Pub Date : 2021-01-01DOI: 10.4114/intartif.vol24iss68pp89-103
João Batista Pacheco Junior, Henrique Mariano Costa do Amaral
The design and manual insertion of new terrestrial roads into geographic databases is a frequent activity in geoprocessing and their demand usually occurs as the most up-to-date satellite imagery of the territory is acquired. Continually, new urban and rural occupations emerge, for which specific vector geometries need to be designed to characterize the cartographic inputs and accommodate the relevant associated data. Therefore, it is convenient to develop a computational tool that, with the help of artificial intelligence, automates what is possible in this respect, since manual editing depends on the limits of user agility, and does it in images that are usually easy and free to access. To test the feasibility of this proposal, a database of RGB images containing asphalted urban roads is presented to the K-Means++ algorithm and the SegNet Convolutional Neural Network, and the performance of each one was evaluated and compared for accuracy and IoU of road identification. Under the conditions of the experiment, K-Means++ achieved poor and unviable results for use in a real-life application involving asphalt road detection in RGB satellite images, with average accuracy ranging from 41.67% to 64.19% and average IoU of 12.30% to 16.16%, depending on the preprocessing strategy used. On the other hand, the SegNet Convolutional Neural Network proved to be appropriate for precision applications not sensitive to discontinuities, achieving an average accuracy of 87.12% and an average IoU of 71.93%.
{"title":"Performance Analysis in the Segmentation of urban asphalted roads in RGB satellite images using K-Means++ and SegNet: Case study in São Luís-MA","authors":"João Batista Pacheco Junior, Henrique Mariano Costa do Amaral","doi":"10.4114/intartif.vol24iss68pp89-103","DOIUrl":"https://doi.org/10.4114/intartif.vol24iss68pp89-103","url":null,"abstract":"The design and manual insertion of new terrestrial roads into geographic databases is a frequent activity in geoprocessing and their demand usually occurs as the most up-to-date satellite imagery of the territory is acquired. Continually, new urban and rural occupations emerge, for which specific vector geometries need to be designed to characterize the cartographic inputs and accommodate the relevant associated data. Therefore, it is convenient to develop a computational tool that, with the help of artificial intelligence, automates what is possible in this respect, since manual editing depends on the limits of user agility, and does it in images that are usually easy and free to access. To test the feasibility of this proposal, a database of RGB images containing asphalted urban roads is presented to the K-Means++ algorithm and the SegNet Convolutional Neural Network, and the performance of each one was evaluated and compared for accuracy and IoU of road identification. Under the conditions of the experiment, K-Means++ achieved poor and unviable results for use in a real-life application involving asphalt road detection in RGB satellite images, with average accuracy ranging from 41.67% to 64.19% and average IoU of 12.30% to 16.16%, depending on the preprocessing strategy used. On the other hand, the SegNet Convolutional Neural Network proved to be appropriate for precision applications not sensitive to discontinuities, achieving an average accuracy of 87.12% and an average IoU of 71.93%.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"24 1","pages":"89-103"},"PeriodicalIF":2.3,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"70874847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-08-27DOI: 10.4114/INTARTIF.VOL23ISS66PP9-25
Adriana Villa-Murillo, A. Carrión, A. Sozzi
We propose a methodology for the improvement of the parameter design that consists of the combination ofRandom Forest (RF) with Genetic Algorithms (GA) in 3 phases: normalization, modelling and optimization.The rst phase corresponds to the previous preparation of the data set by using normalization functions. In thesecond phase, we designed a modelling scheme adjusted to multiple quality characteristics and we have called itMultivariate Random Forest (MRF) for the determination of the objective function. Finally, in the third phase,we obtained the optimal combination of parameter levels with the integration of properties of our modellingscheme and desirability functions in the establishment of the corresponding GA. Two illustrative cases allow us tocompare and validate the virtues of our methodology versus other proposals involving Arti cial Neural Networks(ANN) and Simulated Annealing (SA).
{"title":"Forest-Genetic method to optimize parameter design of multiresponse experiment","authors":"Adriana Villa-Murillo, A. Carrión, A. Sozzi","doi":"10.4114/INTARTIF.VOL23ISS66PP9-25","DOIUrl":"https://doi.org/10.4114/INTARTIF.VOL23ISS66PP9-25","url":null,"abstract":"We propose a methodology for the improvement of the parameter design that consists of the combination ofRandom Forest (RF) with Genetic Algorithms (GA) in 3 phases: normalization, modelling and optimization.The rst phase corresponds to the previous preparation of the data set by using normalization functions. In thesecond phase, we designed a modelling scheme adjusted to multiple quality characteristics and we have called itMultivariate Random Forest (MRF) for the determination of the objective function. Finally, in the third phase,we obtained the optimal combination of parameter levels with the integration of properties of our modellingscheme and desirability functions in the establishment of the corresponding GA. Two illustrative cases allow us tocompare and validate the virtues of our methodology versus other proposals involving Arti cial Neural Networks(ANN) and Simulated Annealing (SA).","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"49 1","pages":"9-25"},"PeriodicalIF":2.3,"publicationDate":"2020-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85091180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-01-01DOI: 10.4114/INTARTIF.VOL23ISS66PP1-8
Yun Shi, Yanyan Zhu
Considering the need for a large number of samples and the long training time, this paper uses deep and transfer learning to identify motion-blurred Chinese character coded targets (CCTs). Firstly, a set of CCTs are designed, and the motion blur image generation system is used to provide samples for the recognition network. Secondly, the OTSU algorithm, the expansion, and the Canny operator are performed on the real shot blurred image, where the target area is segmented by the minimum bounding box. Thirdly, the sample is selected from the sample set according to the 4:1 ratio as the training set and the test set. Under the Tensor Flow framework, the convolutional layer in the AlexNet model is fixed, and the fully-connected layer is trained for transfer learning. Finally, experiments on simulated and real-time motion-blurred images are carried out. The results show that network training and testing only take 30 minutes and two seconds, and the recognition accuracy reaches 98.6% and 93.58%, respectively. As a result, our method has higher recognition accuracy, does not require a large number of trained samples, takes less time, and can provide a certain reference for the recognition of motion-blurred CCTs.
{"title":"Recognition of Motion-blurred CCTs based on Deep and Transfer Learning","authors":"Yun Shi, Yanyan Zhu","doi":"10.4114/INTARTIF.VOL23ISS66PP1-8","DOIUrl":"https://doi.org/10.4114/INTARTIF.VOL23ISS66PP1-8","url":null,"abstract":"Considering the need for a large number of samples and the long training time, this paper uses deep and transfer learning to identify motion-blurred Chinese character coded targets (CCTs). Firstly, a set of CCTs are designed, and the motion blur image generation system is used to provide samples for the recognition network. Secondly, the OTSU algorithm, the expansion, and the Canny operator are performed on the real shot blurred image, where the target area is segmented by the minimum bounding box. Thirdly, the sample is selected from the sample set according to the 4:1 ratio as the training set and the test set. Under the Tensor Flow framework, the convolutional layer in the AlexNet model is fixed, and the fully-connected layer is trained for transfer learning. Finally, experiments on simulated and real-time motion-blurred images are carried out. The results show that network training and testing only take 30 minutes and two seconds, and the recognition accuracy reaches 98.6% and 93.58%, respectively. As a result, our method has higher recognition accuracy, does not require a large number of trained samples, takes less time, and can provide a certain reference for the recognition of motion-blurred CCTs.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"27 1","pages":"1-8"},"PeriodicalIF":2.3,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83266562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.4114/intartif.vol22iss62pp114-133
Paulo Vitor de Campos Souza
This paper presents a novel learning algorithm for fuzzy logic neuron based on neural networks and fuzzy systems able to generate accurate and transparent models. The learning algorithm is based on ideas from Extreme Learning Machine [36], to achieve a low time complexity, and regularization theory, resulting in sparse and accurate models. A compact set of incomplete fuzzy rules can be extracted from the resulting network topology. Experiments considering regression problems are detailed. Results suggest the proposed approach as a promising alternative for pattern recognition with a good accuracy and some level of interpretability.
{"title":"Fuzzy Neural Networks based on Fuzzy Logic Neurons Regularized by Resampling Techniques and Regularization Theory for Regression Problems","authors":"Paulo Vitor de Campos Souza","doi":"10.4114/intartif.vol22iss62pp114-133","DOIUrl":"https://doi.org/10.4114/intartif.vol22iss62pp114-133","url":null,"abstract":"This paper presents a novel learning algorithm for fuzzy logic neuron based on neural networks and fuzzy systems able to generate accurate and transparent models. The learning algorithm is based on ideas from Extreme Learning Machine [36], to achieve a low time complexity, and regularization theory, resulting in sparse and accurate models. A compact set of incomplete fuzzy rules can be extracted from the resulting network topology. Experiments considering regression problems are detailed. Results suggest the proposed approach as a promising alternative for pattern recognition with a good accuracy and some level of interpretability.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"1 1","pages":""},"PeriodicalIF":2.3,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47989804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Juan Murillo-Morera, Carlos Castro-Herrera, J. Arroyo, Rubén Fuentes-Fernández
Today, it is common for software projects to collect measurement data through development processes. With these data, defect prediction software can try to estimate the defect proneness of a software module, with the objective of assisting and guiding software practitioners. With timely and accurate defect predictions, practitioners can focus their limited testing resources on higher risk areas. This paper reports the results of three empirical studies that uses an automated genetic defect prediction framework. This framework generates and compares different learning schemes (preprocessing + attribute selection + learning algorithms) and selects the best one using a genetic algorithm, with the objective to estimate the defect proneness of a software module. The first empirical study is a performance comparison of our framework with the most important framework of the literature. The second empirical study is a performance and runtime comparison between our framework and an exhaustive framework. The third empirical study is a sensitivity analysis. The last empirical study, is our main contribution in this paper. Performance of the software development defect prediction models (using AUC, Area Under the Curve) was validated using NASA-MDP and PROMISE data sets. Seventeen data sets from NASA-MDP (13) and PROMISE (4) projects were analyzed running a NxM-fold cross-validation. A genetic algorithm was used to select the components of the learning schemes automatically, and to assess and report the results. Our results reported similar performance between frameworks. Our framework reported better runtime than exhaustive framework. Finally, we reported the best configuration according to sensitivity analysis.
{"title":"An Automated Defect Prediction Framework using Genetic Algorithms: A Validation of Empirical Studies","authors":"Juan Murillo-Morera, Carlos Castro-Herrera, J. Arroyo, Rubén Fuentes-Fernández","doi":"10.4114/ia.v18i56.1159","DOIUrl":"https://doi.org/10.4114/ia.v18i56.1159","url":null,"abstract":"Today, it is common for software projects to collect measurement data through development processes. With these data, defect prediction software can try to estimate the defect proneness of a software module, with the objective of assisting and guiding software practitioners. With timely and accurate defect predictions, practitioners can focus their limited testing resources on higher risk areas. This paper reports the results of three empirical studies that uses an automated genetic defect prediction framework. This framework generates and compares different learning schemes (preprocessing + attribute selection + learning algorithms) and selects the best one using a genetic algorithm, with the objective to estimate the defect proneness of a software module. The first empirical study is a performance comparison of our framework with the most important framework of the literature. The second empirical study is a performance and runtime comparison between our framework and an exhaustive framework. The third empirical study is a sensitivity analysis. The last empirical study, is our main contribution in this paper. Performance of the software development defect prediction models (using AUC, Area Under the Curve) was validated using NASA-MDP and PROMISE data sets. Seventeen data sets from NASA-MDP (13) and PROMISE (4) projects were analyzed running a NxM-fold cross-validation. A genetic algorithm was used to select the components of the learning schemes automatically, and to assess and report the results. Our results reported similar performance between frameworks. Our framework reported better runtime than exhaustive framework. Finally, we reported the best configuration according to sensitivity analysis.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"24 1","pages":"114-137"},"PeriodicalIF":2.3,"publicationDate":"2016-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79135622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Reinstatement is a principle of argumentation systems that enables the justification of a defeated argument when all its defeaters are in turn ultimately def...
复辟是论证系统的一个原则，它使一个被击败的论点能够被证明是正确的，当它的所有反对者最终都被辩护。
{"title":"Restablecimiento y especificidad en sistemas argumentativos","authors":"C. Alessio","doi":"10.4114/IA.V19I58.1129","DOIUrl":"https://doi.org/10.4114/IA.V19I58.1129","url":null,"abstract":"Reinstatement is a principle of argumentation systems that enables the justification of a defeated argument when all its defeaters are in turn ultimately def...","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"17 1","pages":"39-44"},"PeriodicalIF":2.3,"publicationDate":"2016-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73228494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recent years have witnessed great interest in developing methods for content-based image retrieval (CBIR). Generally, the image search results which are returned by an image search engine contain multiple topics, and organizing the results into different clusters will facilitate users’ browsing. Our aim in this research is to optimize image searching time for a general image database. The proposed procedure consists of two steps. First, it represents each image with a data structure which is based on quadtrees and represented by multi-level feature vectors. The similarity between images is evaluated through the distance between their feature vectors; this distance metric reduces the query processing time. Second, response time is further improved by using a secondary clustering technique to achieve high scalability in the case of a very large image database.
{"title":"Improving Image Retrieval using a Data mining Approach","authors":"Houaria Abed, L. Zaoui","doi":"10.4114/IA.V18I56.1147","DOIUrl":"https://doi.org/10.4114/IA.V18I56.1147","url":null,"abstract":"Recent years have witnessed great interest in developing methods for content-based image retrieval (CBIR). Generally, the image search results which are returned by an image search engine contain multiple topics, and organizing the results into different clusters will facilitate users’ browsing. Our aim in this research is to optimize image searching time for a general image database. The proposed procedure consists of two steps. First, it represents each image with a data structure which is based on quadtrees and represented by multi-level feature vectors. The similarity between images is evaluated through the distance between their feature vectors; this distance metric reduces the query processing time. Second, response time is further improved by using a secondary clustering technique to achieve high scalability in the case of a very large image database.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"14 1","pages":"97-113"},"PeriodicalIF":2.3,"publicationDate":"2015-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90106087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Damny Magdaleno, Yadriel Miranda, Ivett Fuentes, M. M. García
A huge amount of information is represented in XML format. Several tools have been developed to store, and query XML data. It becomes inevitable to develop high performance techniques for efficiently analysing extremely large collections of XML data. One of the methods that many researchers have focused on is clustering, which groups similar XML data, according to their content and structures. In previous work, there has been proposed the similarity function OverallSimSUX, that facilitates to capture the degree of similitude among the documents with a novel methodology for clustering XML documents using both structural and content features. Although this methodology shows good performance, endorsed by experiments with several corpus and statistical tests, on having had impliedly only one clustering algorithm, K-Star, we do not know the effect that it would suffer if we replaced this algorithm by other with dissimilar characteristics. Therefore to endorse completely the methodology, in this work we make a comparative study of the effects of applying the methodology for the OverallSimSUX similarity function calculation, using clustering algorithms of different classifications . Based on our analysis, we arrived to two important results: (1) The Fuzzy-SKWIC clustering algorithm works best both with methodology and without methodology, although there are not present significant differences respect to the K-Star clustering algorithm; (2) For each analysed algorithm when using the methodology, we obtain better results than when it is not taken into account.
{"title":"Comparative Study of Clustering Algorithms using OverallSimSUX Similarity Function for XML Documents","authors":"Damny Magdaleno, Yadriel Miranda, Ivett Fuentes, M. M. García","doi":"10.4114/IA.V18I55.1097","DOIUrl":"https://doi.org/10.4114/IA.V18I55.1097","url":null,"abstract":"A huge amount of information is represented in XML format. Several tools have been developed to store, and query XML data. It becomes inevitable to develop high performance techniques for efficiently analysing extremely large collections of XML data. One of the methods that many researchers have focused on is clustering, which groups similar XML data, according to their content and structures. In previous work, there has been proposed the similarity function OverallSimSUX, that facilitates to capture the degree of similitude among the documents with a novel methodology for clustering XML documents using both structural and content features. Although this methodology shows good performance, endorsed by experiments with several corpus and statistical tests, on having had impliedly only one clustering algorithm, K-Star, we do not know the effect that it would suffer if we replaced this algorithm by other with dissimilar characteristics. Therefore to endorse completely the methodology, in this work we make a comparative study of the effects of applying the methodology for the OverallSimSUX similarity function calculation, using clustering algorithms of different classifications . Based on our analysis, we arrived to two important results: (1) The Fuzzy-SKWIC clustering algorithm works best both with methodology and without methodology, although there are not present significant differences respect to the K-Star clustering algorithm; (2) For each analysed algorithm when using the methodology, we obtain better results than when it is not taken into account.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"44 1","pages":"69-80"},"PeriodicalIF":2.3,"publicationDate":"2015-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86882915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
La explotación hotelera, sin dejar de satisfacer a los clientes, necesita disminuir los requerimientos de energía eléctrica como principal portador energético. Resolver la problemática de la ocupación de un hotel de manera integral, tomando como centro de atención la climatización, la cual provoca los mayores consumos de electricidad, resulta una tarea compleja. Para resolver el problema se implementó un procedimiento para la optimización energética de la operación de los sistemas de climatización centralizados todo-agua. El procedimiento integra, un modelo energético con una estrategia de ocupación bajo criterios energéticos y de fundamento combinatorio-evolutivo. Para la clasificación de la información, la formulación de las tareas y la síntesis de las soluciones, se emplea la metodología de Análisis y Síntesis de Sistemas de Ingeniería. El modelo energético considera la variabilidad de la climatología local y la ocupación de las habitaciones seleccionadas, e incluye: el modelo térmico de la edificación obtenido mediante redes neuronales artificiales, el modelo hidráulico y el modelo del trabajo de compresión. Estos elementos permiten la búsqueda de la variable de decisión ocupación, realizando cálculos intermedios de la velocidad de rotación en la bomba centrífuga y la temperatura de salida del agua del enfriador, minimizando los requerimientos de potencia eléctrica en la climatización centralizada. Para evaluar los estados del sistema se utiliza una optimización combinatoria que emplea los métodos: exhaustivo simple, exhaustivo escalonado o algoritmo genético según la cantidad de variantes de ocupación. Todas las tareas de cálculo y algoritmos del procedimiento se automatizaron mediante una aplicación informática.
{"title":"Optimización energética de la operación de los sistemas de climatización por agua helada en hoteles","authors":"R. M. Laurencio","doi":"10.4114/ia.v18i56.1125","DOIUrl":"https://doi.org/10.4114/ia.v18i56.1125","url":null,"abstract":"La explotación hotelera, sin dejar de satisfacer a los clientes, necesita disminuir los requerimientos de energía eléctrica como principal portador energético. Resolver la problemática de la ocupación de un hotel de manera integral, tomando como centro de atención la climatización, la cual provoca los mayores consumos de electricidad, resulta una tarea compleja. Para resolver el problema se implementó un procedimiento para la optimización energética de la operación de los sistemas de climatización centralizados todo-agua. El procedimiento integra, un modelo energético con una estrategia de ocupación bajo criterios energéticos y de fundamento combinatorio-evolutivo. Para la clasificación de la información, la formulación de las tareas y la síntesis de las soluciones, se emplea la metodología de Análisis y Síntesis de Sistemas de Ingeniería. El modelo energético considera la variabilidad de la climatología local y la ocupación de las habitaciones seleccionadas, e incluye: el modelo térmico de la edificación obtenido mediante redes neuronales artificiales, el modelo hidráulico y el modelo del trabajo de compresión. Estos elementos permiten la búsqueda de la variable de decisión ocupación, realizando cálculos intermedios de la velocidad de rotación en la bomba centrífuga y la temperatura de salida del agua del enfriador, minimizando los requerimientos de potencia eléctrica en la climatización centralizada. Para evaluar los estados del sistema se utiliza una optimización combinatoria que emplea los métodos: exhaustivo simple, exhaustivo escalonado o algoritmo genético según la cantidad de variantes de ocupación. Todas las tareas de cálculo y algoritmos del procedimiento se automatizaron mediante una aplicación informática.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"1 1","pages":"1-4"},"PeriodicalIF":2.3,"publicationDate":"2015-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76512305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Carlos Bermúdez, P. Graglia, Natalia Stark, C. Salto, Hugo Alfonso
The Vehicle Routing Problem (VRP) deals with the assignment of a set of transportation orders to a fleet of vehicles and the sequencing of stops for each vehicle to minimize transportation costs. This paper presents two different genetic algorithm models (panmitic and cellular models) for providing solutions for the Capacitated VRP (CVRP), which is mainly characterized by using vehicles of the same capacity. We propose a new problem dependent recombination operator, called Best Route Better Adjustment recombination (BRBAX), which incorporates problem specific knowledge such as information about the routes constitution. A comparison of its performance is carried out with respect to classical recombination operators for permutations. A complete study of the influence of the recombination operators on the genetic search is presented. The results show that the use of our specialized BRBAX operator outperforms the others more generic operators for all problem instances under all metrics. The deviation between our best solution and the best-known one is very low, under 0.91%.
{"title":"Comparison of Recombination Operators in Panmictic and Cellular GAs to Solve a Vehicle Routing Problem","authors":"Carlos Bermúdez, P. Graglia, Natalia Stark, C. Salto, Hugo Alfonso","doi":"10.4114/IA.V14I46.1520","DOIUrl":"https://doi.org/10.4114/IA.V14I46.1520","url":null,"abstract":"The Vehicle Routing Problem (VRP) deals with the assignment of a set of transportation orders to a fleet of vehicles and the sequencing of stops for each vehicle to minimize transportation costs. This paper presents two different genetic algorithm models (panmitic and cellular models) for providing solutions for the Capacitated VRP (CVRP), which is mainly characterized by using vehicles of the same capacity. We propose a new problem dependent recombination operator, called Best Route Better Adjustment recombination (BRBAX), which incorporates problem specific knowledge such as information about the routes constitution. A comparison of its performance is carried out with respect to classical recombination operators for permutations. A complete study of the influence of the recombination operators on the genetic search is presented. The results show that the use of our specialized BRBAX operator outperforms the others more generic operators for all problem instances under all metrics. The deviation between our best solution and the best-known one is very low, under 0.91%.","PeriodicalId":43470,"journal":{"name":"Inteligencia Artificial-Iberoamerical Journal of Artificial Intelligence","volume":"11 1","pages":"34-44"},"PeriodicalIF":2.3,"publicationDate":"2010-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73593251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：