeability-wall shear stress relation when compared to Schoen gyroid geometries. ConclusionsDirect MultiSearch was proven as an effective tool to aid in the design of tissue engineering scaffolds. This adaptable optimization framework has potential applications beyond bone tissue engineering, including cartilage tissue differentiation.

Banked Primary Progenitor Cells for Allogeneic Intervertebral Disc (IVD) Therapy: Preclinical Qualification and Functional Optimization within a Cell Spheroid Formulation Process.

Background/Objectives: Biological products are emerging as therapeutic management options for intervertebral disc (IVD) degenerative affections and lower back pain. Autologous and allogeneic cell therapy protocols have been clinically implemented for IVD repair. Therein, several manufacturing process design considerations were shown to significantly influence clinical outcomes. The primary objective of this study was to preclinically qualify (chondrogenic potential, safety, resistance to hypoxic and inflammatory stimuli) cryopreserved primary progenitor cells (clinical grade FE002-Disc cells) as a potential cell source in IVD repair/regeneration. The secondary objective of this study was to assess the cell source's delivery potential as cell spheroids (optimization of culture conditions, potential storage solutions). Methods/Results: Safety (soft agar transformation, β-galactosidase, telomerase activity) and functionality-related assays (hypoxic and inflammatory challenge) confirmed that the investigated cellular active substance was highly sustainable in defined cell banking workflows, despite possessing a finite in vitro lifespan. Functionality-related assays confirmed that the retained manufacturing process yielded strong collagen II and glycosaminoglycan (GAG) synthesis in the spheroids in 3-week chondrogenic induction. Then, the impacts of various process parameters (induction medium composition, hypoxic incubation, terminal spheroid lyophilization) were studied to gain insights on their criticality. Finally, an optimal set of technical specifications (use of 10 nM dexamethasone for chondrogenic induction, 2% O2 incubation of spheroids) was set forth, based on specific fine tuning of finished product critical functional attributes. Conclusions: Generally, this study qualified the considered FE002-Disc progenitor cell source for further preclinical investigation based on safety, quality, and functionality datasets. The novelty and significance of this study resided in the establishment of defined processes for preparing fresh, off-the-freezer, or off-the-shelf IVD spheroids using a preclinically qualified allogeneic human cell source. Overall, this study underscored the importance of using robust product components and optimal manufacturing process variants for maximization of finished cell-based formulation quality attributes.

Data-driven drug treatment: enhancing clinical decision-making with SalpPSO-optimized GraphSAGE

Safe drug recommendation systems play a crucial role in minimizing adverse drug reactions and enhancing patient safety. In this research, we propose an innovative approach to develop a safety drug recommendation system by integrating the Salp Swarm Optimization-based Particle Swarm Optimization (SalpPSO) with the GraphSAGE algorithm. The goal is to optimize the hyper parameters of GraphSAGE, enabling more accurate drug-drug interaction prediction and personalized drug recommendations. The research begins with data collection from real-world datasets, including MIMIC-III, Drug Bank, and ICD-9 ontology. The databases provide comprehensive and diverse clinical data related to patients, diseases, and drugs, forming the foundation of a knowledge graph. It represents drug-related entities and their relationships, such as drugs, indications, adverse effects, and drug-drug interactions. The knowledge graph’s integration of patient data, disease ontology, and drug information enhances the system’s accuracy to predict drug-drug interactions as well as identifying potential detrimental drug reactions. The GraphSAGE algorithm is employed as the base model for learning node embeddings in the knowledge graph. To enhance its performance, we propose the SalpPSO algorithm for hyper parameter optimization. SalpPSO combines features from Salp Swarm Optimization and Particle Swarm Optimization, offering a robust and effective optimization process. The optimized hyper parameters lead to more reliable and accurate drug recommendation system. For evaluation, the dataset is split into training and validation sets and compared the performance of the modified GraphSAGE model with SalpPSO-optimized hyper parameters to the standard models. The experimental analysis conducted in terms of various measures proves the efficiency of the proposed safe recommendation system, offering valuable for healthcare experts in making more informed and personalized drug treatment decisions for patients.

Antiacne Gel Containing Aloe vera and Clindamycin Phosphate: Design, Characterization, and Optimization Using Response Surface Methodology

Clindamycin phosphate is a topical antibiotic agent used to treat acne vulgaris, while Aloe vera has both antimicrobial and anti-inflammatory properties. The current study is aimed at formulating an antiacne gel with antioxidant and antimicrobial effects. The antiacne gels were prepared by using polymer HPMC K15M by cold dispersion method. Unveiling the intricacies of gel design, our research harnessed the power of Design Expert 11 to optimize critical parameters—viscosity, spreadability, and permeability. In vitro characterization tests, including pH, spreadability, viscosity, permeability, antimicrobial activity, antioxidant activity, and stability of the gels, were performed. The results of in vitro characterization tests showed that the gels had a mint-like odor, a pH of 6.8, and a spreadability of 21.5 g cm/sec. The gels had a viscosity of 34.2 Pa s and drug content ranging within 90%-110%, as per USP standards. Notably, in vitro permeation assays reveal an exceptional 86% drug release, showcasing the efficacy of our formulation. The uniqueness of our study lies not only in the robust optimization process but also in the multifaceted characterization. Our gel emerges as a promising candidate, exhibiting not only desired antimicrobial and antioxidant properties against acne vulgaris but also demonstrating stability under varied conditions. As we advance toward in vivo studies, our research paves the way for a nuanced understanding of the safety and efficacy of this distinctive antiacne gel.

In-mold condition-centered and explainable artificial intelligence-based (IMC-XAI) process optimization for injection molding

This paper proposes a novel injection molding process optimization method based on the in-mold condition (IMC) and interpreted influence of in-mold condition features on part quality. In-mold condition is crucial for process optimization because it represents the actual process condition in the cavity where the polymer material is formed into the final part shape. Traditionally, the analysis of in-mold condition heavily depends on domain knowledge and insight, which introduces bias and inconsistency in in-mold condition-based process optimization. This study aims at developing an intelligent, objective, and robust process optimization method that concentrates on the highly influential in-mold condition features concerning part-quality, as interpreted by explainable artificial intelligence (XAI). In this in-mold condition-centered modeling approach, the input process parameters and final part quality were associated with the in-mold condition, with their corresponding relationship modeled by two machine learning (ML) models, respectively. The effect of in-mold condition on part quality was interpreted by applying XAI on the ML model that describes the relationship between in-mold condition and part quality. Features in the in-mold condition profiles that have high influence on part quality are given more weight in the search for diverse in-mold conditions that satisfy multiple part-quality objectives. A feasibility check has been implemented to identify, among those potential in-mold conditions, the optimal one that is physically feasible based on the ML model that governs the relationship between the process parameters and in-mold condition. The proposed method not only pinpoints better optimized process parameters than the conventional approach that omits the in-mold condition and only considers the direct relationship between the process parameters and part quality, but also reveals the possibility of further quality improvement. The optimization tool of the proposed method can be found on an online interactive platform https://imc-xai-injmold-optimization-4e26a3e3ef41.herokuapp.com/, which was created to facilitate further research based on the proposed approach. In addition to process optimization, this approach can effectively contribute to intelligent manufacturing management and Industry 4.0.

In-situ Printability Maps (IPM): A new approach for in-situ printability assessment with application to extrusion-based bioprinting

3D Bioprinting is an emerging field with many highly valuable applications. The most common and versatile technology is extrusion-based bioprinting, which requires extensive experimental campaigns to achieve appropriate quality of the bioprinted constructs when new bioinks or complex geometrical constructs need to be considered. This paper presents a new approach to easily guide operators and scientists to evaluate the probability of successful bioprinting in a defined window of the process parameters, starting from a small experimental campaign and relying on in-situ quality data. The proposed method consists of defining printability maps based on a probabilistic approach. These maps assess the printing outcome considering a specified acceptable deviation from the nominal geometry, which is predefined by the end-user depending on the application at hand. Even if shown with reference to extrusion-based bioprinting, the proposed method can be used with any other bioprinting process and any quality index, including categorical assessment classification. Eventually, the paper shows how the map can be combined with different quality criteria (e.g., productivity, cell viability) to define the appropriate setting, depending on the application at hand. Furthermore, the map provides a practical tool for rapid material printability assessment and robust process optimization. It offers an enhanced visual representation of the process domain, acceptable region boundaries, and their resilience to variation and uncertainties. Eventually, in-situ printability maps represent a further leap for the advancement of bioprinting toward the digital transformation, aiming at increasing the controllability and scalability of the bioprinting process.

Bayesian design optimization of biomimetic soft actuators

The design of versatile soft actuators remains a challenging task, as it is a complex trade-off between robotic adaptability and structural complexity. Recently, researchers have used statistical and physical models to simulate the mechanical behavior of soft actuators. These simulations can help identify optimal actuator designs that fulfill specific robotic objectives. However, automated optimization of soft robots is a delicate balance between simplifying assumptions that reduce predictive fidelity and expensive simulations that limit design space exploration. Here we propose a generalized Bayesian optimization method to identify the designs of fiber-based biomimetic soft-robotic arms that minimize the actuation energy under arbitrary robotic control requirements. We use the reduced-order active filament theory as the overarching design paradigm and mechanical model, which enables a computationally robust and efficient optimization process. We evaluate the performance of our Bayesian optimization for a simple control objective in which the actuator has to reach a given target position. We show that our proposed optimization scheme outperforms a random-search baseline; it identifies more desirable designs faster and more frequently. Although we illustrate the performance of our approach for a single actuation problem, the derived method easily generalizes to the design optimization of fiber-based actuators under a large family of robotic applications.

Robust Optimization of Natural Laminar Flow Airfoil Based on Random Surface Contamination

Natural laminar-flow (NLF) airfoils are one of the most promising technologies for extending the range and endurance of aircrafts. However, there is a lack of methods for the optimization of airfoils based on the surface contamination that destroys the laminar flow. In order to solve this problem, a robust optimization process is proposed using the Non-dominated Sorting genetic algorithm- II (NSGA-II) evolutionary algorithm, and Monte Carlo simulation combined with an aerodynamic calculation software Xfoil. Firstly, the airfoil is optimized normally and the aerodynamic performance of optimized airfoil under surface contamination is analyzed. Then, the original airfoil is robustly optimized under random surface contamination based on the assumption that its locations follow triangular and uniform probability distributions. Finally, all the optimized results and original airfoil are compared. It is found that robust optimization reduces the sensitivity of the airfoil to random surface contamination, hence, improving the robustness of the airfoil. The proposed methods make it possible to improve the aerodynamic performance of NLF airfoils considering surface contamination.

Integrating multi-objective superstructure optimization and multi-criteria assessment: a novel methodology for sustainable process design

Abstract This work presents a novel methodology for integrated multi-objective superstructure optimization and multi-criteria assessment. The method is tailored for sustainable process synthesis utilizing mixed-integer linear programming (MILP). The six-step algorithm includes 1) superstructure formulation, 2) criteria definition and implementation, 3) criteria weighting, 4) single-criterion optimization, 5) reformulation and 6) multi-criteria optimization. It is automated in the O pen s U perstruc T ure mo D eling and O ptimizati O n f R amework (OUTDOOR) and tested on integrated power-to-X and biomass-to-X processes for methanol production. Three criteria are considered, namely net production costs (NPC), net production greenhouse gas emissions (NPE) and net production fresh water demand (NPFWD). The optimization indicates NPC of 1307 €/tMeOH with NPE of −2.23 t CO 2 / t MeOH ${\text{t}}_{{\text{CO}}_{2}}/{\text{t}}_{\text{MeOH}}$ and NPFWD of −3.42 t H 2 O / t MeOH ${\text{t}}_{{\text{H}}_{2}\text{O}}/{\text{t}}_{\text{MeOH}}$ for an optimal trade-off plant. The plant configuration features low-pressure alkaline electrolysis for hydrogen supply, absorption-based CO2 capture and steam production from methanol purge gas for internal heat supply. Conducted variation and sensitivity analyses indicate that methanol costs can drop to about 500 €/tMeOH if electricity is free of charge, or to 805 €/tMeOH if biogas is available at large quantities, if a least-cost process layouts are considered. However, all performed multi-criteria analyses imply a robust optimal process design utilizing electricity-based methanol production.

Robust optimization of composite cylindrical shell by a trigonometric mixed response surface method

AbstractThe response surface method (RSM) as a popular fitting method has been applied to numerous engineering issues. However, the existing RSM models cannot satisfy the analysis of complex nonconvex structure problems, especially in composite cylindrical shells. Thus a trigonometric mixed response surface method (TMRSM) is suggested to improve the accuracy and the fitting performance combing optimal Latin hypercube design (LHD), the moving least square method and the trigonometric functions. An improved robust design optimization (RDO) process is performed based on the weighted norm method and reliable constraints using the above TMRSM model to obtain more robust optimal design results. Three example functions and a composite cylindrical pressure shell are given to verify the effectiveness of the TMRSM model. This pressure shell design can check the robustness of the improved RDO process.

Prediction of forming limit curve for AA6061-T6 at room and elevated temperatures

Forming Limit Curve (FLC) has been widely adopted as a practical criterion for evaluating the formability of sheet metals. Predicting a reliable FLC by a virtual methodology could lead to robust process optimization before expensive tool manufacturing. In order to increase the predictive capabilities of the virtual forming tools, an accurate modeling of the forming limit curve should be considered at room and elevated temperatures. In this work, the isothermal forming limit curves of 6000 series aluminum alloy sheet metal are predicted by performing numerical simulations of Nakajima test. A stress triaxiality and Lode angle based ductile fracture criterion is used to determine the forming limit curve. Also, the ductile fracture criterion is extended to add the impact of temperature on ductile fracture prediction. The hybrid experimental-numerical approach is used to calibrate the ductile fracture criterion. The forming limit curve of AA6061-T6 sheet metal, with a thickness of 1 mm, is predicted using the calibrated ductile fracture criterion at room and elevated temperatures. Numerical simulations are performed in 3D with the finite element code Abaqus. The limit strains are determined for specimens undergoing deformation under different strain paths. Influence of temperature on the predicted forming limit curve is discussed.

HiPE: Hierarchical Initialization for Pose Graphs

Pose graph optimization is a non-convex optimization problem encountered in many areas of robotics perception. Its convergence to an accurate solution is conditioned by two factors: the non-linearity of the cost function in use and the initial configuration of the pose variables. In this paper, we present HiPE, a novel hierarchical algorithm for pose graph initialization. Our approach exploits a coarse-grained graph that encodes an abstract representation of the problem geometry. We construct this graph by combining maximum likelihood estimates coming from local regions of the input. By leveraging the sparsity of this representation, we can initialize the pose graph in a non-linear fashion, without computational overhead compared to existing methods. The resulting initial guess can effectively bootstrap the fine-grained optimization that is used to obtain the final solution. In addition, we perform an empirical analysis on the impact of different cost functions on the final estimate. Our experimental evaluation shows that the usage of HiPE leads to a more efficient and robust optimization process, comparing favorably with state-of-the-art methods.

Feature Preserving Non-Rigid Iterative Weighted Closest Point and Semi-Curvature Registration.

Preserving features of a surface as characteristic local shape properties captured e.g. by curvature, during non-rigid registration is always difficult where finding meaningful correspondences, assuring the robustness and the convergence of the algorithm while maintaining the quality of mesh are often challenges due to the high degrees of freedom and the sensitivity to features of the source surface. In this paper, we present a non-rigid registration method utilizing a newly defined semi-curvature, which is inspired by the definition of the Gaussian curvature. In the procedure of establishing the correspondences, for each point on the source surface, a corresponding point on the target surface is selected using a dynamic weighted criterion defined on the distance and the semi-curvature. We reformulate the cost function as a combination of the semi-curvature, the stiffness, and the distance terms, and ensure to penalize errors of both the distance and the semi-curvature terms in a guaranteed stable region. For a robust and efficient optimization process, we linearize the semi-curvature term, where the region of attraction is defined and the stability of the approach is proven. Experimental results show that features of the local areas on the original surface with higher curvature values are better preserved in comparison with the conventional methods. In comparison with the other methods, this leads to, on average, 75%, 8% and 82% improvement in terms of quality of correspondences selection, quality of surface after registration, and time spent of the convergence process respectively, mainly due to that the semi-curvature term logically increases the constraints and dependency of each point on the neighboring vertices based on the point's degree of curvature.

Robust utility maximization under model uncertainty via a penalization approach

This paper addresses the problem of utility maximization under uncertain parameters. In contrast with the classical approach, where the parameters of the model evolve freely within a given range, we constrain them via a penalty function. We show that this robust optimization process can be interpreted as a two-player zero-sum stochastic differential game. We prove that the value function satisfies the Dynamic Programming Principle and that it is the unique viscosity solution of an associated Hamilton-Jacobi-Bellman-Isaacs equation. We test this robust algorithm on real market data. The results show that robust portfolios generally have higher expected utilities and are more stable under strong market downturns. To solve for the value function, we derive an analytical solution in the logarithmic utility case and obtain accurate numerical approximations in the general case by three methods: finite difference method, Monte Carlo simulation, and Generative Adversarial Networks.

A Review of the Robust Optimization Process and Advances with Monte Carlo in the Proton Therapy Management of Head and Neck Tumors.

In intensity-modulated proton therapy, robust optimization processes have been developed to manage uncertainties associated with (1) range, (2) setup, (3) anatomic changes, (4) dose calculation, and (5) biological effects. Here we review our experience using a robust optimization technique that directly incorporates range and setup uncertainties into the optimization process to manage those sources of uncertainty. We also review procedures for implementing adaptive planning to manage the anatomic uncertainties. Finally, we share some early experiences regarding the impact of uncertainties in dose calculation and biological effects, along with techniques to manage and potentially reduce these uncertainties.

A Guideline for Implementing a Robust Optimization of a Complex Multi-Stage Manufacturing Process

In the industrial production scenario, the goal of engineering is focused on the continuous improvement of the process performance by maximizing the effectiveness of the manufacturing and the quality of the products. In order to address these aims, the advanced robust process optimization techniques have been designed, implemented, and applied to the manufacturing process of ultrasound (US) probes for medical imaging. The suggested guideline plays a key role for improving a complex multi-stage manufacturing process; it consists of statistical methods applied for improving the product quality, and for achieving a higher productivity, jointly with engineering techniques oriented to problem solving. Starting from the Six Sigma approach, the high definition of the production process was analyzed through a risk analysis, and thus providing a successful implementation of the PDCA (plan-do-check-act) methodology. Therefore, the multidisciplinary analysis is carried out by applying statistical models and by detecting the latent failures by means of NDT (non-destructive testing), i.e., scanning acoustic microscopy (SAM). The presented approach, driven by the statistical analysis, allows the engineers to distinguish the potential weak points of the complex manufacturing, in order to implement the corrective actions. Furthermore, in this paper we illustrate this approach by considering a pilot study, e.g., a process of US probes for medical imaging, by detailing all the guideline steps.

Design of porous and graded NiTi smart energy absorbers considering synthetic uncertainty in parameters

The synthetic uncertainty (SU) method is introduced and applied to the optimal design of energy absorbing NiTi shape memory alloy (SMA) bars. A sensitivity analysis for a large number of stochastic parameters identifies geometrical grading, porosity, and imposed maximum nominal stress as critical design parameters for the energy dissipation capacity of the bars. Parametric uncertainty on the optimal design of the energy absorber is incorporated and estimated through the SU formalism. The SU method provides a unified approach to discover the critical design parameters, quantify uncertainty, and optimally design a system around its extreme response (maximum or minimum). Therefore, the SU method is placed next to the robust design optimization (RDO) process, yet with a discovery component. It is found that variations in porosity and shape factor can significantly alter the stress-strain response and energy dissipation capacity. For a given value of maximum nominal stress, it is found that there exists an optimal combination of shape factor and porosity which maximizes the energy dissipation capacity of a tapered and porous NiTi bar.

Optimización de armaduras espaciales de acero utilizando algoritmos genéticos auto-adaptados: una primera aproximación

En las últimas décadas, la optimización estructural mediante metaheurísticas ganó acogida en la comunidad científica; sin embargo, para garantizar buenos resultados se requiere una correcta selección de los parámetros de la metaheurísticas. En este trabajo se propone un algoritmo genético multi-cromosoma auto-adaptado para optimizar armaduras de acero tridimensionales. Las variables de diseño corresponden a las secciones asignadas a cada elemento en la armadura. El objetivo es la minimización del peso de la armadura, considerando desplazamientos y esfuerzos máximos como restricciones. El algoritmo propuesto se aplicó a la optimización de dos armaduras, produciendo diseños que pesan hasta un 35% menos que el mejor diseño inicial y son valores comparables al resultado obtenidos en otros trabajos. No obstante, la adaptación de los parámetros permite mayor robustez cuando se desea optimizar diferentes tipos de armadura y evita las ejecuciones del algoritmo de optimización que son necesarias para la calibración de sus parámetros.

Robust-Oriented Optimization Design for Permanent Magnet Motors Considering Parameter Fluctuation

In this article, a robust-oriented optimization design method is proposed for the permanent magnet (PM) motors. The key of the method is to introduce the concept of parameter fluctuation into motor design process, which aims at solving the uncertainties during motor manufacturing process. The optimization design process is purposely divided into two layers. In the Layer I, except for the motor performance objectives, the influence of parameters fluctuation on motor performances is also served as the design objectives to obtain the deterministic motor structure. In the Layer II, the allowable manufacture errors caused by the parameter fluctuation and the manufacturing qualification rate of the motor with the errors are calculated and optimized simultaneously. Then, an interior PM (IPM) motor is chosen as a design example during robust optimization process. For verifying the effectiveness of the optimization design, the performances are evaluated in detail and the prototype motor is built. Finally, both theoretical analysis and experimental results validate the effectiveness of design method and the rationality of PM motor.