Data analytics for facility management work orders
The cost invested in the maintenance of the facilities at the Faculty of Terrassa of the polytechnical university of Catalonia (UPC) is high, and the time and resources invested can also be improved, therefore there is an evident need to improve these results in order to optimize maintenance managem...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2023 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/397259 |
| Acceso en línea: | https://hdl.handle.net/2117/397259 |
| Access Level: | acceso abierto |
| Palabra clave: | College buildings Intelligent buildings Edificis universitaris--Gestió Edificis intel·ligents Escola Tècnica Superior d’Enginyeries Industrial i Aeronàutica de Terrassa Àrees temàtiques de la UPC::Edificació::Manteniment d'edificis::Gestió del manteniment d'edificis |
| Sumario: | The cost invested in the maintenance of the facilities at the Faculty of Terrassa of the polytechnical university of Catalonia (UPC) is high, and the time and resources invested can also be improved, therefore there is an evident need to improve these results in order to optimize maintenance management; saving money, reducing execution time to repair failures and reducing environmental impact. Throughout the school year it is possible to register incidents of different types to be solved by maintenance personnel. All this information is stored in a database. This master thesis will focus on analysing those failures related to the facilities registered during a year. Through the analysis of data from registered incidents and the reprocessing of these data, they are made easy to read by grouping them by type of incident and type of failure thanks to keyword search for the categorization and simplification of the stored data. In this database, it is possible to obtain the type of failure and how often it occurs, although it is not possible to record the severity of the failure being reported. After an arduous analysis, it is found that the most frequent failures related to facilities are those related to lighting, toilets and taps, and heating, ventilation and air conditioning (HVAC). There are alternatives to improve these results. - Allowing the software to register equipment and assign preventive maintenance to it, as is done with safety equipment such as elevators or sprinklers that must pass periodic function checks, therefore, reinforcing preventive maintenance in those failures with high frequency can improve the results. - Introducing the use of preventive maintenance through machine learning. This allows, through algorithms, to record the use of the equipment and detect prematurely when a piece of equipment or element is close to failure. It is concluded that through a deep analysis based on an extensive database, very relevant information is obtained, although sometimes it is necessary to transform and decipher it in order to interpret it. Thanks to this process, it is possible to find the necessary information to execute alternatives and improvements in the management and maintenance process. |
|---|