{"id":{"repo_id":"rgu","oai_identifier":"oai:rgu-repository.worktribe.com:1880200"},"canonical_url":"https://search.dev.ndltd.org/etd/rgu/oai:rgu-repository.worktribe.com:1880200","repository":{"repo_id":"rgu","name":"Robert Gordon University","base_url":"https://rgu-repository.worktribe.com/oaiprovider"},"display":{"title":"Bi-level optimisation and machine learning in the management of large service-oriented field workforces.","abstract":"The tactical planning problem for members of the service industry with large multi-skilled workforces is an important process that is often underlooked. It sits between the operational plan - which involves the actual allocation of members of the workforce to tasks - and the strategic plan where long term visions are set. An accurate tactical plan can have great benefits to service organisations and this is something we demonstrate in this work. Sitting where it does, it is made up of a mix of forecast and actual data, which can make effectively solving the problem difficult. In members of the service industry with large multi-skilled workforces it can often become a very large problem very quickly, as the number of decisions scale quickly with the number of elements within the plan. In this study, we first update and define the tactical planning problem to fit the process currently undertaken manually in practice. We then identify properties within the problem that identify it as a new candidate for the application of bi-level optimisation techniques. The tactical plan is defined in the context of a pair of leader-follower linked sub-models, which we show to be solvable to produce automated solutions to the tactical plan. We further identify the need for the use of machine learning techniques to effectively find solutions in practical applications, where limited detail is available in the data due to its forecast nature. We develop neural network models to solve this issue and show that they provide more accurate results than the current planners. Finally, we utilise them as a surrogate for the follower in the bi-level framework to provide real world applicable solutions to the tactical planning problem. The models developed in this work have already begun to be deployed in practice and are providing significant impact. This is along with identifying a new application area for bi-level modelling techniques.","abstract_html":"The tactical planning problem for members of the service industry with large multi-skilled workforces is an important process that is often underlooked. It sits between the operational plan - which involves the actual allocation of members of the workforce to tasks - and the strategic plan where long term visions are set. An accurate tactical plan can have great benefits to service organisations and this is something we demonstrate in this work. Sitting where it does, it is made up of a mix of forecast and actual data, which can make effectively solving the problem difficult. In members of the service industry with large multi-skilled workforces it can often become a very large problem very quickly, as the number of decisions scale quickly with the number of elements within the plan. In this study, we first update and define the tactical planning problem to fit the process currently undertaken manually in practice. We then identify properties within the problem that identify it as a new candidate for the application of bi-level optimisation techniques. The tactical plan is defined in the context of a pair of leader-follower linked sub-models, which we show to be solvable to produce automated solutions to the tactical plan. We further identify the need for the use of machine learning techniques to effectively find solutions in practical applications, where limited detail is available in the data due to its forecast nature. We develop neural network models to solve this issue and show that they provide more accurate results than the current planners. Finally, we utilise them as a surrogate for the follower in the bi-level framework to provide real world applicable solutions to the tactical planning problem. The models developed in this work have already begun to be deployed in practice and are providing significant impact. This is along with identifying a new application area for bi-level modelling techniques.","abstract_has_math":false,"creators":["Ainslie, Russell Thomas"],"institution":"Robert Gordon University","degree_name":"PhD","degree_level":"Doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["J. McCall, S. Shakya and G. Owusu"],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022","date_published":"2022","updated_at":"2026-07-24T04:10:03Z","subjects":["Decision sciences","Information systems and management","Machine learning","Neural networks"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rgu-repository.worktribe.com:1880200","https://doi.org/10.48526/rgu-wt-1880200"],"render_values":[{"text":"oai:rgu-repository.worktribe.com:1880200","href":null,"code":true},{"text":"https://doi.org/10.48526/rgu-wt-1880200","href":"https://doi.org/10.48526/rgu-wt-1880200","code":true}]},{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000-0001-5398-1269"],"render_values":[{"text":"0000-0001-5398-1269","href":"https://orcid.org/0000-0001-5398-1269","code":true}]}]},"links":{"outbound_url":"https://rgu-repository.worktribe.com/1880200/1/AINSLIE%202022%20Bi-level%20optimisation","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["J. McCall, S. Shakya and G. Owusu"]},{"key":"dc:contributor.sponsor","label":"Sponsor","values":["No Funder Acknowledged (Outputs)"]},{"key":"dc:creator","label":"Author","values":["Ainslie, Russell Thomas"]},{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000-0001-5398-1269"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022-10-31"]},{"key":"dc:date.issued","label":"Date","values":["2022"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["Robert Gordon University"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://rgu-repository.worktribe.com/output/1880200"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Decision sciences","Information systems and management","Machine learning","Neural networks"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rgu-repository.worktribe.com:1880200","https://doi.org/10.48526/rgu-wt-1880200"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://rgu-repository.worktribe.com/1880200/1/AINSLIE%202022%20Bi-level%20optimisation"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The tactical planning problem for members of the service industry with large multi-skilled workforces is an important process that is often underlooked. It sits between the operational plan - which involves the actual allocation of members of the workforce to tasks - and the strategic plan where long term visions are set. An accurate tactical plan can have great benefits to service organisations and this is something we demonstrate in this work. Sitting where it does, it is made up of a mix of forecast and actual data, which can make effectively solving the problem difficult. In members of the service industry with large multi-skilled workforces it can often become a very large problem very quickly, as the number of decisions scale quickly with the number of elements within the plan. In this study, we first update and define the tactical planning problem to fit the process currently undertaken manually in practice. We then identify properties within the problem that identify it as a new candidate for the application of bi-level optimisation techniques. The tactical plan is defined in the context of a pair of leader-follower linked sub-models, which we show to be solvable to produce automated solutions to the tactical plan. We further identify the need for the use of machine learning techniques to effectively find solutions in practical applications, where limited detail is available in the data due to its forecast nature. We develop neural network models to solve this issue and show that they provide more accurate results than the current planners. Finally, we utilise them as a surrogate for the follower in the bi-level framework to provide real world applicable solutions to the tactical planning problem. The models developed in this work have already begun to be deployed in practice and are providing significant impact. This is along with identifying a new application area for bi-level modelling techniques."]},{"key":"dc:title","label":"Title","values":["Bi-level optimisation and machine learning in the management of large service-oriented field workforces."]}]}],"canonical_facts":{"dc:contributor.advisor":["J. McCall, S. Shakya and G. Owusu"],"dc:contributor.sponsor":["No Funder Acknowledged (Outputs)"],"dc:creator":["Ainslie, Russell Thomas"],"dc:creator.authoridentifier":["0000-0001-5398-1269"],"dc:date":["2022-10-31"],"dc:date.issued":["2022"],"dc:description.abstract":["The tactical planning problem for members of the service industry with large multi-skilled workforces is an important process that is often underlooked. It sits between the operational plan - which involves the actual allocation of members of the workforce to tasks - and the strategic plan where long term visions are set. An accurate tactical plan can have great benefits to service organisations and this is something we demonstrate in this work. Sitting where it does, it is made up of a mix of forecast and actual data, which can make effectively solving the problem difficult. In members of the service industry with large multi-skilled workforces it can often become a very large problem very quickly, as the number of decisions scale quickly with the number of elements within the plan. In this study, we first update and define the tactical planning problem to fit the process currently undertaken manually in practice. We then identify properties within the problem that identify it as a new candidate for the application of bi-level optimisation techniques. The tactical plan is defined in the context of a pair of leader-follower linked sub-models, which we show to be solvable to produce automated solutions to the tactical plan. We further identify the need for the use of machine learning techniques to effectively find solutions in practical applications, where limited detail is available in the data due to its forecast nature. We develop neural network models to solve this issue and show that they provide more accurate results than the current planners. Finally, we utilise them as a surrogate for the follower in the bi-level framework to provide real world applicable solutions to the tactical planning problem. The models developed in this work have already begun to be deployed in practice and are providing significant impact. This is along with identifying a new application area for bi-level modelling techniques."],"dc:identifier":["oai:rgu-repository.worktribe.com:1880200","https://doi.org/10.48526/rgu-wt-1880200"],"dc:identifier.uri":["https://rgu-repository.worktribe.com/1880200/1/AINSLIE%202022%20Bi-level%20optimisation"],"dc:language":["en"],"dc:publisher.institution":["Robert Gordon University"],"dc:relation.isreferencedby":["https://rgu-repository.worktribe.com/output/1880200"],"dc:subject":["Decision sciences","Information systems and management","Machine learning","Neural networks"],"dc:title":["Bi-level optimisation and machine learning in the management of large service-oriented field workforces."],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["PhD"]},"updated_at":"2026-07-24T04:10:03Z"}