{"id":{"repo_id":"nott-trent","oai_identifier":"oai:irep.ntu.ac.uk:28279"},"canonical_url":"https://search.dev.ndltd.org/etd/nott-trent/oai:irep.ntu.ac.uk:28279","repository":{"repo_id":"nott-trent","name":"Nottingham Trent University","base_url":"https://irep.ntu.ac.uk/cgi/oai2"},"display":{"title":"Baldwinian-based meta-heuristic for robust engineering optimisation","abstract":"The aim of this research was to identify problems in engineering optimisation and then to develop novel solutions to the identified problems. First, principles of computational optimisation were studied and a literature review was conducted. It emerged that the latest research in the area of automated engineering design optimisation tends to combine different optimisation algorithms to improve either effectiveness or efficiency. Three basic types of such hybrid configurations were identified: nested algorithms, sequential algorithms and metaoptimiser. Two problems were then identified that inexperienced practitioners encounter when trying to apply computational optimisation to real‐world engineering problems. The first is the problem of parameter tuning and the second is the problem of finding robust solutions. A wellknown engineering design problem, the pressure vessel problem, was selected as a case study. A problem of engineering optimisation is that the theoretical solutions have to be implemented in the physical world using manufacturing processes, which have a limited accuracy. If an optimum is too narrow or located too close to a constraint, slight deviations from that location will result in a dramatic drop in fitness. In the real‐world this could have catastrophic consequences for practical engineering applications, such as designing a bridge. To overcome these problems, a Baldwinian‐based meta‐heuristic (BMH) was proposed. As well as identifying the fitness of a solution, it also probes its neighbourhood in order to estimate the goodness of the region of the solution. This meta‐heuristic can be combined with any arbitrary optimisation algorithm. SASS was chosen because it only has one control parameter to tune, which makes it most suitable to overcome the problems with parameter tuning. It was shown that BMH/SASS was able to outperform standard SASS as well as particle swarm optimisation (PSO) and hybrid particle swarm branch‐and‐bound (HPB). In conclusion, it can be said that the new method proposed in this work has the potential to find more robust solutions for engineering optimisation applications.","abstract_html":"The aim of this research was to identify problems in engineering optimisation and then to develop novel solutions to the identified problems. First, principles of computational optimisation were studied and a literature review was conducted. It emerged that the latest research in the area of automated engineering design optimisation tends to combine different optimisation algorithms to improve either effectiveness or efficiency. Three basic types of such hybrid configurations were identified: nested algorithms, sequential algorithms and metaoptimiser. Two problems were then identified that inexperienced practitioners encounter when trying to apply computational optimisation to real‐world engineering problems. The first is the problem of parameter tuning and the second is the problem of finding robust solutions. A wellknown engineering design problem, the pressure vessel problem, was selected as a case study. A problem of engineering optimisation is that the theoretical solutions have to be implemented in the physical world using manufacturing processes, which have a limited accuracy. If an optimum is too narrow or located too close to a constraint, slight deviations from that location will result in a dramatic drop in fitness. In the real‐world this could have catastrophic consequences for practical engineering applications, such as designing a bridge. To overcome these problems, a Baldwinian‐based meta‐heuristic (BMH) was proposed. As well as identifying the fitness of a solution, it also probes its neighbourhood in order to estimate the goodness of the region of the solution. This meta‐heuristic can be combined with any arbitrary optimisation algorithm. SASS was chosen because it only has one control parameter to tune, which makes it most suitable to overcome the problems with parameter tuning. It was shown that BMH/SASS was able to outperform standard SASS as well as particle swarm optimisation (PSO) and hybrid particle swarm branch‐and‐bound (HPB). In conclusion, it can be said that the new method proposed in this work has the potential to find more robust solutions for engineering optimisation applications.","abstract_has_math":false,"creators":["Krause, R"],"institution":"Nottingham Trent University","degree_name":"mphil","degree_level":"masters","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2015,"date_issued":"2015-05","date_published":"2015-05","updated_at":"2026-07-24T06:30:54Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Krause, R"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2015-05"]},{"key":"dc:date.issued","label":"Date","values":["2015-05"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["Nottingham Trent University"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://irep.ntu.ac.uk/id/eprint/28279/"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["masters"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["mphil"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://irep.ntu.ac.uk/id/eprint/28279/1/MPhil_Ralph_Krause_finalVersion.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The aim of this research was to identify problems in engineering optimisation and then to develop novel solutions to the identified problems. First, principles of computational optimisation were studied and a literature review was conducted. It emerged that the latest research in the area of automated engineering design optimisation tends to combine different optimisation algorithms to improve either effectiveness or efficiency. Three basic types of such hybrid configurations were identified: nested algorithms, sequential algorithms and metaoptimiser. Two problems were then identified that inexperienced practitioners encounter when trying to apply computational optimisation to real‐world engineering problems. The first is the problem of parameter tuning and the second is the problem of finding robust solutions. A wellknown engineering design problem, the pressure vessel problem, was selected as a case study. A problem of engineering optimisation is that the theoretical solutions have to be implemented in the physical world using manufacturing processes, which have a limited accuracy. If an optimum is too narrow or located too close to a constraint, slight deviations from that location will result in a dramatic drop in fitness. In the real‐world this could have catastrophic consequences for practical engineering applications, such as designing a bridge. To overcome these problems, a Baldwinian‐based meta‐heuristic (BMH) was proposed. As well as identifying the fitness of a solution, it also probes its neighbourhood in order to estimate the goodness of the region of the solution. This meta‐heuristic can be combined with any arbitrary optimisation algorithm. SASS was chosen because it only has one control parameter to tune, which makes it most suitable to overcome the problems with parameter tuning. It was shown that BMH/SASS was able to outperform standard SASS as well as particle swarm optimisation (PSO) and hybrid particle swarm branch‐and‐bound (HPB). In conclusion, it can be said that the new method proposed in this work has the potential to find more robust solutions for engineering optimisation applications."]},{"key":"dc:format","label":"Dc Format","values":["text"]},{"key":"dc:title","label":"Title","values":["Baldwinian-based meta-heuristic for robust engineering optimisation"]}]}],"canonical_facts":{"dc:creator":["Krause, R"],"dc:date":["2015-05"],"dc:date.issued":["2015-05"],"dc:description.abstract":["The aim of this research was to identify problems in engineering optimisation and then to develop novel solutions to the identified problems. First, principles of computational optimisation were studied and a literature review was conducted. It emerged that the latest research in the area of automated engineering design optimisation tends to combine different optimisation algorithms to improve either effectiveness or efficiency. Three basic types of such hybrid configurations were identified: nested algorithms, sequential algorithms and metaoptimiser. Two problems were then identified that inexperienced practitioners encounter when trying to apply computational optimisation to real‐world engineering problems. The first is the problem of parameter tuning and the second is the problem of finding robust solutions. A wellknown engineering design problem, the pressure vessel problem, was selected as a case study. A problem of engineering optimisation is that the theoretical solutions have to be implemented in the physical world using manufacturing processes, which have a limited accuracy. If an optimum is too narrow or located too close to a constraint, slight deviations from that location will result in a dramatic drop in fitness. In the real‐world this could have catastrophic consequences for practical engineering applications, such as designing a bridge. To overcome these problems, a Baldwinian‐based meta‐heuristic (BMH) was proposed. As well as identifying the fitness of a solution, it also probes its neighbourhood in order to estimate the goodness of the region of the solution. This meta‐heuristic can be combined with any arbitrary optimisation algorithm. SASS was chosen because it only has one control parameter to tune, which makes it most suitable to overcome the problems with parameter tuning. It was shown that BMH/SASS was able to outperform standard SASS as well as particle swarm optimisation (PSO) and hybrid particle swarm branch‐and‐bound (HPB). In conclusion, it can be said that the new method proposed in this work has the potential to find more robust solutions for engineering optimisation applications."],"dc:format":["text"],"dc:identifier.uri":["https://irep.ntu.ac.uk/id/eprint/28279/1/MPhil_Ralph_Krause_finalVersion.pdf"],"dc:language":["en"],"dc:publisher.institution":["Nottingham Trent University"],"dc:relation.isreferencedby":["https://irep.ntu.ac.uk/id/eprint/28279/"],"dc:title":["Baldwinian-based meta-heuristic for robust engineering optimisation"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["masters"],"dc:type.qualificationname":["mphil"]},"updated_at":"2026-07-24T06:30:54Z"}