Correlation Analysis

In this section, a statistical analysis on the data in the selected Time Interval is presented. The entire page is dominated by a heatmap, which graphically represents the level of cost correlation present between the different elements included in the analysis. On the side, it is possible to note the percentage of correlation that each individual aggregate demonstrates with respect to the total cost.

The heatmap is easily interpretable: each row is associated with a parameter to be monitored, considered as a consumable, while each column is associated with an aggregate. The intersection between a row and a column represents the correlation between the consumable and the associated aggregate. The color scale used varies from white to red passing through blue, where white represents a null correlation value, blue represents a low-moderate value, and red represents a high value.

EXAMPLE

To better clarify the potential of the heatmap as an analytical tool, an example of its use is provided.

Let's suppose we want to evaluate the production data of a plant that deals with the production of spreadable cheese.

The consumables to be evaluated are: production time, production waste, packaging waste, setup time, downtime, raw material, water, energy.

The aggregates useful for the analysis to be considered are: supplier, product code, production line, waiting time on the line.

Observing the chart, it is immediately noticeable how the production line used has a strong impact on the final cost, particularly the line used influences both setup and downtime, but also on the consumption of water and energy.

The supplier affects the cost of raw materials and the amount of waste, while having a lesser impact on the consumption of water and energy and time expenses.

The type of spreadable cheese that was produced has an impact on the cost of water, energy, and raw materials, but also on setup and downtime, while the waiting time of the product on the line does not have a strong influence on the final cost.

All the consumables present in the analysis do not significantly influence the cost related to production time and packaging waste.

By double-clicking on a cell, it is possible to view additional information on the correlation for each of the possible values assumed by the corresponding aggregate.