Source-linked AI summary
Genetic Algorithm: Reviews, Implementations, and Applications
Tanweer Alam, Shamimul Qamar, Amit Dixit, Mohamed Benaida
TL;DR
The paper addresses how genetic algorithms can support engineering pedagogy and be implemented in emerging areas. It reviews GA concepts, operators, methodologies, applications, and implementations, reporting a close-optimal route in its traveling-salesman discussion while noting that the described fitness approach needs improvement for optimal solutions.
Problem
The paper examines GA usefulness in engineering pedagogy and implementation across emerging areas and applications.
Method
The paper reviews GA concepts and applications while describing population-based search through selection, crossover, mutation, fitness evaluation, and termination.
Results
A close-optimal route is identified in the GA traveling-salesman discussion, with the smallest optimum route reported as P1, P2, P3, P4, P5, P6, and P1.
Takeaways & Limitations
GAs are presented as applicable to optimization, artificial intelligence, robotics, neural networks, bioinformatics, blockchain, and other emerging technologies.
Takeaways & Limitations
The target-string GA approach has issues with optimal solutions and requires further improvement to its fitness score.
Abstract
from arXiv · showhide
Nowadays genetic algorithm (GA) is greatly used in engineering pedagogy as an adaptive technique to learn and solve complex problems and issues. It is a meta-heuristic approach that is used to solve hybrid computation challenges. GA utilizes selection, crossover, and mutation operators to effectively manage the searching system strategy. This algorithm is derived from natural selection and genetics concepts. GA is an intelligent use of random search supported with historical data to contribute the search in an area of the improved outcome within a coverage framework. Such algorithms are widely used for maintaining high-quality reactions to optimize issues and problems investigation. These techniques are recognized to be somewhat of a statistical investigation process to search for a suitable solution or prevent an accurate strategy for challenges in optimization or searches. These techniques have been produced from natural selection or genetics principles. For random testing, historical information is provided with intelligent enslavement to continue moving the search out from the area of improved features for processing of the outcomes. It is a category of heuristics of evolutionary history using behavioral science-influenced methods like an annuity, gene, preference, or combination (sometimes refers to as hybridization). This method seemed to be a valuable tool to find solutions for problems optimization. In this paper, the author has explored the GAs, its role in engineering pedagogies, and the emerging areas where it is using, and its implementation.
1 Introduction
The paper presents genetic algorithms as evolution-inspired meta-heuristics for optimization and search, highlighting their robustness and use in engineering pedagogy. It frames the paper around GA usefulness in engineering education and implementation in emerging areas.
- Genetic algorithms are meta-heuristics motivated by evolution and classified within evolutionary algorithms for informatics and computational mathematics.
- GAs address optimization and search problems lacking well-defined efficient solutions, including scheduling, shortest paths, modeling, and simulation.
- Candidate solutions are represented by characteristics called genes or phenotypes, commonly encoded as binary strings.
- The paper identifies robustness and strong performance as GA strengths, including optimistic solutions over large populations.
- The paper asks how GAs support engineering pedagogy and how they can be implemented in emerging areas.
2 Operators used in GA
GA operators evolve candidate solutions through selection, crossover, and mutation. These operators respectively favor fitter individuals, recombine genes, and introduce random genetic variation.
- After an initial generation, GAs evolve populations using genetic operators.
- Selection: Selection favors individuals with better fitness scores so they can pass genes to the succeeding generation.
- Crossover: Crossover exchanges genes at selected sites between two individuals to create a new offspring.
- Mutation: Mutation introduces random genes into offspring to preserve genetic heterogeneity and prevent excessive divergence.
3 Genetic algorithm
The GA models evolution by repeatedly evaluating and transforming a population of candidate strings through selection, crossover, and mutation until convergence. The paper notes that its target-string fitness approach requires improvement for optimal solutions.
- The algorithm begins with a random population, evaluates fitness, selects parents, generates offspring through crossover, applies mutation, and reevaluates fitness until convergence.
- Candidate strings are treated as chromosomes whose fitness is evaluated against a target string.
- Fitness equals the number of characters differing from the target at corresponding string positions, with lower values receiving preference.
- The described approach has issues with optimal solutions and needs further improvement to its fitness score.
4 Methodologies
GA methodologies generate populations, evaluate solution fitness, select individuals, and reproduce new populations through genetic operators. The process continues until a termination condition is met and the best individuals are identified.
- Initial populations are randomly generated, often containing hundreds or thousands of potential solutions, with promising areas optionally seeded.
- Individuals are selected for breeding using health-based evaluation, while stochastic mechanisms preserve population heterogeneity by retaining some less-fit approaches.
- Reproduction: Reproduction selects parents, applies crossover and mutation, and generates new populations with different chromosomes and increased average fitness.
- The population is repeatedly replaced by newly generated populations until a termination condition occurs, after which the best individual results are identified.
5 Genetic Programming
Genetic programming is presented as a major GA application that evolves populations of software programs to find solutions without programmers specifying the technique.
- 5 Genetic Programming: Genetic programming evolves populations of software programs through biologically inspired processes.The process is based on natural selection and evolutionary theory.
- 5 Genetic Programming: Its output is a virtual machine process rather than the numerical result produced by a GA.
- 5 Genetic Programming: Genetic programming can find solutions without requiring programmers to determine the solving technique.The paper contrasts this with intelligent coding through genetic modification of automated systems.
6 Role of GA in Emerging Areas
The paper identifies engineering pedagogy and several emerging technology domains where GA is applied to optimization, selection, routing, balancing, and distributed computing challenges.
- 6 Role of GA in Emerging Areas: GA is useful in engineering pedagogy for timetable generation, mathematical modeling, and optimization principles and practices.
- 6 Role of GA in Emerging Areas: In IoT, GA supports smart-device or node selection and mining intelligent objects.
- 6 Role of GA in Emerging Areas: GA is applied to intelligent traffic systems combining IoT with highway infrastructure to address journey times, emissions, and traffic.
- 6 Role of GA in Emerging Areas: For wireless networks and MANETs, GA is used for intelligent routing and dynamic optimum-path discovery.
- 6 Role of GA in Emerging Areas: GA is also discussed for cloud load balancing and blockchain-based edge computing and distributed-network processing.
7 GA Implementation in traveling salesman problem
The paper implements GA for the Traveling Salesman Problem by encoding routes as chromosomes and applying selection, crossover, and mutation to identify a close-optimal route.
- 7 GA Implementation in traveling salesman problem: Routes in the Traveling Salesman Problem are encoded as chromosomes whose genes represent visited places.The example route begins and ends at P1 while visiting P2 through P6.
- 7 GA Implementation in traveling salesman problem: The implementation generates chromosomes, evaluates fitness, calculates selection probabilities, and produces new chromosomes through crossover and mutation.The supplied tables document chromosome generation, fitness calculation, probability values, random numbers, and post-operator chromosomes.
- 7 GA Implementation in traveling salesman problem: Crossover generates children from parent chromosomes, with the crossover probability determining how many chromosomes undergo the operation.
- 7 GA Implementation in traveling salesman problem: Mutation swaps genes between chromosomes to alter routes and is intended to improve fitness.The mutation results are recorded in the new-chromosome and fitness-after-mutation tables.
- 7 GA Implementation in traveling salesman problem: The reported close-optimal route is P1, P2, P3, P4, P5, P6, and P1.The paper states that the smallest fitness level remains in place through the iterations described.
8 Applications and Area Coverage
The paper surveys GA applications across machine learning, image processing, vehicle routing, and general optimization problems with constrained objectives.
- 8 Applications and Area Coverage: In genetics-based machine learning, GA operates in discrete spaces, supports reinforcement learning techniques, and can train multi-agent structures.The paper also mentions rulesets, neural-network structures, and cellular automation machines.
- 8 Applications and Area Coverage: GA is applied to image segmentation by dividing digital images into segments based on genetic or conceptual similarities.
- 8 Applications and Area Coverage: Vehicle-routing problems involving multiple time frames, depots, and heterogeneous fleets are addressed through GA.The described setting includes deliveries from a warehouse, vehicle capacity, and travel-distance considerations.
- 8 Applications and Area Coverage: GA is widely used for optimization tasks that maximize or reduce an objective function under conditions such as fairness and inequality restrictions.The paper identifies research, marketing, and manufacturing operations as relevant contexts.
8.5 Multimodal optimization
Genetic algorithms support multimodal optimization and machine-learning applications through population-based, bio-inspired search. Their flexibility allows different local and global approaches, while parallelism can address the runtime costs of large populations.
- 8.5 Multimodal optimization: Multiple local and global approaches can be considered for difficult optimization problems, allowing the execution strategy to change with limited design disruption.The passage relates this flexibility to technical challenges involving physical and expense limitations.
- 8.5 Multimodal optimization: GAs use selection, mutation, and crossover to generate alternatives through biologically inspired evolutionary optimization.The approach is described as a meta-heuristic influenced by natural selection and used for high-quality computational solutions.
- 8.5 Multimodal optimization: GAs are used to train neural-network systems, including recursive neural networks, as population-based computational optimization techniques.The method is associated with machine learning and neural-network training.
- 8.5 Multimodal optimization: Parallel computing can theoretically reduce the large runtime associated with maintaining many individuals in the GA population.The computational burden is linked to the population size.
8.9 Scheduling applications
The paper describes GA applications across scheduling, robotic trajectory planning, aircraft design, and bioinformatics. These applications use optimization to balance competing objectives, construct feasible paths, evaluate designs, or analyze genetic information.
- 8.9 Scheduling applications: Scheduling problems balance contradictory priorities involving individual costs, operator costs, consumer service, and provider income.The passage contrasts consumer preferences for shorter waits and less crowding with provider concerns about operating costs and revenue.
- 8.9 Scheduling applications: GAs can design collision-free robotic-arm paths between specified starting and target positions using coding and fitness analysis.The passage identifies fitness evaluation as a main feature of the trajectory-design algorithm.
- 8.9 Scheduling applications: Aircraft design with GAs evaluates competing measures including value, control surfaces, production, fuel consumption, disturbance, and aviation performance.The final design is presented as a compromise because cheaper, strongest, most effective, and most convenient aircraft may differ.
- 8.9 Scheduling applications: GA-based DNA evaluation uses specimen spectrometric information in a bioinformatics context shaped by rapidly increasing genetic data.The passage cautions that limited awareness of genetic characteristics can make hypotheses introduced during construction unreliable or redundant.
9 Conclusion
The conclusion presents GA as a probabilistic optimization method modeled on biological genetic evaluation. It positions the algorithm as applicable to global optimization and diverse artificial-intelligence domains.
- 9 Conclusion: GA is presented as a probabilistic method for finding global optimum solutions across many problem types.Its search process is modeled on biological genetic evaluation.
- 9 Conclusion: The algorithm is used in artificial-intelligence applications including object-oriented systems, robotics, and emerging technologies.These applications are listed as examples of GA’s broad use.