...
The presentation is made by the user creating the different pages according to the specific case that the user wants to present. A summary panel is automatically updated showing the indicators grouped by a user-selectable dimension, and a natural language description of the nature of the process (stable, unstable), patterns that can be detected in the data, and other contextual information. The presentation is made by the user creating the different pages according to the specific case that the user wants to present. A summary panel is automatically updated showing the indicators grouped by a user-selectable dimension, and a natural language description of the nature of the process (stable, unstable), patterns that can be detected in the data, and other contextual information. A tree-shaped structure allows to visualize and select the signals detected in the data corresponding to any indicator present in the Story
...
You proceed to the analysis of indicators interpreting the presence of signals of variation in the data. You can do this by using the previous analysis performed by the application, which comes at two levels:
Statistical: analyzes statistical variation in a normal distribution (+-3 sigma variation). It looks for recognized statistical variation patterns (we called dominant, moderate, weak, ascendant, and near average). The indicators are grouped by type of project and in order to analyze period of homogeneous data. You can see the result of this analysis displaying the tab "Summary" or see it in a new window (navigable) from the option menu. The report shows the nature of the process in relation to the detected signal level: it says stable if no signals are detected, in these cases you can query the process parameters, such as maximum, minimum, average. For indicators with statistical variation signals, you can access a possible diagnostic process that is showing on each shift pattern. Can use in your analysis the qualification level of information (or lack of entropy) present in the data.
Fuzzy: grouping qualitative data sets ("low", "medium", "high", and its qualifiers "very", "slightly", "quite") with respect to another data set taken as a comparison (currently, consolidated data by project type). You can see the information from this "point of view" by selecting the "labels" on the graph of the indicator to get additional insight, "nuanced" change
Since the change in your organization is a dynamic phenomenon, in order to study it you should monitor the capacity of each process using "baseline" for comparison. By setting a new baseline, you persist current data for future comparison (values, statistical limits, statistical and logical signs, context). You may in the future open this "photo" so imprinted on the indicator and thus visualize the evolution of the parameters (called the "voice of the process") and compare it with the desired values (we call the "voice of the customer")Fig. 9. Membership functions used to calculate fuzzy membership grades
...
Normally indicators show a single variable, whether by graphic curves, pie or bar type. You have two ways to include another variable in the same graph (curve):
Add an extra series
Adding a polymetric
With the first technique you can project more than one data series on the same graph X. Usually aggregate series will be the same indicator for another group of projects, or in some cases another indicator to observe a possible correlation.
...