Objective Function Of Model Research Articles

Hypergraph-based convex semi-supervised unconstraint symmetric matrix factorization for image clustering

Semi-supervised symmetric nonnegative matrix factorization (SNMF) has been extensively utilized in both linear and nonlinear data clustering tasks. However, the current SNMF model's non-convex objective function faces challenges in global optimization and time efficiency. In this study, we leverage label information to propose a convex and unconstrained symmetric matrix factorization (SMF) model that is thoroughly analyzed for its convexity properties. In order to capture high-order relationships among data, a hypergraph is utilized in the model, which is computationally simple, translation invariant, and naturally normalized. Moreover, based on the analysis and the corresponding experiments in the paper, the model exhibits robustness towards outliers to some extent. Due to the convexity of our proposed model without constraint, it can be efficiently optimized using the Conjugate Gradient (CG) method, one of the most efficient methods available. Therefore, we propose a novel Convex Combination-based Sufficient Descent CG (CSDCG) method, which outperforms other methods across 284 optimization problems within the CUTEst library. In order to evaluate the effectiveness of the proposed method, the semi-supervised clustering experiments are conducted on the eight datasets by comparison with ten state-of-the-art matrix factorization (MF) methods. The experiment results demonstrate its superiority over the other compared methods to handle the clustering problem with better performance and less computational time. The code is available at https://github.com/Pokemer/HCSSMF.

Integrated optimization of crude oil procurement planning and blending scheduling for property stabilization

Crude oil procurement and blending are key processes in refinery production. When solving the integrated optimization problem of simple oil procurement and blending, the following issues are mainly considered: how to establish an integrated optimization model of crude oil procurement and blending process under different time scales to minimize the cost of procurement under the premise of ensuring the stability of the properties of blended crude oil. The paper proposes an integrated optimization model, takes cost minimization as the model's objective function, and provides stability through the yield and property constraints of blended crude oil. Then, the paper describes the procurement process by an event-based representation and the blending process by a continuous event representation and proposes a hybrid event-time-based representation to describe the model. Finally, this paper verifies the model's effectiveness in solving real production problems through case simulations.

A data-driven robust optimization in viable supply chain network design by considering Open Innovation and Blockchain Technology

The research proposes a new method for Viable Supply Chain Network Design (VSCND) that incorporates Open Innovation (OI) and Blockchain Technology (BCT). A robust stochastic optimization and Conditional Value at Risk (CVaR) in the cost function is utilized to develop the model. This model's objective function incorporates the expected value, maximum cost, and CVaR cost. The OI and BCT platforms are suggested for an antifragile policy against disruption. In addition, CO2 emissions and energy consumption are proposed as sustainability requirements. Eventually, minimum demand satisfaction and resilience facilities by incorporating capacity based on specific scenarios are added to the model for an agile strategy. This research entails several parties, including vendors supplying the primary components and manufacturers creating products based on customer preferences. OI and BCT platforms aim to receive demand based on customer specifications and facilitate rapid transactions between components. Risk-averse decision-makers utilize a polyhedral Data-Driven Robust Optimization (DDRO) approach to manage uncertainty and flexible sets. Incorporating OI and BCT as antifragility instruments resulted in a 0.2% cost reduction for VSCNDOIBCT compared to when OI and BCT were not considered. This research suggests that integrating OI and BCT positively affects SC performance overall. The decreasing rate, DDRO coefficient, agility rate, demand variation, and problem size were subjected to a sensitivity analysis.

Optimizing base isolation system parameters using a fuzzy reinforced butterfly optimization: A case study of the 2023 Kahramanmaras earthquake sequence

This study aims to obtain the viable parameters of base isolation systems that are utilized in buildings with seismic isolation and specially located in Turkey. To provide a more accurate representation of real-world scenarios, the study accounts for severe ground motion records simultaneously during the dynamic response analysis of the structure. Three records are taken from the 2023 Turkey earthquake (records for Elbistan, Nurdagi, and Pazarcik). The optimization model's objective function is to minimize the ratio of the roof acceleration to the peak ground acceleration. To ensure realistic outcomes, certain isolation parameters are constrained, such as maintaining the damping ratio and period within a predetermined range as well as the maximum lateral drift of the structure. An upgraded variant of the Butterfly Optimization Algorithm, known as the Fuzzy Butterfly Optimization Algorithm (FBOA), is employed to address the presented optimization model. The FBOA technique incorporates a fuzzy decision-making mechanism that adjusts its search strategy according to the prevailing conditions of the present problem. In addition, the FBOA method utilizes a new auxiliary vector called virtual butterfly, which provides information on the entire population based on each agent's quality. Consequently, FBOA transforms into a self-regulating search algorithm, eliminating the requirement for any external parameters. Results have verified the effectiveness of FBOA in detecting optimal parameters of the proposed isolation system.

Data augmentation strategy for power inverter fault diagnosis based on wasserstein distance and auxiliary classification generative adversarial network

With the rapid development of new energy vehicles, the brushless DC motor (BLDCM) drive system's reliability and safety have attracted extensive attention. The three-phase full-bridge inverter (TFI) of the BLDCM drive system has a high fault occurrence rate under actual working conditions. It is difficult to identify the fault directly, which leads to imbalanced fault datasets. In addition, it is challenging to obtain fault samples directly, which increases the difficulty of fault diagnosis. In response to these problems, a data augmentation method based on Wasserstein distance and auxiliary classification generative adversarial network (WAC-GAN) for TFI fault diagnosis has been proposed. First, based on the Auxiliary Classification Generative Adversarial Network (ACGAN), one-dimensional convolutions are constructed to replace two-dimensional convolutions for the characteristics of a three-phase current signal to improve the extraction efficiency of signal features. Then, the Wasserstein distance is introduced to improve the model's objective function. Based on the principle of the mutual game between the generator and discriminator, the generator can mine the sample distribution characteristics from few fault mode samples and generate numerous fault samples of specific categories to accomplish the purpose of data augmentation. The experimental results show that the fault diagnosis accuracy of the WAC-GAN model under different datasets and different fault modes can achieve satisfactory fault recognition performance. Compared with other data augmentation methods, the effectiveness and superiority of the proposed method has been verified.

A light robust optimization model for a hierarchical healthcare network

PurposeA growing body of evidence points to the influence of location and allocation decisions on the structure of healthcare networks. The authors introduced a three-level hierarchical facility location model to minimize travel time in the healthcare system under uncertainty.Design/methodology/approachMost healthcare networks are hierarchical and, as a result, the linkage between their levels makes it difficult to specify the location of the facilities. In this article, the authors present a hybrid approach according to data envelopment analysis and robust programming to design a healthcare network. In the first phase, the efficiency of each potential location is calculated based on the non-radial range-adjusted measure considering desirable and undesirable outputs based on a number of criteria such as the target area's population, proximity to earthquake faults, quality of urban life, urban decrepitude, etc. The locations deemed suitable are then used as candidate locations in the mathematical model. In the second phase, based on the proposed robust optimization model, called light robustness, the location and allocation decisions are adopted.FindingsThe developed model is evaluated using an actual-world case study in District 1 of Tehran, Iran and relevant results and different sensitivity analyses were presented as well. When the percentage of referral parameters changes, the value of the robust model's objective function increases.Originality/valueThe contributions of this article are listed as follows: Considering desirable and undesirable criteria to selecting candidate locations, providing a robust programming model for building a service network and applying the developed model to an actual-world case study.

Shilling attack detection for collaborative recommender systems: a gradient boosting method.

Organized malicious shilling attackers influence the output of the collaborative filtering recommendation systems by inserting fake users into the rating matrix within the database. The existence of shilling attack poses a serious risk to the stability of the system. To counter this specific security threat, many attack detection methods are proposed. Some of the past methods suffer from two disadvantages, the first being that they only analyze the rating matrix from a single perspective of user rating values and ignore other perspectives. Another is that some methods only use a single classifier to handle the classification of malicious attackers. Considering the above disadvantages, this paper proposes a gradient boosting method (named XGB-SAD) to achieve attack detection by combining double-view and gradient boosting. We first analyze the rating matrix with a double-view of time and item, which in turn defines the TPUS collection. Then our method uses eXtreme Gradient Boosting to perform heuristic iterative optimization of the model's objective function and uses the idea of ensemble learning to integrate multiple sets of base classifiers into strong classifier. The integrated strong classifiers are used to complete the detection of malicious attackers. Finally, we perform several experiments and the results demonstrate that XGB-SAD outperforms the comparison methods in terms of small-scale attack detection and overall detection, which proves the performance of our method.

Development of Order Quantity Optimization Model for Growing Item Considering the Imperfect Quality and Incremental Discount in Three Echelon Supply Chain

This research develops an optimization model for determining the order quantity for growing items by considering the imperfect quality and incremental discount by involving three supply chain members: farmers, processors, and retailers. The farmers are responsible for caring for the newborn items until they reach their ready-to-eat weight. The processors perform two roles, namely processing and screening. In the processing role, the processors process the grown items by a slaughtering and packaging process. Afterward, they inspected the processed items and categorized the items into good and poor quality. Finally, they shipped the end products to retailers. The retailers are responsible for selling good-quality items to the final consumers. This research considers two kinds of poor quality. First is the poor quality of growing items in terms of mortality rate. The second is the poor quality of final products on the processor side. The processed items with poor quality are then sold to the secondary market at lower prices in one batch at the end of the period. This model also considers the incremental discounts offered by vendors to farmers and retailers to consumers for specific amounts of purchases. The model's objective function is to maximize the total supply chain profit, with the number of orders quantity, cycle time, and the number of batches delivery set as the decision variables. The sensitivity analysis results show that the most sensitive parameter in the model is the probability that the live items survive throughout the growth period.

Predictive models with end user preference

AbstractClassical machine learning models typically try to optimize the model based on the most discriminatory features of the data; however, they do not usually account for end user preferences. In certain applications, this can be a serious issue as models not aware of user preferences could become costly, untrustworthy, or privacy‐intrusive to use, thus becoming irrelevant and/or uninterpretable. Ideally, end users with domain knowledge could propose preferable features that the predictive model could then take into account. In this paper, we propose a generic modeling method that respects end user preferences via a relative ranking system to express multi‐criteria preferences and a regularization term in the model's objective function to incorporate the ranked preferences. In a more generic perspective, this method is able to plug user preferences into existing predictive models without creating completely new ones. We implement this method in the context of decision trees and are able to achieve a comparable classification accuracy while reducing the use of undesirable features.

A fuzzy multi-objective multiple-pollutant model for rivers using an ant colony algorithm

A fuzzy multi-objective optimisation model was investigated for water quality management in a river under uncertain conditions. In this study, to deal with multiple pollutants simultaneously, the National Sanitation Foundation's water quality index was considered as one of the model's objective functions based on fuzzy set theory. This made it possible to investigate the overall effect of uncertainties on simultaneous changes. Another objective function was the total treatment costs for wastewater discharged into the river. A water quality simulation model and a non-dominated-archiving ant colony optimisation algorithm were used to determine the values of water quality parameters and the model's optimal solutions, respectively. Furthermore, a simulation–optimisation approach was adopted for facilitating the problem-solving process and applied to a hypothetical case study resembling a river system in Iran. The results show that the proposed model significantly reduced the total wastewater treatment costs compared with a similar single-objective model with a more cautious and a cost-effective approach. Although the treatment costs were increased compared with the similar deterministic model, more feasible approach was taken by considering the uncertainties associated with the objectives. With suitable modifications, the model could be easily adapted for other river systems.

A Multiobjective Aggregate Production Planning Model for Lean Manufacturing: Insights From Three Case Studies

This article proposes a multiobjective mathematical model to optimize the multiperiod aggregate production planning (APP) of multiproduct companies. Although there are many studies of lean manufacturing (LM), its integration with APP has not been studied. The present article is intended to integrate APP and LM, including an analysis of market winners, market qualifiers, and waste. The model's objective functions include the cost, lead time, and waste minimization in addition to maximizing the product quality. A solution procedure is suggested to solve the model using IBM CPLEX 12.4 software. The model is investigated in three different case studies to check its applicability and generalizability. According to the obtained results, the proposed model provides an optimized APP with regard to the major concerns of LM, including waste, overproduction, time, and sourcing. In addition, according to the sensitivity analysis, lean weighting of the objective functions provides a better output than using equal weighting.

A green closed loop supply chain design using queuing system for reducing environmental impact and energy consumption

Abstract Due to increased environmental impacts and their important role in human life, reduction of impacts made by human has attracted more attention, recently. Green supply chains are among the most effective issues related to environmental impacts and increased number of studies in this area verifies this opinion. Transportation fleets transfer products between supply chain's centers and are one of the important factors which increase environmental impacts. Transportation fleets which transfer products between supply chain's centers are one of the important factors which increase environmental impacts while transferring products between centers and waiting in loading queue. Decreasing environmental impacts which are created by transportation fleets, from this point of view, is not investigated comprehensively in forward and reverse logistic supply chains. In order to deal with this gap, in this article a green supply chain with forward and reverse logistic consideration is designed and queuing system is used to optimize the transportation and waiting time of transportation fleets' network. This optimization model will lead to the reduction in environmental impacts. Our network consists of supplier, production system, distribution center, repair center, recycling center, disposal center, and collection center. Returned products from customers are collected in the collection center and transferred to other centers based on their type. Transportation fleets in the network are assumed to be customers of loading system in each center where each of these loading systems has a multi-server queuing system with finite sources. It is assumed that a sufficient number of servers are available in unloading centers, therefore, no queue will exist there. The proposed model will reduce the created environmental impacts and energy consumption of transportation fleets by determining loading, unloading and production rates, which affect waiting and transportation time. A numerical example is discussed for the NLP model in small size and solved with the exact methods. In addition, a meta-heuristics approach is employed to solve the large size of problem. Finally, the sensitivity analysis is performed to investigate the effects of change in parameters on model's decision variables and objective function.

Modular Assessment of Rainfall–Runoff Models Toolbox (MARRMoT) v1.2: an open-source, extendable framework providing implementations of 46 conceptual hydrologic models as continuous state-space formulations

Abstract. This paper presents the Modular Assessment of Rainfall–Runoff Models Toolbox (MARRMoT): a modular open-source toolbox containing documentation and model code based on 46 existing conceptual hydrologic models. The toolbox is developed in MATLAB and works with Octave. MARRMoT models are based solely on traceable published material and model documentation, not on already-existing computer code. Models are implemented following several good practices of model development: the definition of model equations (the mathematical model) is kept separate from the numerical methods used to solve these equations (the numerical model) to generate clean code that is easy to adjust and debug; the implicit Euler time-stepping scheme is provided as the default option to numerically approximate each model's ordinary differential equations in a more robust way than (common) explicit schemes would; threshold equations are smoothed to avoid discontinuities in the model's objective function space; and the model equations are solved simultaneously, avoiding the physically unrealistic sequential solving of fluxes. Generalized parameter ranges are provided to assist with model inter-comparison studies. In addition to this paper and its Supplement, a user manual is provided together with several workflow scripts that show basic example applications of the toolbox. The toolbox and user manual are available from https://github.com/wknoben/MARRMoT (last access: 30 May 2019; https://doi.org/10.5281/zenodo.3235664). Our main scientific objective in developing this toolbox is to facilitate the inter-comparison of conceptual hydrological model structures which are in widespread use in order to ultimately reduce the uncertainty in model structure selection.

Decision model for planning material supply channels in construction

Abstract Material supply-related decisions strongly affect economic performance of construction projects. To facilitate planning, the authors put forward a mixed integer linear programming model for optimizing supplies of materials or components that are consumed irregularly. The model's objective function is minimizing the total inventory management cost. As material prices change over time and cannot be predicted with certainty, they are expressed as fuzzy values with triangular membership functions. The planning horizon is divided into units of time to allow for seasonal changes and scope variations that affect consumption of materials. The model enables the user to determine economic order quantities for consecutive periods of construction works and to select most economical supply channels of a particular material/component. The model allows for material/component substitution. Substitutes may differ in level of prefabrication and thus in costs of on-site processing. To solve the problem the fuzzy model is converted into a three-objective linear program with crisp coefficients. An example illustrates benefits of using the proposed approach in reducing the total inventory cost.

Supplier selection in closed loop supply chain by an integrated simulation-Taguchi-DEA approach

Purpose – Reverse logistics refer to processes related to the reuse of products. The role of suppliers’ performance is crucial in achieving quality, cost, service and delivery aims of a supply chain. The selection of suppliers is regarded as one of the critical issues encountered by purchasing and operations managers in a supply chain to enhance organization’s global marketplace competitiveness. Most of the supply chain models are rather complex problems. Consequently, it is impossible to propose systematic models to handle them. Therefore, in this paper a new integrated approach based on experimental design and computer simulation is proposed for supplier selection. The paper aims to discuss these issues. Design/methodology/approach – In this study, a simulation approach is implemented to determine certain equivalent of parameters values in a CLSC network design which cannot be computed through mathematical model. Suppliers’ order quantities are investigated by Taguchi method for planning time horizon. Moreover, data envelopment analysis (DEA) is applied to assess suppliers based on quality, cost, delivery time, production capabilities, services and technology. Findings – In the numerical example, there are three suppliers for different regions. Purchase value of each supplier is measured in three years successively. According to the proposed method, the authors find the minimum level of costs together with the maximum number of high-quality products. Practical implications – The objective functions of model are minimizing the costs and maximizing number of high-quality products. The integrated approach introduced in this paper enables managers to select their suppliers effectively in their real system. Originality/value – Most supply chain models are complex and the identification of proper and optimal solutions in complex real-world systems often requires the solution of multi-objective problems involving multiple stochastic variables. Therefore, the paper introduces a new integrated approach for supplier selection in closed loop supply chain. To the best of the authors knowledge, this is the first study that integrates DEA, computer simulation and Taguchi method for supplier selection in closed loop supply chains.

Minimally-Constrained Multilingual Embeddings via Artificial Code-Switching

We present a method that consumes a large corpus of multilingual text and produces a single, unified word embedding in which the word vectors generalize across languages. In contrast to current approaches that require language identification, our method is agnostic about the languages with which the documents in the corpus are expressed, and does not rely on parallel corpora to constrain the spaces. Instead we utilize a small set of human provided word translations---which are often freely and readily available. We can encode such word translations as hard constraints in the model's objective functions; however, we find that we can more naturally constrain the space by allowing words in one language to borrow distributional statistics from context words in another language. We achieve this via a process we term artificial code-switching. As the name suggests, we induce code-switching so that words across multiple languages appear in contexts together. Not only do embedding models trained on code-switched data learn common cross-lingual structure, the common structure allows an NLP model trained in a source language to generalize to multiple target languages (achieving up to 80% of the accuracy of models trained with target-language data).

Determining optimal primary sawing and ripping machine settings in the wood manufacturing chain

for wood manufacturers around the world, the single biggest cost factor is known to be its raw material. Maximum volume recovery of this raw material is, therefore, of key importance for the industry. The wood products industry consists of several interrelated manufacturing steps for converting trees into logs and logs into finished lumber. Each operation usually optimises its functionality in isolation from the preceding and following operations. It is a well documented fact that the optimisation of decisions through the whole chain of operations is considerably more profitable than the optimisation of individual operations. The objective of this study was to determine the optimal machine settings for two interrelated operations, namely the sawing and ripping operations which have tradition‑ ally been optimised individually. A model, having two decision variables, was developed which aims to satisfy market demand at a minimal cost. The first decision was how to saw the log supply into different thicknesses by choosing specific sawing patterns. The second was to decide on a rip saw's priority value settings, which determines how the products from the primary sawing operation are ripped into products of a certain thickness and width. The techniques used to determine the machine settings included static simulation with the SIMSAW software to represent the sawing operation and mixed integer programming to model the ripping operation. A metaheuristic, namely the Population Based Incremental Learning algorithm, was used to link the simulation and mixed integer models and to determine the optimal settings for the combined process. The model's objective function was to minimise the cost of production. This cost included the raw material waste cost and the over or under production cost. The over production cost included the stock keeping costs and the under production cost was estimated as the buy‑in cost of under supplied products from another wood supplier. The model performed well against current decision software available, namely the Sawmill Production Planning System package, which combines simulation and mixed integer programming techniques to maximise profit. The model added further value by modelling and determining the ripping priority settings in addition to the primary sawing patterns.

수치해석적 토지이용-교통모형의 이론연구 도구화: 교통수요의 내생화를 중심으로

토지이용-교통 통합모형은 실증연구와 정책연구에 적합한 풍부하고 유연한 모형요소를 가지고 있다. 그러나 모형의 구성이 복잡한 만큼 후생함수의 정책변수에 대한 변화율 또한 통상 복잡하다. 따라서 최적 정책수단이 충족시켜야 할 1계 도함수 조건을 푸는 최적 정책수단의 수식을 명시적으로 유도하는 것이 매우 어렵고, 그 결과 수치해석적 모형은 이론연구 도구로서 활용하는데 근본적 한계를 가지고 있었다. 이 문제를 해결하고자 Yu and Rhee(2011) 및 Rhee (2012)는 이들 모형에서 목적함수인 후생함수의 정책변수에 대한 변화율을 간단한 수식으로 변환하는 방법론을 제시한 바 있다. 그러나 이들이 사용한 모형에서 교통수요는 고정된 것으로 가정하고 있어, 이들 모형 또한 교통 계획적 관점에서 보았을 때 상당한 한계를 지니고 있다. 이에 본 연구는 이들의 방법론을 교통수요가 탄력적인 토지이용-교통 모형으로 확장한다. 이 방법론을 이용하면 보다 현실적인 모습의 토지시장과 교통망이 구현된 모형에서 정책수단의 다양한 영향을 분석할 수 있게 된다. 이러한 분석은 종래에 존재하지 않았거나 있었다 하더라도 매우 제한된 범위에서 수행되었던 분석들이다. The land use-transportation models typically have complicated model structure that is good for empirical execution but bad for theoretical probe. This complexity makes it very difficult to derive the first-order conditions for system optimization in tractable forms. Yu and Rhee (2011) and Rhee (2012) show how to simplify the derivative of the model's objective function with respect to policy variables in the computable general equilibrium model of land use and transportation. However, the travel demand in their model was fixed. This drawback fundamentally limits the applicability of their methodology in the planning field. We relax this restriction. Once this is done, we can employ the methodology developed in analyzing the impacts of various types of policy instruments in the models where land market is treated endogenously and transportation network is embedded.

Performance analysis of a human–robot collaborative target recognition system

SUMMARYThis paper presents a comprehensive analysis of time and action operational costs on an objective function developed by Becharet al. (A. Bechar, J. Meyer and Y. Edan, “An objective function to evaluate performance of human–robot collaboration in target recognition tasks,”IEEE Trans. Syst. Man Cybern. Part C39(6), 611–620 (2009)) for collaborative target recognition systems. Different task types, system reaction types, and environments were evaluated. Results reveal two types of task and system reactions – one focused on minimizing false alarms, and the second on detecting a target when one is presented. In addition, the analysis reveals a new property of the objective function based on a specific ratio between the weight differences that generalizes the model's objective function and facilitates its analysis. Results indicate that human decision time strongly influences system performance.

Combined particle swarm optimization and fuzzy inference system model for estimation of current-induced scour beneath marine pipelines

In this paper the capability of Particle Swarm Optimization (PSO) is employed to deal with an Adaptive Network based Fuzzy Inference System (ANFIS) model's inherent shortcomings to extract optimum fuzzy if–then rules in noisy areas arising from the application of nondimensional variables to estimate scour depth. In the model, a PSO algorithm is employed to optimize the clustering parameters controlling fuzzy if–then rules in subtractive clustering while another PSO algorithm is employed to tune the fuzzy rule parameters associated with the fuzzy if–then rules. The PSO model's objective function is the Root Mean Square (RMSE), by which the model attempts to minimize the error in scour depth estimation with respect to its generalization capability. To evaluate the model's performance, the experimental datasets are used as training, checking and testing datasets. Two-dimensional and nondimensional models are developed such that in the dimensional model the mean current velocity, mean grain size, water depth, pipe diameter and shear boundary velocity are used as input variables while in the nondimensional model the pipe, boundary Reynolds numbers, Froude number and normalized depth of water are set as input variables. The results show that the model provides an alternative approach to the conventional empirical formulae. It is evident that the developed PSO–FIS–PSO is superior to the ANFIS model in the noisy area in which the input and output variables are slightly related to each other.