A Comparative Study of State-of-the-Art Evolutionary Multi-objective Algorithms for Optimal Crop-mix planning

In this paper we propose a methodology based on an evolutionary multiobjective optimization algorithm for the problem of up-grading (expansion) the physical topology of transparent optical networks. The aim is to define the new topology layout and the specification of the new optical devices that should be deployed in an existing network in order to improve its performance, considering simultaneously the total installation cost (capital cost) and the total network blocking probability (performance criteria). To accomplish that, we propose a capital cost model for the network. We considered the following physical layer impairments: losses in optical devices, amplified spontaneous emission in optical amplifier and homodyne crosstalk in optical cross connect. The multiobjective optimization evolutionary algorithm used in the optimization process is the NSGA-II. Our proposed methodology can solve the expand physical topology design problem taking into account the physical layer impairments and the capital costs simultaneously. We also present a case study to show the effectiveness of our methodology.

Using Pareto-based multi-objective Evolution algorithms in decision structure to transfer the hazardous materials to safety storage centre

Multi-objective Cascade Control of Regulatory Process with Two Conflicting Objectives

Substation automation systems (SAS) are critical infrastructures whose design and maintenance must be optimised to guarantee a suitable performance. In order to provide a collection of solutions that balance availability and cost, this paper explores the optimisation of the design and maintenance of a section of SAS. Multiobjective evolutionary algorithms are combined with discrete event simulation while the performance of two state‐of‐the‐art multiobjective evolutionary algorithms is studied. On the one hand, the nondominated sorting genetic algorithm II (NSGA‐II), and on the other hand, the S‐metric selection evolutionary multiobjective optimisation algorithm (SMS‐EMOA). Such a problem is solved from 2 and 3‐objective approaches by attending to the multiobjectivisation concept. The robustness of the methodology is brought to light, and benefits were observed from the multiobjectivisation approach. Decision‐makers can employ this knowledge to make informed decisions based on economic and reliability criteria.

The non-dominated sorting genetic algorithm II (NSGA-II) is the most intensively used multi-objective evolutionary algorithm (MOEA) in real-world applications. However, in contrast to several simple MOEAs analyzed also via mathematical means, no such study exists for the NSGA-II so far. In this work, we show that mathematical runtime analyses are feasible also for the NSGA-II. As particular results, we prove that with a population size larger than the Pareto front size by a constant factor, the NSGA-II with two classic mutation operators and three different ways to select the parents satisfies the same asymptotic runtime guarantees as the SEMO and GSEMO algorithms on the basic OneMinMax and LOTZ benchmark functions. However, if the population size is only equal to the size of the Pareto front, then the NSGA-II cannot efficiently compute the full Pareto front (for an exponential number of iterations, the population will always miss a constant fraction of the Pareto front). Our experiments confirm the above findings.

Almost all real-world and engineering problems involve multi-objective optimization of some sort that is often constrained. To solve these constrained multi-objective optimization problems, constrained multi-objective optimization evolutionary algorithms (CMOEAs) are enlisted. These CMOEAs require specific constraint handling techniques. This study aims to address and test the most successful constraint handling techniques, seven different penalty constraint techniques, as applied to the Non-dominated Sorting Genetic Algorithm II (NSGA-II). In this paper, NSGA-II is chosen because of its high popularity amongst evolutionary algorithms. Inverted Generational Distance and Hypervolume are the main metrics that are discussed to compare the constraint handling techniques. NSGA-II is applied on 13 constrained multi-objective problems known as CF1-CF10, C1-DTLZ1, C2-DTLZ2, and C3-DTLZ4. The result of IGD and HV values are compared and the feasibility proportions of each combination on each problem are shown. The results of simulation present interesting findings that have been presented at the end of paper as discussion and conclusion.

Mathematical runtime analysis for the non-dominated sorting genetic algorithm II (NSGA-II)

A multi-objective genetic optimization for fast, fuzzy rule-based credit classification with balanced accuracy and interpretability

Distribution system (DS) service restoration is a very complex multi-objective and multi-constraint optimization problem, which requires high-quality Pareto fronts for helping the DS operators' work. This paper proposes a multi-objective evolutionary algorithm (MOEA) that combines the advantages of MOEAs in Tables with the properties of the Non-dominated Sorting Genetic Algorithm-II (NSGA-II) to recover Pareto fronts of the service problem. Its main differentials are its ability of recovering high-quality Pareto fronts, even in large-scale DSs, and of prioritizing switching operations in Remotely Controlled Switches (RCSs), independently of amount of RCSs. In opposite of Manually Controlled Switches (MCSs), RCS can be faster and cheapest operated from operating center. Several tests were carried out to validate and to compare the proposed MOEA to three MOEAs published in literature. Four metrics were used for assessing the quality of Pareto fronts generated and a Welch's t hypothesis test enabled the comparison of MOEAs' performance. Test results indicate the proposed MOEA outperforms those three MOEAs published in literature.

Water distribution networks (WDN) model optimization is an important part of smart water systems to achieve optimal strategies. WDN optimization focuses on the nonlinearity of the discharge head loss equation, the availability of discrete properties of pipe sizes, and the conservation of constraints. Multi-objective evolutionary algorithms (MOEAs) have been proposed and successfully applied in the field of WDN design optimization. Previous studies have focused on comparing the optimization effects of algorithms in water distribution networks, ignoring the problems of unbalanced pressure distribution and water hammer at the nodes of the pipe network caused by the complex terrain in mountainous areas. In this paper, a multi-objective water supply optimization model that integrated cost, reliability, and water quality was established for a mountainous WDN in real engineering. The method of traversing the nodes to solve the water age was introduced to find a more scientific and practical water age solution model, with setting the weight function to evaluate the water age of the water supply model comprehensively. Strength Pareto Evolutionary Algorithm II (SPEA-II) and Non-dominated Sorting Genetic Algorithm II (NSGA-II) were adopted to optimize the WDN design model in the mountainous complex terrain. The significance levels of the number of Pareto solutions (NOPS) and running time are 0.029 and 0.001, respectively, indicating that the two algorithms have significant differences. Compared to NSGA-II, SPEA-II has a better convergence rate and running time in multi-objective water supply optimization design. The solution set distribution of SPEA-II is more concentrated than that of NSGA-II, also the numerical value is better. The number of SPEA-II optimization schemes is larger and the scheme is more effective. Among them, the Pareto solution set of SPEA-II can obtain more desirable optimization results on cost, reliability index (RI) and water age. In summary, the study provides valuable information for decision makers in WDN with complex terrain.

In this study, we define the criteria for fund allocation in an investment portfolio based on three key issues: maximizing returns, minimising risk, and optimal asset allocation. The context of solving these issues reveals that the best solutions are not those that sequentially maximise or minimise each criterion but rather those that achieve an optimal compromise between them, known in the specialised literature as the Pareto front. To identify a set of nondominated solutions, we utilise a specialized evolutionary algorithm for multi-objective optimisation, the Nondominated Sorting Genetic Algorithm II (NSGA-II). This is a fast and elitist evolutionary algorithm based on a process of sorting and selecting the best agents for the repopulation of new solving sets. By using this algorithm, we generate different sets of possible solutions, also testing various mutation rates of the agents to study different approaches to favourable combinations for fund allocation. The subjects of these iterations will be a set of some of the most successful assets listed on the Bucharest Stock Exchange, simultaneously including a considerable part of the Bucharest Exchange Trading Index, over a period that encompasses both the COVID-19 pandemic and the Ukrainian war shocks. Subsequently, we evaluate the performance of these portfolio weights over time, analysing their performance and identifying differences in the evolutionary genome behaviour in comparison to the traditional Markovitz method of quadratic mean-variance equation.

Most evolutionary multiobjective optimization algorithms are designed to approximate the entire Pareto front. During the last decade, a series of preference-based evolutionary algorithms have been developed, where a part of the Pareto front is approximated by incorporating the preferences of a Decision Maker. However, only a few such algorithms are able to obtain well-distributed solutions covering the complete “region of interest” that is determined by a reference point. In this paper, a preference-based evolutionary algorithm for approximating the region of interest is proposed. It is based on the state-of-the-art genetic algorithm NSGA-II and the CHIM approach introduced in the NBI method which is used to obtain uniformly distributed solutions in the region of interest. The efficiency of the proposed algorithm has been experimentally evaluated and compared to other state-of-the-art multiobjective preference-based evolutionary algorithms by solving a set of multiobjective optimization benchmark problems. It has been shown that the incorporation of the Decision Maker’s preferences and the CHIM approach into the NSGA-II algorithm allows approximating the whole region of interest accurately while maintaining a good distribution of the obtained solutions.

In recent years, multi-objective evolutionary optimization algorithms have shown success in different areas of research. Due to their efficiency and power, many researchers have concentrated on adapting evolutionary algorithms to generate Pareto solutions. This paper proposes a new memetic adaptive multi-objective evolutionary algorithm that is based on a three-term backpropagation network (MAMOT). This algorithm is an automatic search method for optimizing the parameters and performance of neural networks, and it relies on the use of the adaptive non-dominated sorting genetic algorithm-II integrated with the backpropagation algorithm, being used as a local search method. The presented MAMOT employs a self-adaptive mechanism toward improving the performance of the proposed algorithm and a local optimizer improving all the individuals in a population in order to obtain better accuracy and connection weights. In addition, it selects an appropriate number of hidden nodes simultaneously. The proposed method was applied to 11 datasets representing pattern classification problems, including two-class, multi-class and complex data reflecting real problems. Experiments were performed, and the results indicated that the proposed method is viable in pattern classification tasks compared to a multi-objective genetic algorithm based on a three-term backpropagation network (MOGAT) and some of the methods mentioned in the literature. The statistical analysis results of the t test and Wilcoxon signed-ranks test also show that the performance of the proposed method is significantly better than MOGAT.

Solving a tri-objective supply chain problem with modified NSGA-II algorithm

Multi-objective optimization of a stacked neural network using an evolutionary hyper-heuristic