Abstract
Problem solving in continuous industrial processes is
often a time-consuming task where the readings from many
online measurements have to be studied simultaneously. To
aid the process operators of a lime kiln in this task, an
application utilizing case-based reasoning (CBR) was
developed. One of the factors affecting system design was
the number of training cases available, which was quite
small for a CBR application. The diagnostic accuracy of
the system was significantly improved with weight factor
optimization, where the cost function was defined to find
a suitable balance between false positives and false
negatives. The results indicate that there are
economically profitable possibilities in using the CBR
approach and tools in everyday pulp & paper production
processes.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2009 International Conference on Information & Knowledge Engineering, IKE 2009 |
| Publisher | CSREA Press |
| Pages | 351-356 |
| ISBN (Electronic) | 1-60132-115-5 |
| ISBN (Print) | 1-60132-114-7 |
| Publication status | Published - 2009 |
| MoE publication type | A4 Article in a conference publication |
| Event | 2009 International Conference on Information & Knowledge Engineering, IKE 2009 - Las Vegas, United States Duration: 13 Jul 2009 → 16 Jul 2009 |
Conference
| Conference | 2009 International Conference on Information & Knowledge Engineering, IKE 2009 |
|---|---|
| Abbreviated title | IKE 2009 |
| Country/Territory | United States |
| City | Las Vegas |
| Period | 13/07/09 → 16/07/09 |
Keywords
- case-based reasoning
- weight factor
- optimization
- lime kiln
- statistical process control