next up previous contents
Next: Paper machines and pulp Up: Pulp and Paper Industry Previous: Pulp and Paper Industry

Pulp and paper technology

There were three data sets of the technological aspects of pulp and paper industry. The first one included information on the production capacities of the mills, the second included information on the technology of the paper machines and the third on the technology of the pulp lines.

Each mill could contain several paper machines and pulp lines and therefore a hierarchical structure of maps was used (see figure 5.1). First two low level maps were constructed from the paper machine and pulp line data sets. These maps provided a clustering of the different machine types. The actual technology-map was trained using the mill-specific info in the mill data set and the data histograms from the two low-level maps.

Figure 5.1: (a) There were three technological data sets: one of mill production capacities, one of paper machines and one of pulp lines. Each mill could contain several paper machines and pulp lines. (b) The hierarchical map structure. Data histrograms from the two smaller maps were utilized in the training of the third map. The arrows show which data sets were used in training the maps.

To get a better idea of the precision of the maps for different vector components, the mean deviations were calculated for each component. The deviation for component k in vector tex2html_wrap_inline3158 is the absolute difference between the component values in the vector and in its BMU tex2html_wrap_inline3166 : tex2html_wrap_inline3506 . It is averaged over all testing vectors and finally returned to the original value range by multiplying it by the component scaling factor tex2html_wrap_inline3508 :


where tex2html_wrap_inline3510 is the value of the kth component of input vector tex2html_wrap_inline3158 and tex2html_wrap_inline3166 is its BMU.

Juha Vesanto
Tue May 27 12:40:37 EET DST 1997