Pub Date : 2016-12-01DOI: 10.1109/SSCI.2016.7850204
Fergal Lane, R. Muhammad, Atif Azad, C. Ryan, Ireland Email, Fergal Lane, Ie
Configuring an Evolutionary Algorithm (EA) can be a haphazard and inefficient process. An EA practitioner may have to choose between a plethora of search operator types and other parameter settings. In contrast, the goal of EA principled design is a more streamlined and systematic design methodology, which first seeks to better understand the problem domain, and only then uses such acquired insights to guide the choice of parameters and operators.
{"title":"Principled Evolutionary Algorithm search operator design and the kernel trick","authors":"Fergal Lane, R. Muhammad, Atif Azad, C. Ryan, Ireland Email, Fergal Lane, Ie","doi":"10.1109/SSCI.2016.7850204","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7850204","url":null,"abstract":"Configuring an Evolutionary Algorithm (EA) can be a haphazard and inefficient process. An EA practitioner may have to choose between a plethora of search operator types and other parameter settings. In contrast, the goal of EA principled design is a more streamlined and systematic design methodology, which first seeks to better understand the problem domain, and only then uses such acquired insights to guide the choice of parameters and operators.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129106461","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7849997
Bruno Cunha, A. Madureira, J. Pereira, I. Pereira
The ability to adjust itself to users' profile is imperative in modern system, given that many people interact with a lot of information in different ways. The creation of adaptive systems is a complex domain that requires very specific methods and the integration of several intelligent techniques, from an intelligent systems development perspective. Designing an adaptive system requires planning and training of user modelling techniques combined with existing system components. Based on the architecture for user modelling on Intelligent and Adaptive Scheduling Systems, this paper presents an analysis of using the mentioned architecture to characterize user's behaviours and a case study comparing the employment of different user classifiers. Bayesian and Artificial Neural Networks were selected as the elements of the computational study and this paper presents a description on how to prepare them to deal with user information.
{"title":"Evaluating the effectiveness of Bayesian and Neural Networks for Adaptive Schedulling Systems","authors":"Bruno Cunha, A. Madureira, J. Pereira, I. Pereira","doi":"10.1109/SSCI.2016.7849997","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7849997","url":null,"abstract":"The ability to adjust itself to users' profile is imperative in modern system, given that many people interact with a lot of information in different ways. The creation of adaptive systems is a complex domain that requires very specific methods and the integration of several intelligent techniques, from an intelligent systems development perspective. Designing an adaptive system requires planning and training of user modelling techniques combined with existing system components. Based on the architecture for user modelling on Intelligent and Adaptive Scheduling Systems, this paper presents an analysis of using the mentioned architecture to characterize user's behaviours and a case study comparing the employment of different user classifiers. Bayesian and Artificial Neural Networks were selected as the elements of the computational study and this paper presents a description on how to prepare them to deal with user information.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131035918","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7849973
Jiangnan Qiu, Wenjing Gu, Q. Kong, Qiuyan Zhong, Jilei Hu
In order to solve the emergency decision management problem with uncertainty, an Emergency Bayesian decision network (EBDN) model is used in this paper. By computing the probability of each node, the EBDN can solve the uncertainty of different response measures. Using Gray system theory to determine the weight of all kinds of emergency losses. And then use genetic algorithm to search the best combination measure by comparing the value of output loss. For illustration, a typhoon example is utilized to show the feasibility of EBDN model. Empirical results show that the EBDN model can combine expert's knowledge and historic data to predict expected effects under different combinations of response measures, and then choose the best one. The proposed EBDN model can combine the decision process into a diagrammatic form, and thus the uncertainty of emergency events in solving emergency dynamic decision making is solved.
{"title":"The emergency response management based on Bayesian decision network","authors":"Jiangnan Qiu, Wenjing Gu, Q. Kong, Qiuyan Zhong, Jilei Hu","doi":"10.1109/SSCI.2016.7849973","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7849973","url":null,"abstract":"In order to solve the emergency decision management problem with uncertainty, an Emergency Bayesian decision network (EBDN) model is used in this paper. By computing the probability of each node, the EBDN can solve the uncertainty of different response measures. Using Gray system theory to determine the weight of all kinds of emergency losses. And then use genetic algorithm to search the best combination measure by comparing the value of output loss. For illustration, a typhoon example is utilized to show the feasibility of EBDN model. Empirical results show that the EBDN model can combine expert's knowledge and historic data to predict expected effects under different combinations of response measures, and then choose the best one. The proposed EBDN model can combine the decision process into a diagrammatic form, and thus the uncertainty of emergency events in solving emergency dynamic decision making is solved.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130537903","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7849965
Jayne Eaton, Shengxiang Yang
Train delays at stations are a common occurrence in complex, busy railway networks. A delayed train will miss its scheduled time slot on the platform and may have to be reallocated to a new platform to allow it to continue its journey. The problem is a dynamic one because while reallocating a delayed train further unanticipated train delays may occur, changing the nature of the problem over time. Our aim in this study is to apply ant colony optimisation (ACO) to a dynamic platform reallocation problem (DPRP) using a model created from real-world train schedule data. To ensure that trains are not unnecessarily reallocated to new platforms we introduce a novel best-ant-replacement scheme that takes into account not only the objective value but also the physical distance between the original and the new platforms. Results showed that the ACO algorithm outperformed a heuristic that places the delayed train in the first available time-slot and that this improvement was more apparent with high-frequency dynamic changes.
{"title":"Railway platform reallocation after dynamic perturbations using ant colony optimisation","authors":"Jayne Eaton, Shengxiang Yang","doi":"10.1109/SSCI.2016.7849965","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7849965","url":null,"abstract":"Train delays at stations are a common occurrence in complex, busy railway networks. A delayed train will miss its scheduled time slot on the platform and may have to be reallocated to a new platform to allow it to continue its journey. The problem is a dynamic one because while reallocating a delayed train further unanticipated train delays may occur, changing the nature of the problem over time. Our aim in this study is to apply ant colony optimisation (ACO) to a dynamic platform reallocation problem (DPRP) using a model created from real-world train schedule data. To ensure that trains are not unnecessarily reallocated to new platforms we introduce a novel best-ant-replacement scheme that takes into account not only the objective value but also the physical distance between the original and the new platforms. Results showed that the ACO algorithm outperformed a heuristic that places the delayed train in the first available time-slot and that this improvement was more apparent with high-frequency dynamic changes.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130825586","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7849975
U. Yolcu, Yaochu Jin, E. Eğrioğlu
Inference systems basically aim to provide and present the knowledge (outputs) that decision-makers can take advantage of in their decision-making process. Nowadays one of the most commonly used inference systems for time series prediction is the computational inference system based on artificial neural networks. Although they have the ability of handling uncertainties and are capable of solving real life problems, neural networks have interpretability issues with regard to their outputs. For example, the outputs of neural networks that are difficult to interpret compared to statistical inference systems' outputs that involve a confidence interval and probabilities about possible values of predictions on top of the point estimations. In this study, an ensemble of single multiplicative neuron models based on bootstrap technique has been proposed to get probabilistic predictions. The main difference of the proposed ensemble model compared to conventional neural network models is that it is capable of getting a bootstrap confidence interval and probabilities of predictions. The performance of the proposed model is demonstrated on different time series. The obtained results show that the proposed ensemble model has a superior prediction performance in addition to having outputs that are more interpretable and applicable to probabilistic evaluations than conventional neural networks.
{"title":"An ensemble of single multiplicative neuron models for probabilistic prediction","authors":"U. Yolcu, Yaochu Jin, E. Eğrioğlu","doi":"10.1109/SSCI.2016.7849975","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7849975","url":null,"abstract":"Inference systems basically aim to provide and present the knowledge (outputs) that decision-makers can take advantage of in their decision-making process. Nowadays one of the most commonly used inference systems for time series prediction is the computational inference system based on artificial neural networks. Although they have the ability of handling uncertainties and are capable of solving real life problems, neural networks have interpretability issues with regard to their outputs. For example, the outputs of neural networks that are difficult to interpret compared to statistical inference systems' outputs that involve a confidence interval and probabilities about possible values of predictions on top of the point estimations. In this study, an ensemble of single multiplicative neuron models based on bootstrap technique has been proposed to get probabilistic predictions. The main difference of the proposed ensemble model compared to conventional neural network models is that it is capable of getting a bootstrap confidence interval and probabilities of predictions. The performance of the proposed model is demonstrated on different time series. The obtained results show that the proposed ensemble model has a superior prediction performance in addition to having outputs that are more interpretable and applicable to probabilistic evaluations than conventional neural networks.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130443887","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7850150
M. Pérez-Ortiz, Pedro Antonio Gutiérrez, Mariano Carbonero-Ruz, C. Hervás‐Martínez
The recently coined term “learning from label proportions” refers to a new learning paradigm where training data is given by groups (also denoted as “bags”), and the only known information is the label proportion of each bag. The aim is then to construct a classification model to predict the class label of an individual instance, which differentiates this paradigm from the one of multi-instance learning. This learning setting presents very different applications in political science, marketing, healthcare and, in general, all fields in relation with anonymous data. In this paper, two new strategies are proposed to tackle this kind of problems. Both proposals are based on the optimisation of pattern class memberships using the data distribution in each bag and the known label proportions. To do so, linear discriminant analysis has been reformulated to work with non-crisp class memberships. The experimental part of this paper sets different objetives: 1) study the difference in performance, comparing our proposals and the fully supervised setting, 2) analyse the potential benefits of refining class memberships by the proposed approaches, and 3) test the influence of other factors in the performance, such as the number of classes or the bag size. The results of these experiments are promising, but further research should be encouraged for studying more complex data configurations.
{"title":"Adapting linear discriminant analysis to the paradigm of learning from label proportions","authors":"M. Pérez-Ortiz, Pedro Antonio Gutiérrez, Mariano Carbonero-Ruz, C. Hervás‐Martínez","doi":"10.1109/SSCI.2016.7850150","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7850150","url":null,"abstract":"The recently coined term “learning from label proportions” refers to a new learning paradigm where training data is given by groups (also denoted as “bags”), and the only known information is the label proportion of each bag. The aim is then to construct a classification model to predict the class label of an individual instance, which differentiates this paradigm from the one of multi-instance learning. This learning setting presents very different applications in political science, marketing, healthcare and, in general, all fields in relation with anonymous data. In this paper, two new strategies are proposed to tackle this kind of problems. Both proposals are based on the optimisation of pattern class memberships using the data distribution in each bag and the known label proportions. To do so, linear discriminant analysis has been reformulated to work with non-crisp class memberships. The experimental part of this paper sets different objetives: 1) study the difference in performance, comparing our proposals and the fully supervised setting, 2) analyse the potential benefits of refining class memberships by the proposed approaches, and 3) test the influence of other factors in the performance, such as the number of classes or the bag size. The results of these experiments are promising, but further research should be encouraged for studying more complex data configurations.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129288545","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7849957
Sinem Sav, David J. D. Hampson, Herbert H. Tsang
Most of the biological processes including expression levels of genes and translation of DNA to produce proteins within cells depend on RNA sequences, and the structure of the RNA plays vital role for its function. RNA design problem refers to the design of an RNA sequence that folds into given secondary structure. However, vast number of possible nucleotide combinations make this an NP-Hard problem. To solve the RNA design problem, a number of researchers have tried to implement algorithms using local stochastic search, context-free grammars, global sampling or evolutionary programming approaches. In this paper, we examine SIMARD, an RNA design algorithm that implements simulated annealing techniques. We also propose QPS, a mutation operator for SIMARD that pre-selects high quality sequences. Furthermore, we present experiment results of SIMARD compared to eight other RNA design algorithms using the Rfam datset. The experiment results indicate that SIMARD shows promising results in terms of Hamming distance between designed sequence and the target structure, and outperforms ERD in terms of free energy.
{"title":"SIMARD: A simulated annealing based RNA design algorithm with quality pre-selection strategies","authors":"Sinem Sav, David J. D. Hampson, Herbert H. Tsang","doi":"10.1109/SSCI.2016.7849957","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7849957","url":null,"abstract":"Most of the biological processes including expression levels of genes and translation of DNA to produce proteins within cells depend on RNA sequences, and the structure of the RNA plays vital role for its function. RNA design problem refers to the design of an RNA sequence that folds into given secondary structure. However, vast number of possible nucleotide combinations make this an NP-Hard problem. To solve the RNA design problem, a number of researchers have tried to implement algorithms using local stochastic search, context-free grammars, global sampling or evolutionary programming approaches. In this paper, we examine SIMARD, an RNA design algorithm that implements simulated annealing techniques. We also propose QPS, a mutation operator for SIMARD that pre-selects high quality sequences. Furthermore, we present experiment results of SIMARD compared to eight other RNA design algorithms using the Rfam datset. The experiment results indicate that SIMARD shows promising results in terms of Hamming distance between designed sequence and the target structure, and outperforms ERD in terms of free energy.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128823655","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7849939
D. Koutsoukos, Georgios Alexandridis, Georgios Siolas, A. Stafylopatis
In this paper a new algorithm for session identification in web logs is outlined, based on the fuzzy c-means clustering of the available data. The novelty of the proposed methodology lies in the initialization of the partition matrix using subtractive clustering, the examination of the effect a variety of distance metrics have on the clustering process (in addition to the widely-used Euclidean distance), the determination of the number of user sessions based on candidate sessions and the representation of the session data. The experimental results show that the proposed methodology is effective in the reconstruction of user sessions and can distinguish individual sessions more accurately than baseline time-heuristic methods proposed in literature.
{"title":"A new approach to session identification by applying fuzzy c-means clustering on web logs","authors":"D. Koutsoukos, Georgios Alexandridis, Georgios Siolas, A. Stafylopatis","doi":"10.1109/SSCI.2016.7849939","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7849939","url":null,"abstract":"In this paper a new algorithm for session identification in web logs is outlined, based on the fuzzy c-means clustering of the available data. The novelty of the proposed methodology lies in the initialization of the partition matrix using subtractive clustering, the examination of the effect a variety of distance metrics have on the clustering process (in addition to the widely-used Euclidean distance), the determination of the number of user sessions based on candidate sessions and the representation of the session data. The experimental results show that the proposed methodology is effective in the reconstruction of user sessions and can distinguish individual sessions more accurately than baseline time-heuristic methods proposed in literature.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126219398","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7850103
M. Vasile, Lorenzo A. Ricciardi
This paper proposes a memetic direct transcription algorithm to solve Multi-Objective Optimal Control Problems (MOOCP). The MOOCP is first transcribed into a Non-linear Programming Problem (NLP) with Direct Finite Elements in Time (DFET) and then solved with a particular formulation of the Multi Agent Collaborative Search (MACS) framework. Multi Agent Collaborative Search is a memetic algorithm in which a population of agents combines local search heuristics, exploring the neighbourhood of each agent, with social actions exchanging information among agents. A collection of all Pareto optimal solutions is maintained in an archive that evolves towards the Pareto set. In the approach proposed in this paper, individualistic actions run a local search, from random points within the neighbourhood of each agent, solving a normalised Pascoletti-Serafini scalarisation of the multi-objective NLP problem. Social actions, instead, solve a bi-level problem in which the lower level handles only the constraint equations while the upper level handles only the objective functions. The proposed approach is tested on the multi-objective extensions of two well-known optimal control problems: the Goddard Rocket problem, and the maximum energy orbit rise problem.
{"title":"A direct memetic approach to the solution of Multi-Objective Optimal Control Problems","authors":"M. Vasile, Lorenzo A. Ricciardi","doi":"10.1109/SSCI.2016.7850103","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7850103","url":null,"abstract":"This paper proposes a memetic direct transcription algorithm to solve Multi-Objective Optimal Control Problems (MOOCP). The MOOCP is first transcribed into a Non-linear Programming Problem (NLP) with Direct Finite Elements in Time (DFET) and then solved with a particular formulation of the Multi Agent Collaborative Search (MACS) framework. Multi Agent Collaborative Search is a memetic algorithm in which a population of agents combines local search heuristics, exploring the neighbourhood of each agent, with social actions exchanging information among agents. A collection of all Pareto optimal solutions is maintained in an archive that evolves towards the Pareto set. In the approach proposed in this paper, individualistic actions run a local search, from random points within the neighbourhood of each agent, solving a normalised Pascoletti-Serafini scalarisation of the multi-objective NLP problem. Social actions, instead, solve a bi-level problem in which the lower level handles only the constraint equations while the upper level handles only the objective functions. The proposed approach is tested on the multi-objective extensions of two well-known optimal control problems: the Goddard Rocket problem, and the maximum energy orbit rise problem.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"211 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121474727","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 : 2016-12-01DOI: 10.1109/SSCI.2016.7850084
Ruibin Bai, J. Woodward, N. Subramanian
The service network design problem (SNDP) is a fundamental problem in consolidated freight transportation. It involves the determination of an efficient transportation network and the scheduling details of the corresponding services. Compared to vehicle routing problems, SNDP can model transfers and consolidations on a multi-modal freight network. The problem is often formulated as a mixed integer programming problem and is NP-Hard. In this research, we propose a new efficient large neighbourhood search function that can handle the constraints more efficiently. The effectiveness of this new neighbourhood is evaluated in a tabu search metaheuristic (TS) and a GLS guided local search (GLS) method. Experimental results based on a set of well-known benchmark instances show that the new neighbourhood performs significantly better than the previous arc-flipping neighbourhood. The neighbourhood function is also applicable in other optimisation problems with similar discrete constraints.
{"title":"A new fast large neighbourhood search for service network design with asset balance constraints","authors":"Ruibin Bai, J. Woodward, N. Subramanian","doi":"10.1109/SSCI.2016.7850084","DOIUrl":"https://doi.org/10.1109/SSCI.2016.7850084","url":null,"abstract":"The service network design problem (SNDP) is a fundamental problem in consolidated freight transportation. It involves the determination of an efficient transportation network and the scheduling details of the corresponding services. Compared to vehicle routing problems, SNDP can model transfers and consolidations on a multi-modal freight network. The problem is often formulated as a mixed integer programming problem and is NP-Hard. In this research, we propose a new efficient large neighbourhood search function that can handle the constraints more efficiently. The effectiveness of this new neighbourhood is evaluated in a tabu search metaheuristic (TS) and a GLS guided local search (GLS) method. Experimental results based on a set of well-known benchmark instances show that the new neighbourhood performs significantly better than the previous arc-flipping neighbourhood. The neighbourhood function is also applicable in other optimisation problems with similar discrete constraints.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121475729","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：