Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance
In computer-integrated manufacturing, machines are equipped with sensors and memory storage. This provides a large corpus of information retrievable for data mining and knowledge discovery. The case study is focused on the investigation of machine breakdown and the effectiveness of scheduled m...
| Main Author: | |
|---|---|
| Format: | Monograph |
| Language: | English |
| Published: |
Universiti Sains Malaysia
2019
|
| Subjects: | |
| Online Access: | http://eprints.usm.my/58290/ http://eprints.usm.my/58290/1/Data%20Mining%20On%20Machine%20Breakdowns%20And%20Effectiveness%20Of%20Scheduled%20Maintenance.pdf |
| _version_ | 1848883859483000832 |
|---|---|
| author | Tan, Su Sze |
| author_facet | Tan, Su Sze |
| author_sort | Tan, Su Sze |
| building | USM Institutional Repository |
| collection | Online Access |
| description | In computer-integrated manufacturing, machines are equipped with sensors
and memory storage. This provides a large corpus of information retrievable for data
mining and knowledge discovery. The case study is focused on the investigation of
machine breakdown and the effectiveness of scheduled maintenance with the
application of data mining. The research methodology involves eight steps. The initial
step is performed business understanding to gain insight of data mining on machine
performances. Several questions in the stage of macro-level and micro-level are
generated. Second, a proposed simulation model for the operational process was
designed based on a real medical tool manufacturing plant. The production is a job
shop whereby products in batch have to go through a number of processes and multi�stations. Each process alters particular attributes of the product. The production
simulation would be constructed in Witness Horizon V21. Six different machine
breakdown scenarios were modelled. Different feature processing strategies would be
devised, in particular time-related data. Third, a relational database is developed to
store the information from the simulation. The next step is involved data pre�processing which includes data selection, data cleaning and data transformation. Data
mining is the sixth step in which software of Orange is used as the tool. Seventh, the
pattern evaluation is developed to present the discovery of data which helps in
decision-making. From the research, it is found that there have ten types of
breakdowns affecting the performance of machine and the breakdown of coolant
leaking is the main contributor as compared to others breakdown. Besides, the
frequency of breakdown especially for coolant leaking has decreased after the
maintenance is scheduled on the machine. Hence, the application of maintenance on
machine is effective in controlling the frequency of breakdown. From the results of
data, it is able to evaluate the breakdown of machine and support the decision on
scheduling maintenance on machine. However, it requires a large amount of cost to be
invested in maintenance. Last but not least, some of the complex data mining tasks
are not able to perform because of the limited algorithms and machine learning in
Orange software. |
| first_indexed | 2025-11-15T18:57:30Z |
| format | Monograph |
| id | usm-58290 |
| institution | Universiti Sains Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T18:57:30Z |
| publishDate | 2019 |
| publisher | Universiti Sains Malaysia |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | usm-582902023-04-30T04:17:53Z http://eprints.usm.my/58290/ Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance Tan, Su Sze T Technology TJ Mechanical engineering and machinery In computer-integrated manufacturing, machines are equipped with sensors and memory storage. This provides a large corpus of information retrievable for data mining and knowledge discovery. The case study is focused on the investigation of machine breakdown and the effectiveness of scheduled maintenance with the application of data mining. The research methodology involves eight steps. The initial step is performed business understanding to gain insight of data mining on machine performances. Several questions in the stage of macro-level and micro-level are generated. Second, a proposed simulation model for the operational process was designed based on a real medical tool manufacturing plant. The production is a job shop whereby products in batch have to go through a number of processes and multi�stations. Each process alters particular attributes of the product. The production simulation would be constructed in Witness Horizon V21. Six different machine breakdown scenarios were modelled. Different feature processing strategies would be devised, in particular time-related data. Third, a relational database is developed to store the information from the simulation. The next step is involved data pre�processing which includes data selection, data cleaning and data transformation. Data mining is the sixth step in which software of Orange is used as the tool. Seventh, the pattern evaluation is developed to present the discovery of data which helps in decision-making. From the research, it is found that there have ten types of breakdowns affecting the performance of machine and the breakdown of coolant leaking is the main contributor as compared to others breakdown. Besides, the frequency of breakdown especially for coolant leaking has decreased after the maintenance is scheduled on the machine. Hence, the application of maintenance on machine is effective in controlling the frequency of breakdown. From the results of data, it is able to evaluate the breakdown of machine and support the decision on scheduling maintenance on machine. However, it requires a large amount of cost to be invested in maintenance. Last but not least, some of the complex data mining tasks are not able to perform because of the limited algorithms and machine learning in Orange software. Universiti Sains Malaysia 2019-05-01 Monograph NonPeerReviewed application/pdf en http://eprints.usm.my/58290/1/Data%20Mining%20On%20Machine%20Breakdowns%20And%20Effectiveness%20Of%20Scheduled%20Maintenance.pdf Tan, Su Sze (2019) Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance. Project Report. Universiti Sains Malaysia, Pusat Pengajian Kejuruteraan Mekanik. (Submitted) |
| spellingShingle | T Technology TJ Mechanical engineering and machinery Tan, Su Sze Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance |
| title | Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance |
| title_full | Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance |
| title_fullStr | Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance |
| title_full_unstemmed | Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance |
| title_short | Data Mining On Machine Breakdowns And Effectiveness Of Scheduled Maintenance |
| title_sort | data mining on machine breakdowns and effectiveness of scheduled maintenance |
| topic | T Technology TJ Mechanical engineering and machinery |
| url | http://eprints.usm.my/58290/ http://eprints.usm.my/58290/1/Data%20Mining%20On%20Machine%20Breakdowns%20And%20Effectiveness%20Of%20Scheduled%20Maintenance.pdf |